paper_name,article_authors,year_of_publication,article_title,paper_pdf,sample_size,sample_gender,sample_handedness,patient_sample,patient_group_details,simulated_data,simulated_data_details,question_of_interest___1,question_of_interest___2,question_of_interest___3,question_of_interest___4,question_of_interest___5,question_of_interest___6,question_of_interest___7,question_of_interest___8,question_of_interest___9,question_of_interest___10,main_message,study_info_complete,fmri_design,activity_measure,li_calculation_method,calcmethods_compared1___1,calcmethods_compared1___2,calcmethods_compared1___3,calcmethods_compared1___4,calcmethods_compared1___5,calcmethods_compared1___6,calcmethods_compared,li_range,threshold,thresholds_compared1___1,thresholds_compared1___2,thresholds_compared1___3,thresholds_compared1___4,thresholds_compared1___5,thresholds_compared,li_calc_method_description,dominance_cut_offs,dominance_categorisation,region_of_interest,roi_definition,roi_definition_method,no_regions,regions,multiple_regions___1,multiple_regions___2,multiple_regions___3,multiple_regions___4,multiple_regions___5,multiple_regions___6,specific_regions,no_test_sessions,time_between_sessions,fmri_li_protocol_complete,no_lang_tasks,language_tasks,language_tasks_multiple1___1,language_tasks_multiple1___2,language_tasks_multiple1___3,language_tasks_multiple1___4,language_tasks_multiple1___5,language_tasks_multiple1___6,language_tasks_multiple1___7,language_tasks_multiple1___8,specific_language_tasks,language_tasks_description,task_analysis_type,overt_covert,expressive_receptive,baseline_type,baseline_tasks,task_difficulty,task_modality,language_process___1,language_process___2,language_process___3,language_process___4,language_process___5,language_process___6,language_process___7,lang_processes_notes,language_paradigms_complete,verbal_fluency_task,baseline_vf,no_left_lat_vf,no_right_lat_vf,no_bilateral_vf,no_crossed_vf,proportion_typical_vf,li_defined_roi_vf,li_vf_frontal,li_vf_temp,li_global_vf,li_strength_vf,li_reliability_vf,intersubject_variability_vf,li_robustness_vf,roi_effect_vf,calc_effect_vf,baseline_effect_vf,laterality_measures_verbal_fluency_complete,semantic_decision_task,baseline_sd,no_left_lat_sd,no_right_lat_sd,no_bilateral_sd,no_crossed_sd,proportion_typical_sd,li_defined_roi_sd,li_sd_frontal,li_sd_temp,li_global_sd,li_strength_sd,li_reliability_sd,intersubject_variability_sd,li_robustness_sd,roi_effect_sd,calc_effect_sd,baseline_effect_sd,laterality_measures_semantic_decision_complete,phoneme_task,baseline_phoneme,no_left_lat_phoneme,no_right_lat_phoneme,no_bilateral_phoneme,no_crossed_phoneme,proportion_typical_phoneme,li_defined_roi_phoneme,li_frontal_phoneme,li_temp_phoneme,li_global_phoneme,li_strength_phoneme,li_reliability_phoneme,intersubject_variability_phoneme,li_robustness_phoneme,roi_effect_phoneme,calc_effect_phoneme,baseline_effect_phoneme,laterality_measures_phonetic_judgement_complete,word_read_task,baseline_word_read,no_left_lat_read,no_right_lat_read,no_bilateral_read,no_crossed_read,proportion_typical_read,li_defined_roi_read,li_frontal_read,li_temp_read,li_global_read,li_strength_read,li_reliability_read,intersubject_variability_read,li_robustness_read,roi_effect_read,calc_effect_read,baseline_effect_read,laterality_measures_word_reading_complete,naming_task,baseline_naming,no_left_lat_naming,no_right_lat_naming,no_bilateral_naming,no_crossed_naming,proportion_typical_naming,li_defined_roi_naming,li_frontal_naming,li_temp_naming,li_global_naming,li_strength_naming,li_reliability_naming,intersubject_variability_naming,li_robustness_naming,roi_effect_naming,calc_effect_naming,baseline_effect_naming,laterality_measures_naming_complete,speech_listening_task,baseline_speech_listening,no_left_lat_listening,no_right_lat_listening,no_bilateral_listening,no_crossed_listening,proportion_typical_listening,li_defined_roi_listening,li_frontal_speech_listening,li_temp_speech_listening,li_global_speech_listening,li_strength_speech_listening,li_reliability_speech_listening,intersubject_variability_listening,li_robustness_speech_listening,roi_effect_speech_listening,calc_effect_speech_listening,baseline_effect_speech_listening,laterality_measures_passive_speech_listening_complete,sentence_comp_task,baseline_sc,no_left_lat_sc,no_right_lat_sc,no_bilateral_sc,no_crossed_sc,proportion_typical_sc,li_defined_roi_sc,li_frontal_sc,li_temp_sc,li_global_sc,li_strength_sc,li_reliability_sc,intersubject_variability_sc,li_robustness_sc,roi_effect_sc,calc_effect_sc,baseline_effect_sc,laterality_measures_sentence_comprehension_complete,language_task,baseline_other,no_left_lat_other,no_right_lat_other,no_bilateral_other,no_crossed_other,proportion_typical_other,li_defined_roi_other,li_frontal_lang_other,li_temp_lang_other,li_global_lang_other,li_strength_other,li_reliability_other,intersubject_variability_other,li_robustness_other,roi_effect_other,calc_effect_other,baseline_effect_other,laterality_measures_other_language_task_complete,language_tasks_cta,cta_method,baselines_cta,no_left_lat_cta,no_right_lat_cta,no_bilateral_cta,no_crossed_cta,proportion_typical_cta,li_defined_roi_cta,li_frontal_lang_cta,li_temp_lang_cta,li_global_lang_cta,li_strength_cta,correlation_cta_tasks,li_reliability_cta,intersubject_variability_cta,li_robustness_cta,roi_effect_cta,calc_effect_cta,baseline_effect_cta,laterality_measures_cta_complete,task_correlations,simulated_findings,patient_findings,colateralization,crossed_lateralisation,stimulus_modality_effect,handedness_effect_li,study_findings,study_conclusions,other_findings_complete
3,"Abbott, Waites, Lillywhite and Jackson",2010,"fMRI assessment of language lateralisation: an objective approach",,"34 ","20 male, 14 female",6,1,"4 epilepsy patients",1,"Took real resting data from healthy subjects and added a known signal response. Added signal time courses within three ROIs centred on peak clusters from actual group activation data. Constructed a series of datasets with varying percentage signal change. Each dataset contained lateralization equivalent to an LI of 0.3.",0,1,1,0,0,0,0,0,0,0,"Laterality should not be determined at a single threshold but using thresholds that yield a fixed number of active voxels i.e. an individualy determined variable threshold, as this is more robust. Argue for the utility of an approach to dominance classification in which an LI distribution across voxel counts for a subject can be compared to a normative distribution. ",2,1,1,1,0,0,0,0,0,0,,1,6,0,1,1,0,1,"Variable threshold,     Multiple thresholds,     Unthresholded","Calculated the LI at multiple thresholds, and then calculated at a threshold that yielded a fixed number of active voxels. Calculated the distribution of LI as function of the number of voxels above threshold i.e. obtained a voxel count distribution.   ",,"A threshold independent method was used to assess whether lateralisation was atypical. This involved statistically comparing the LI distribution of a subject to a group of LI distributions for a normative group- calculate the probability of the laterality of a subject being consistent with the control group. ",2,3,"Defined using a combinationof functional and anatomical landmarks. ","1 ",1,0,0,0,0,0,0,"Inferior frontal and middle frontal gyri. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Letter cued word generation","Covert word generation cued by visually displayed letters. ",1,2,1,1,"Visual cross fixation (rest)",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Passive cross fixation",,,,,,,,,,"LI strength varied as a function of threshold level and active voxel threshold, from about 0.4 to 1. ",,,"Simulated data indicated that the use of a single fixed threshold was not robust against the effect of noise, whereas thresholding at a fixed number of voxels was robust against variability in signal strength.",,"Overall, calculation of LI using a threshold that yields a fixed number of active voxels was more robust against noise. The distribution of LIs across different voxel counts was much tighter than the distribution of LIs across different threshold levels. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,"Simulated data indicated that the use of a single fixed threshold was not robust against the effect of noise, whereas thresholding at a fixed number of voxels was robust against variability in signal strength. ",,,,,,,"Laterality should not be determined at a single threshold but using thresholds that yield a fixed number of active voxels i.e. an individualy determined variable threshold, as this is more robust. Argue for the utility of an approach to dominance classification in which an LI distribution across voxel counts for a subject can be compared to a normative distribution. ",2
4,"Adcock, Wise, Oxbury, Oxbury, Matthews",2003,"Quantitative fMRI assessment of the differences in lateralisation of language-related brain activation in patients with temporal lobe epilepsy.",,"12 ","6 male, 6 female",1,1,"19 right handed temporal lobe epilepsy patients. ",0,,0,1,0,0,0,0,0,0,0,0,"Aimed to compare different methods of determining laterality from fMRI data, and to investigate reproducibility of fMRI LIs. Found high reproducibility across two testing sessions and reported that LIs based on magnitude of signal were more robust than those based on signal extent. ",2,1,3,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Four methods of calculation were compared:  Voxel counts, anatomically defined ROI  Voxel counts, functionally defined ROI  Signal magnitude, anatomically defined ROI  Signal magnitude, functionally defined ROI  ","Within 2 SDs of the mean","The categorical definition of laterality (as right, left or bilateral) was based on cut-off values defined at two standard deviations on either side of the mean for an LI. ",2,3,"Anatomically defined VOI: a large area including frontal, parietal and temporal cortex (excluding occipital cortex and cerebellum).   Functionally defined VOI: derived from a map of language activity, including frontal and temporal areas (flipped symmetrically to give right homologues).",2,5,1,1,0,1,0,0,"Inferior frontal gyrus,     Superior temporal gyri,    Premotor areas,    Anterior cingulate gyrus",2,"3 months",2,1,1,0,0,0,0,0,0,0,0,"Visually letter cued covert word generation","Silent word generation cued by visually displayed letters. ",1,2,1,1,"Cross fixation",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Cross fixation",12,0,0,0,1,"0.633, 0.78","0.759 ",,,"Strength of LI extent measures were dependent on z-score threshold used. ","In all cases, categorical definitions of laterality agreed across the two testing sessions. There was a good correlation across testing sessions for LI magnitude measure, and no significant difference between sessions at any threshold level for extent LIs. Reported that magnitude measures had a smaller relative intersessional difference than extent measures. ",,"LI extent strength varied according to the threshold level. LI magnitude measures were described as more robust than LI extent measures as they had a smaller mean relative intersessional difference. ","LI extent based on functionally defined ROI was higher than LI extent based on anatomically defined VOI, at a threshold of 2.3 (not at threshold of 5.3). The most active ROI for the verbal fluency task was the IFG. There was no significant difference in strength of LI magnitude measure according to whether a functional or anatomical ROI was used. ","Argued for the superiority of magnitude measures. LI extent measures were threshold dependent and although reproducible showed slightly more variation across testing sessions than magnitude measures. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"fMRI LIs are highly reproducible. Magnitude LIs were more robust than extent LIs in terms of lower variability across testing sessions. ROI had little effect on strength of LI. ",2
5,"Baciu, Juphard, Cousin, Le Bas",2005,"Evaluating fMRI methods for assessing hemisphere language dominance in healthy subjects",,10,"5 male, 5 female",1,2,,0,,0,1,0,0,1,0,0,0,0,0,"Both the standard LI method and the flip method were robust for measuring language lateralisation; however, the reliability of each method depends on the task used. ",2,1,3,6,1,1,0,0,0,0,"Standard LI equation    Flip method",1,1,0,0,0,0,0,,"Compared two methods:    Standard LI method: LIs calculated based on voxel counts within a VOI consisting of inferior frontal, temporal and parietal cortex.     Flip method: First, two sets of images are created; right side images (normalised L and R) and a mirror images set (flipped so that LH is on the right). Calculated the following statistical contrast: [task vs control] for right side images versus [task vs control] for mirror images. Then calculate LIs as normal using voxel counts in new images within ROI. ",0.2,"Typical 0.2 cut-offs. ",2,2,"Based on functional maps provided by the task vs control contrast. ","1 large VOI",3,0,0,0,0,0,0,"Inferior frontal cortex (44, 45, 47)    Temporal cortex (22, 21, 37)    Parietal cortex (40, 39)",1,,2,2,9,1,0,1,0,0,0,0,0,"Word generation task,    Rhyme detection task","Word generation- covert generation from visually presented letters.     Rhyme detection- decide if pairs of words rhyme or not using response keys.",1,2,3,3,"Passive cross fixation for word generation.     Visual detection task for rhyme detection- perceptual judgement on unreadable words. ",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Cross fixation",10,0,0,0,1,"0.935 for FM, 0.901 for LIM",,,,"LI values were generally high- lowest was 0.48, highest was 1. ",,"Low intersubject variability; 7/10 subjects showed identical LI values across both methods. LI ranged from 0.48 to 1. SD of LIM values was 0.19. ","Laterality measures using word generation were not robust against the effects of normalisation, smoothing and clustering; these procedures increased laterality estimates for LIM from 0.55 to 0.9. ","Lateralisation was strong in the inferior frontal region, but the task also engaged an inferior temporal region. ","Using the word fluency task, neither the LIM or FM individually, nor methods confounded yielded LI values that correlated significantly with manual LIs (Edinburgh inventory). Argued therefore to be poor at indicating the dominant hemisphere for language. ",,2,,,,,,,,,,,,,,,,,,,2,"Rhyme detection task","Visual detection task",,,,,,"0.844 (FM), 0.745 (LIM)","0.844 (FM), 0.745 (LIM)","0.844 (FM), 0.745 (LIM)",,"LI strength was relatively high, but lower on average than LIs from the word generation task. ",,"Larger inter-subject variability in LI values than for word generation (SD for LIM = 0.28)","Laterality measures using rhyme detection were robust against the effects of normalisation, smoothing and clustering; these procedures did not significantly change the LI value. ","Rhyme detection yielded lateralised activation within both Broca's and Wernicke's areas; was good at engaging posterior temporal regions. ","For rhyme detection, the flip method showed greater correlation with manual LI values (Ediburgh inventory) than the standard LI method. With methods confounded, rhyme detection LIs showed significant correlation with MLIs (not the case for word generation). ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"Word generation and rhyme detection LIs not compared directly, however the word generation task did yield slightly higher LIs than the rhyme detection task. ",,,,,,,,"The rhyme detection task appears more robust than the word generation task in that LI values were more resistant to the effects of normalisation, smoothing and clustering. For rhyme detection, FM appeared superior to LIM in terms of correlation with manual LIs; neither method correlated with manual LIs for word generation. ",2
6,"Bethmann, Tempelmann, Bleser, Scheich, Brechmann",2007,"Determining language laterality by fMRI and dichotic listening",,30,"15 male, 15 female",5,2,,0,,0,1,0,1,0,0,0,0,0,0,"Analysing only one ROI is not sufficient to determine language dominance within a single subject. ",2,1,3,1,0,0,0,0,0,0,,2,1,0,0,0,0,0,,"Used the standard LI equation using both the BOLD signal intensity and the number of activated voxels. Calculated for each region and then an average taken across regions. ",-20,"LI above +20 = left  LI below -20 = right  Li between +2- and -20 = bilateral",2,2,"ROIs defined according to activation peaks obtained. ",4,3,0,0,0,0,0,0,"Inferior frontal sulcus,     Inferior frontal gyrus,     Superior temporal sulcus,     Angular gyrus",1,,2,1,2,0,0,0,0,0,0,0,0,"Semantic decision task (from Fernandez et al, 2001)","Decide if two visually presented words are semantically related or unrelated. ",1,2,2,2,"Perceptual matching decision- decide if two non-word letter strings were identical. ",3,2,1,1,0,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,"Semantic decision","Perceptual decision",26,3,1,4,0.8666666666666667,84,,,,"LI strength was generally high; only 1 case of bilateral (weak) lateralisation. LI strength did not differ significantly across LI method (extent or magnitude) or across ROIs at a group level. ",,"In general, fairly low variability in LI (84+-15). Greater variability in language laterality when multiple regions were considered i.e. two individuals could show concordant laterality for one ROI but discordant for another ROI. ",,"Found no significant difference in magnitude of laterality across frontal and temporal regions. Laterality across the four regions was significantly correlated, except between STS and IFG (only a trend). Concordance between dominance categories across all ROIs was found in 26/30 subjects. However, there were four cases of crossed dominance across regions. ","Voxel count LIs and signal change magnitude LIs were strongly correlated and always categorised dominance concordantly. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,"Four out of 30 subjects showed evidence of crossed lateralisation in which one region differed in its laterality category from that of other regions. In one case, bilateral categorisation reflected a left frontal dominance but a right temporal dominance.   ",,"Overall, handedness did not correlate with the general LI; but language dominance was generally concordant with hand dominance. ",,"Classifying language dominance based on a single ROI is not always appropriate for characterising an individual's pattern of language laterality, in light of cases of crossed regional dominance. LI extent and magnitude measures did not differ significantly in magnitude and were strongly correlated. ",2
7,"Berl, Zimmaro, Khan, Dustin, Ritzl, Duke et al.",2014,"Characterisation of atypical language activation patterns in focal epilepsy. ",,118,"60 male, 58 female",1,1,"220 patients with focal epilepsy",0,,0,0,1,1,0,0,0,0,0,0,"A data-driven clustering method indicated the existence of meaningful categories of language dominance versus the arbitrary categorisation cut-offs traditionally used. ",2,1,2,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used Wilke and Schmithorst's (2006) bootrapping method within LI toolbox. ",0.2,"Compared two methods of classifying language dominance:  Traditional 0.2 cut-offs.     Hierarchical cluster analysis method, using Ward's method to measure similarity between cases. ",2,1,"Anatomically defined using the Wake Forest PickAtlas.",2,3,0,0,0,0,0,0,"Broca's area,     Wernicke's area",1,,2,1,7,0,0,0,0,0,0,0,0,"Semantic definition verification","Decide whether a word definition is semantically correct or not e.g. 'A King's hat is a crown.'",1,2,2,2,"Listen to reverse speech and identify tones. ",3,1,0,1,0,1,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,"Semantic definition verification","Tone identification ",115,0,1,2,0.9745762711864406,,,,,,,,,"Reported a case of crossed dominance across frontal and temporal regions, and therefore argued that laterality should be assessed on a regional rather than a global basis. ","Clustering analysis also reported that nearly all healthy volunteers were left dominant. Found one case of crossed dominance, otherwise all atypically categorised subjects were on or adjacent to the 0.2 threshold. Overall, this data-driven method largely agreed with the traditional method. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Semantic definition verification","Tone identification",115,0,1,2,0.9745762711864406,,,,,,,,,"Reported a case of crossed dominance across frontal and temporal regions, and therefore argued that laterality should be assessed on a regional rather than a global basis. ","Clustering analysis also reported that nearly all healthy volunteers were left dominant. Found one case of crossed dominance, otherwise all atypically categorised subjects were on or adjacent to the 0.2 threshold. Overall, this data-driven method largely agreed with the traditional method. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,"The lateralisation categories yielded by the cluster analysis showed correlations with relevant clinical features. ","For the majority of individuals, Broca's area and Wernicke's area were both left dominant (87/118 cases). ","Reported one case of crossed lateralisation within the control group. ",,,,"A data-driven clustering method indicated the existence of meaningful categories of language dominance versus the arbitrary categorisation cut-offs traditionally used. Laterality should be assessed on a regional rather than a global level. ",2
8,"Fernandez, de Greiff, von Oertzen, Reuber, Lun, Klaver et al.",2001,"Language mapping in less than 15 minutes: Real-time functional MRI during routine clinical investigation.",,12,"6 male, 6 female",1,1,"12 epilepsy patients. ",0,,0,1,0,1,0,0,0,0,0,0,"Demonstrated the reliability of a real-time fMRI lateralisation analysis. ",2,1,2,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used a variable threshold for each subject set at half the mean maximum t-value, defined as the mean of those 5% voxels showing the highest level of activation. Then calculated a weighted LI using the sum of t-values of supra-threshold voxels multiplied by the set of activated voxels. Compared two methods of LI calculation; online and offline. Laterality raters viewed intermixed real-time and offline datasets and rated them as left, right or bilateral, as well as comparing the two datasets in terms of quantified LI. ",,"Language dominance was classified by visual inspection of contrast images by neurologists who rated language dominance as left, right, bilateral or unclear.  ",3,3,"Three ROIs defined in a data-derived manner by looking at activation maps generation (with mirror images to obtain right homologues). A global VOI was defined by masking the whole brain, excluding three sagittal midline planes.",3,3,0,0,0,0,0,0,"Broca's area,     Prefrontal areas outside Broca's area,     Temporoparietal area",1,,2,1,2,0,0,0,0,0,0,0,0,"Semantic decision task","Decide if two visually presented words are semantically related or unrelated. ",1,2,2,2,"Perceptual matching decision- decide if two non-word letter strings were identical. ",3,2,0,1,0,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,"Semantic decision","Perceptual decision",11,1,0,0,0.9166666666666666,0.82,0.73,0.83,0.66,"Note that LI mean values given above are excluding 1 subject showing strong right lateralisation. In general, laterality is strong (between 0.66 and 0.82 depending on ROI)",,"SDs of between 0.22 and 0.31. ",,"Interestingly, found that the temporoparietal LI mean was higher than the frontal LI. Global LI was also lower than regional LIs. ","When comparing the two methods of LI calculation (online and offline), both regional and global LI values were significantly correlated. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,"Generally, language lateralisation in patients and controls was similar. ","Lateralisation within frontal and temporoparietal ROIs not directly compared; no data explicitly given to enable comparison. ",,,,,"Laterality analyses based on offline and online image processing were highly correlated. For this semantic decision task, temporoparietal mean LI was higher than frontal mean LI. ",2
9,"Jansen, Menke, Sommer, Forster, Bruchmann, Hempleman, Weber, Knecht.",2006,"The assessment of hemispheric lateralization in functional MRI- robustness and reproducibility",,10,"6 male, 4 female",4,2,,0,,0,1,0,1,1,1,0,0,0,0,"The robustness and reproducibility of fMRI laterality indices are affected by multiple methodological aspects, including the threshold used, the activity measure used, the ROI used and the task used. ",2,1,3,1,0,0,0,0,0,0,,1,6,0,1,1,0,0,"Multiple fixed thresholds,    Variable threshold","Used the standard LI equation, and varied the choice of activity measure and definition of ROI:  Extent of activation (at both fixed and variable thresholds)  Magnitude of activation (either all within ROI, or only those exceeding a threshold level)    Fixed thresholds- LIs were calculated for a range of statistical thresholds.   Variable thresholds- threshold defined at level to yield fixed number of activated voxels (different criterion number chosen for different tasks). ",0.2,"Positive LI value = Left dominance  Negative LI value = Right dominance  LI < 0.2 = Bilateral",3,3,"Functionally derived ROI- Took significantly activated clusters and mirrored them to get RH homologues.     Anatomically derived ROI- included all cerebral regions excluding the cerebellum. ",3,5,1,1,0,0,0,0,"Broca's area    Rest of prefrontal cortex    Temporoparietal cortex    Whole cerebral hemispheres (anatomical ROI)",2,"2-2.5 hours apart (same day)",2,3,9,1,1,0,0,1,0,0,0,"Word generation,    Semantic decision,    Picture naming.","Word generation- covert letter cued generation.     Semantic decision- decide if words are synonyms.    Picture naming- silent naming of pictures.",2,2,3,3,"Word generation- silent repetition of non-word 'baba'.    Semantic decision- perceptual matching for non-word letter strings.     Picture naming- cross fixation. ",4,2,1,1,0,0,0,1,0,,2,"Letter cued word generation","Silent nonword repetition",,,,,,,,,,,"LI extent based on either fixed or variable thresholds determined dominance reliably; but variable thresholds yielded more reliable LIs.   For, LI magnitude, reproducibility for both degree of LI and dominance was higher when only those voxels reaching a threshold level within a ROI were used rather than all voxels within a ROI.  Pearson's correlation across testing sessions for magnitude LIs was 0.61 within Broca's area.   Reported that bilaterality was not reproducible in over 90%, independent of ROI and method of calculation.   Overall, WGT was showed slightly higher reproducibility than SDT. ","LI extent was not suitable for determining individual variability in degree of lateralization (often yielded an LI of 1), whereas LI magnitude yielded greater variation in LI values. ","LI extent not robust against changes in threshold- reported some cases in which a change in threshold used reversed the sign of the LI within Broca's area. ","Reproducibility of LI extent was unexpectedly higher for the whole hemisphere than functionally defined ROIs when LI extent was used with fixed thresholds (intersessional correlation of 0.18 for Broca's area, 0.63 for whole hemisphere).  ","Consistent dominance classifications were achieved across all approaches to LI calculation for word generation.  However, reliability of LI was greatly affected by choice of activity measure and ROI; these two factors interacted i.e. one ROI may be more or less reproducible than another depending on the choice of activity measure.   ",,2,"Synonym decision","Perceptual decision",,,,,,,,,,,"Reproducibility of LI extent between the two sessions was slightly lower for the synonym task compared to the word generation task. ","LI extent was not suitable for determining individual variability in degree of lateralization (often yielded an LI of 1), whereas LI magnitude yielded greater variation in LI values. ",,"Reproducibility of LI magnitude (using only above threshold voxels) was higher within temporoparietal ROI than within Broca's area (0.49 vs 0.32). ","Consistent dominance classifications were achieved across all approaches to LI calculation using semantic decision. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Picture naming","Cross fixation",,,,,,,,,,"Weaker lateralisation was found for the picture naming task compared to word generation and semantic decision. ","Picture naming did not determine dominance reliably in about a third of subjects. ","Higher variability in LI values using the picture naming task compared to other language tasks. ",,,"Strong effect of calculation method on laterality/dominance measurements- different methods of calculation yielded different dominance classifications. Not the case for the other tasks (word generation and semantic decision). ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Word generation,    Synonym decision,     Picture naming. ","Used conjunction analysis implemented in SPM99 to calculate a combined language activation pattern for each subject (Ramsey et al, 2001). ","Non-word repetition,     Perceptual decision,     Cross-fixation.",,,,,,,,,,"Generally, CTA yielded higher LI values than those seen for the individual tasks. Within the language ROIs, CTA LI values were 1 for most subjects. ",,"Could reliably determine language dominance using either LI extent or LI magnitude. Could not reliably determine bilaterality. Overall reported that the reproducibility for the CTA was comparable to that of the word generation task alone. ","Reduced intersubject variability in LI compared to individual tasks alone; within the language ROIs, CTA LI values were 1 for most subjects. ",,,"CTA yielded LIs with reduced threshold dependence- stable activity could still be observed at higher thresholds than for the individual tasks. ",,2,"Correlations between tasks not given.",,,,"Graphs indicate some evidence of crossed lateralization e.g. there are cases in which LIs are positive for temporoparietal cortex but negative for Broca's area using the SDT. Not discussed in paper. ",,,,"LI extent and LI magnitude measures have similar reproducibility.   But, recommended that the most reproducible and robust LIs could be obtained using magnitude of signal change from those voxels exceeding a pre-set threshold value within a defined ROI.   The word generation task may be just as reproducible as CTA, which are both more reproducible than semantic decision. ",2
10,"Binder, Swanson, Hammeke, Sabsevitz",2008,"A comparison of five fMRI protocols for mapping speech comprehension systems",,26,"13 male, 13 female",1,2,,0,,0,0,0,0,1,0,0,0,0,0,"Speech comprehension paradigms yield strongest lateralisation when an active non-linguistic baseline task is used, to interrupt resting activation of conceptual representations. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used the standard LI equation using voxel counts within global ROIs of the left and right hemisphere (excluding cerebellum and brainstem) at a single fixed threshold. ","Not given",,1,1,"Created individual masks for each subject by excluding voxels outside the brain and those within the cerebellum and brainstem. Then divided this mask at the midline to produce separate right and left ROIs. ","1 in each hemisphere",6,0,0,0,0,0,1,"Global hemispheres",1,,2,3,9,0,1,1,0,0,1,0,0,"Passive word listening,    Semantic decision,    Phoneme decision.","Word listening- passively listened to animal words.    Semantic decision- decide if animal words are both found in the USA and used by people.     Phoneme decision- decide if triplet of CV syllables included both the consonants /b/ and /d/.",1,2,2,3,"Rest.    Passive tone listening.    Active tone decision- decide if sequence contained two 750Hz tones.     ",4,1,1,1,0,1,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,"Semantic decision","Rest, Tone decision, Phoneme decision",,,,,,,,,"0.117 - 0.624","LI strength obtained depended on baseline/contrast condition used. Out of all receptive speech comprehension protocols tested, a contrast of semantic decision versus active tone decision gave the stronger laterality (compared to passive word listening versus rest and semantic decision versus rest). ",,"LIs for the semantic decision versus either tone decision or phoneme decision protocols showed reduced inter-subject variability (lower SD) than protocols using passive paradigms. ","LI obtained with semantic decision is not robust against changes in the baseline task used- heavily dependent on control task used. ",,,"Large effect of baseline used. Changing the baseline from rest to active tone decision raised the LI value from 0.117 to 0.624. Single sample t-test found that LI values for SD versus rest did not differ significantly from zero. Use of phoneme decision as a baseline yielded similarly strong lateralisation (LI = 0.62). ",2,"Phoneme decision","Compared to semantic decision",,,,,,,,,0.62,"Phoneme decision was used as a control for phonemic processing within the semantic decision task (see notes on semantic decision task). This yielded strong laterality. ",,"Using phoneme decision as a control for semantic decision yielded high consistency across subjects in LI (smaller SD than passive paradigms). ",,,,"Phoneme decision was a good baseline for semantic decision, yielding high and consistent laterality. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Passive word listening","Rest, passive tone listening",,,,,,,,,"0.1 - 0.522","Strength of laterality obtained depended on the baseline task used; higher LI was found when contrasted to a baseline of tone listening (0.522) versus rest (0.1). Passive words versus rest yielded LIs that did not differ significantly from zero (single samples t-test). ",,"Higher variability in LI was found for these passive protocols compared to active speech comprehension protocols such as semantic decision/tone decision. SDs of between 0.415 and 0.457. ",,,,"Large effect of baseline task; higher LI was found when contrasted to a baseline of tone listening (0.522) versus rest (0.1). ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,"The same brain areas were activated by resting versus tone decision and semantic decision versus phoneme decision protocols. This suggests that resting must activate similar conceptual representations to semantic decision (hence the need for an active baseline). ","Resting baselines should not be used for speech comprehension protocols, due to resting activation of conceptual representations. Overall, a semantic decision- tone decision protocol was optimal in terms of producing the strongest and most consistent lateralisation. ",2
11,"Allendorfa, Hernando, Hossain, Nenert, Holland, Szaflarski",2016,"Arcuate fasciculus asymmetry has a hand in language function but not handedness. ",,214,"108 male, 132 female",5,2,,0,,0,0,0,0,1,0,0,0,0,0,"Focus of paper was on comparing structural and functional asymmetries with handedness. Provides a measurement of LI with verb generation.",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used LI toolbox on fMRI data within a functional mask. Threshold adaptively defined as the mean intensity of voxels within that mask. Number of voxels surviving this threshold in R and L hemispheres then used for standard LI calculation. Removed statistical outliers using data clustering and variance weighting options. ",0.1,"LI >= 0.1 = left lateralised  LI <= -0.1 = right lateralised  -0.1< LI <0.1 = bilateral ",2,2,"Clusters of significant activation during the language paradigm were used to create a functional mask for each subject. Voxels above threshold in the left hemisphere were mirrored over to the right hemisphere and vice versa to obtain a symmetrical mask. This included frontal, temporal and parietal areas. ","Multiple in a single VOI",3,0,0,0,0,0,0,"Single large mask including frontal, temporal and parietal regions. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Covert verb generation.","Subjects had to covertly generate related verbs for a auditorily presented noun.",1,2,1,2,"Bilateral finger tapping in response to a 400 Hz tone. ",2,1,1,0,0,1,0,0,0,,2,"Verb generation","Bilateral finger tapping",180,18,16,,0.8411214953271028,"0.44 ",,,,"LI values ranged from 0.35 for left handers to 0.44 for right handers. ",,"Standard deviation of LI values for right handers was 0.3 and for left handers was 0.44. ",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"LIs were not significantly different between right handed subjects and subjects with atypical handedness. However, there was an association between LIs and Edinburgh handedness inventory scores. ",,"Laterality strength did not differ between right and left handers for verb generation; was on average 0.44 for right handers and 0.35 for left handers. ",2
12,"Backes, Deblaere, Vonck, Kessels, Boon, Hofman, Wilmink et al.",2005,"Language activation distributions revealed by fMRI in post-operative epilepsy patients: Differences between left- and right-sided resections. ",,9,"3 male, 6 female",1,1,"12 right temporal lobe and 9 left temporal lobe resected epilepsy patients. ",0,,0,0,0,0,1,0,0,0,0,0,"Focus of study was on patient group. Demonstration of lateralising ability of word generation task/word reading task. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts within right and left hemispheres for the standard LI equation. Calculated LIs across a range of thresholds for each subject. When group comparisons were made, threshold was set at Z>3.8. ",0.2,"Standard 0.2 cut-offs:  LH = >0.2  RH = <-0.2  BL = between -0.2 and +0.2",2,3,"Used digital brain masks to select ROIs that had previously been identified as being activated in controls for the same language tasks (Deblaere et al, 2002). ","Multiple in a single VOI",3,0,0,0,0,0,0,"Inferior and middle frontal cortex,     Superior frontal cortex,     Anterior temporal lobe,     Posterior and inferior temporal lobe,     Temporo-parietal junction. ",1,,2,2,9,1,0,0,1,0,0,0,0,"Word generation,     Text reading. ","Word generation- covert generation in response to a visually presented letter.     Text reading- covert reading of meaningful text.",1,2,3,3,"Word generation- passive cross fixation.     Text reading- reading of text composed of pronounceable non-words. ",2,1,1,1,1,0,0,1,0,,2,"Letter cued word generation","Passive cross fixation",8,0,1,0,0.8888888888888888,0.66,,,,"Strong positive LI value found. ",,"Low inter-subject variability (SD = 0.09)",,"High LI value was mainly mediated by strong left lateralised activation in the prefrontal cortex.",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Covert text reading","Covert nonsense text reading",8,0,1,0,0.8888888888888888,0.59,,,,"Relatively strong lateralisation (0.59) but lower mean LI value than for word generation (0.66). ","Low inter-subject variability in LI (SD = 0.12)",,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,"Findings suggested interhemispheric reorganisation of language function in left temporal lobe patients. ",,,,,,"Strong mean LIs obtained for both word generation and text reading covert paradigms. ",2
14,"Brennan, Whalen, Branco, O'Shea, Norton, Golby",2007,"Object naming is a more sensitive measure of speech localization than number counting: Converging evidence from direct cortical stimulation and fMRI. ",,7,"2 male, 5 female",1,1,"Fifteen left hemisphere damage patients. ",0,,0,0,0,0,1,0,0,0,0,0,"Aimed to compare automated (number counting) and non-automated (object naming) speech tasks to assess their lateralising ability. ",2,1,,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Standard LI formula used (not clear if using voxel extent or magnitude), from voxels surviving threshold within ROIs. LIs were calculated at a range of thresholds.",0.2,"Standard 0.2 cut-offs. ",2,1,"Defined anatomically using the Pick Atlas. ",2,3,0,0,0,0,0,0,"Broca's area (inferior and middle frontal gyri),    Wernicke's area (superior temporal, supramarginal and angular gyri)",1,,2,2,9,0,0,0,0,1,0,0,1,"Object naming,    Number counting","Object naming- covertly name visually presented objects.     Number counting- covertly generate the number sequence.",1,2,1,3,"Passive fixation,     Viewing of non-semantic nonsense objects (perceptual control)",3,2,1,1,0,0,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Object naming","Number counting",,,,,,,0.35,0.76,,"When object naming was contrasted with number counting, strong left lateralisation was found within Wernicke's area. Left lateralisation within Broca's area was weaker. ",,,,"Higher LI value obtained within Wernicke's area (0.76) than within Broca's area (0.35). ",,"No data comparing the two types of baselines (fixation versus perceptual control). ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Number counting","Object naming",,,,,,,-0.23,0.35,,"Relatively weak laterality found for number counting subtracting object naming; only just left lateralised in Wernicke's area, weakly right lateralised in Broca's area. ",,,,"Almost crossed dominance across Broca's area and Wernicke's area. Relatively weak left lateralisation in Wernicke's area (0.35) compared to very weak right lateralisation in Broca's area (-0.23). ",,,2,,,,,,,,,,,,,,,,,,,,,2,"No correlation computed between task LIs. ",,,,"Possible group crossed lateralisation for number counting>object naming- weakly right lateralised in Broca's area but left lateralised in Wernicke's area. ",,,,"Automated language paradigms appear to be more weakly lateralising than non-automated speech tasks. ",2
16,"Clements, Rimrodt, Abel, Blankner, Mostofsky, Pekar et al.",2006,"Sex differences in cerebral laterality of language and visuospatial processing. ",,30,"15 male, 15 female",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Gives LI values for a rhyming task within a language and a non-language ROI. Main focus of paper was on sex differences. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts within the standard LI equation for voxels surviving a single fixed threshold within each ROI independently. ","Not given","Not specified. ",2,1,"Defined anatomically using the Wake Forest University PickAtlas. ",2,5,1,0,0,0,0,1,"Inferior frontal gyrus,     Inferior parietal lobe. ",1,,2,1,3,0,0,0,0,0,0,0,0,"Rhyming decision","Decide if a pair of pronounceable non-word strings rhymed or not, using a button press. ",1,2,2,2,"Perceptual decision- decide if two sets of oriented lines had the same pattern or not. ",3,2,1,0,0,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Rhyming decision","Perceptual decision",,,,,,"0.41 (males), 0.50 (females)","0.89 (males), 0.66 (females)",,,"Strong laterality found within frontal ROI for rhyming decision. Weaker laterality found within inferior parietal lobe. ",,"Greater variability in LIs within the IPL ROI than within the IFG- SD for IFG were 0.16 for males and 0.4 for females, SD for IPL were 0.63 for males and 0.56 for females. ",,"Laterality was stronger within the IFG than the inferior parietal lobe for rhyming decision task. ",,"Note that an active baseline was used- yielded strong laterality for rhyming decision task within frontal ROI. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,"Males showed greater lateralization for the phonological task, whereas females showed greater lateralization for the visuospatial task (results not reported here). ","Strong laterality obtained for a rhyming decision task using an active perceptual decision baseline within a frontal ROI. ",2
17,"Cousin, Peyrun, Pichat, Lamalle, Le Bas, Baciu",2007,"Functional MRI approach for assessing hemispheric predominance of regions activated by a phonological and a semantic task. ",,11,"11 male, 0 female",1,2,,0,,0,1,0,0,1,0,0,0,0,0,"Investigated the utility of the flip method of LI calculation using a phonological and a semantic task. ",2,2,2,2,0,0,0,0,0,0,,,1,0,0,0,0,0,,"Used the flip method (Baciu et al, 2005). First, two sets of images are created; right side images (normalised L and R) and a mirror images set (flipped so that LH is on the right). Then did direct statistical comparisons between hemispheres for each task without taking into account the control condition e.g. (rhyming unflipped + non-rhyming unflipped) > (rhyming flipped + nonrhyming unflipped). Then added analyses using the flip method in which the control condition was taken into account.   ",,"Visual inspection of contrast images (flipped versus unflipped). ",2,2,"Based on functional maps provided by the task vs control contrast. ","1 large VOI",5,1,1,1,1,0,0,"Frontal cortex,     Temporal cortex,     Parietal cortex,     Cerebellum,     Thalamus. ",1,,2,2,9,0,1,1,0,0,0,0,0,"Rhyme detection task,     Semantic Categorization task. ","Rhyme detection- decide if pseudowords rhymed with /e/.    Semantic categorization- decide if words belonged to 'animal' or 'plant' categories. ",1,2,2,2,"Visual detection task- perceptual judgement of whether words of unreadable characters had at least one character which overshot the others. ",3,2,1,1,0,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,"Semantic categorisation ","Perceptual decision",,,,,,,,,,"Leftward asymmetry found across various areas during direct statistical comparison between the hemispheres.  ",,,,"Left hemispheric dominance shown in the IFG, superior and inferior temporal gyrus and occipito-parietal cortex (superior parietal lobule). ","Two methods used: Comparison of flipped and unflipped images for living and non-living items, and comparison of flipped and unflipped images including the control condition. Revealed that left dominance for IFG was related to categorization (living vs non-living conditions), whereas ITG and occipitoparietal cortex were likely to be right dominant regions related to the control condition. ","When the control condition was taken into account, obtained more left hemisphere activated regions than when only hemispheres were contrasted for the task.     Inclusion of the control task when comparing flipped and unflipped images for each condition (living vs non-living) revealed that asymmetry of some regions (ITG and occipito-parietal cortex) was attributable to the control and likely to in fact be right lateralised. ",2,"Rhyming decision","Perceptual decision",,,,,,,,,,"Leftward asymmetry found across various areas during direct statistical comparison between the hemispheres.",,,,"Leftward asymmetry found in the IFG, motor cortex, inferior temporal gyrus, lateral occipital gyrus, SMG, occipito-parietal cortex, precuneas, and parahippocampal gyrus. ","Two methods used: Comparison of flipped and unflipped images for living and non-living items, and comparison of flipped and unflipped images including the control condition. Revealed that left dominance for ventral IFG and SMG was related to rhyming (task condition) whereas the other areas were likely to be right dominant regions related to the control condition. ","When the control condition was taken into account, obtained more left hemisphere activated regions than when only hemispheres were contrasted for the task.",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"The flip method can be used to assess the significance of hemispheric asymmetries across different areas, and whether such asymmetries relate to the condition of interest or to the control task. Taking the control task into account when comparing the hemispheres yielded more left lateralised areas. ",2
18,"Deblaere, Backes, Hofman, Vandemaele, Boon, Vonck et al. ",2002,"Developing a comprehensive presurgical functional MRI protocol for patients with intractable temporal lobe epilepsy. ",,9,"3 male, 6 female",1,2,,0,,0,0,0,0,1,0,0,0,0,0,"Compared 4 language paradigms in terms of strength and reliability of lateralisation; evidence for the particular strength and reliability of word generation. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts above a single fixed threshold within the standard LI equation. ","Not given","Not specified. ",1,1,"Whole hemisphere used. ",1,6,0,0,0,0,0,1,"Whole hemispheres. ",1,,2,4,9,1,1,0,1,1,0,0,0,"Word generation,     Semantic decision,     Reading,     Naming.      ","Covert letter cued word generation.     Semantic decision of whether animals are mammals or not using button press.     Covert reading of meaningful text with semantic content.     Overt naming of drawings of objects.",1,3,3,3,"Cross fixation for word generation.     Vowel decision (decide if vowel /a/ is in vowel string) for semantic decision.     Reading of nonsense word text for reading.     Line task (name position of intersection of four lines) for naming. ",4,2,1,1,1,0,1,1,0,,2,"Letter cued word generation","Cross fixation",,,,,,,,,0.67,"Gave the strongest lateralisation in all subjects out of all language tasks tested. ",,"Achieved the most consistent language lateralisation across subjects out of the language tasks tested. ",,"LI only calculated for whole hemispheres, but reported left dominant activity in the IFG, temporo-occipital region, posterior parietal lobule and anterior thalamus.",,"Standard baseline of cross fixation gave strong laterality (0.67). ",2,"Semantic decision","Vowel decision",,,,,,,,,0.32,"Relatively weaker lateralisation found. Group mean indicates clear left lateralisation, but failure to demonstrate strong lateralisation in four subjects. ","Less reliably lateralising- failure to lateralise activity in four out of nine subjects. ","Greater variability- less consistency in lateralisation. Failure to lateralise in four out of nine subjects. ",,,,"Attributed weaker and less consistent lateralisation for semantic decision task to the choice of baseline task. The vowel decision task may have generated too much language related left activation. ",2,,,,,,,,,,,,,,,,,,,2,"Covert text reading","Covert nonsense text reading",,,,,,,,,0.21,"Relatively weak lateralisation (0.21). ",,,,"Generated bilateral fusiform gyrus activation and leftward activation in the middle temporal gyrus and anterior temporal pole in all subjects. Argued that this task is suitable for assessing hemispheric dominance in the temporal lobe.  ",,"Baseline of covert nonsense text reading yielded weak lateralisation. ",2,"Overt picture naming","Line position naming",,,,,,,,,0.08,"Very weak lateralisation (0.08)- poorest lateralising task out of those tested. ",,,,"Yielded symmetrical bilateral mesial temporal lobe activation. ",,"Baseline of line position naming yielded very weak lateralisation. Argued that this baseline may induce similar speech/language processes (speech production) and so subtract out true lateralisation. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"Did not calculate cross-task correlations. ",,,,,,,,"Rank ordering of language tasks from strongest LI to weakest: Word generation, semantic decision, reading, naming. Confirms the strongly lateralising power of word generation, and provides some suggestions of the effects of baseline on LI; in particular, argued their baselines for the other tasks may have engaged language processes to too high an extent. ",2
20,"Dodoo-Schittko, Rosengarth, Doenitz, Greenlee",2012,"Assessing language dominance with functional MRI: The role of control tasks and statistical analysis. ",,11,"3 male, 9 female",1,2,,0,,0,1,0,0,1,0,0,0,0,0,"Argue for the need for combined task analysis and the use of high-level control tasks in order to obtain strong, unambiguous LIs. Found that strength of LIs and the effect of parameters such as the threshold used varied according to the language paradigm used. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts within ROIs for the standard LI equation. Group LIs were calculated at a single fixed threshold but also calculated LIs for each subject across a range of threshold levels to show threshold dependence. ","Not given","Not specified or reported. ",2,1,"Anatomically defined ROIs associated with primary language functions using the Anatomical Automatic Labelling atlas. ",6,3,0,0,0,0,0,0,"Pars opercularis and pars triangularis,     Inferior frontal gyrus,    Superior temporal gyrus,     Angular and supramarginal gyri. ",1,,2,4,9,1,1,0,0,0,0,0,0,"Verb generation,    Antonym generation,     Verb congruency decision,    Antonym decision. ","Generation tasks: Generate a verb/antonym to go with the presented noun/adjective.     Decision tasks: Decide if noun matches a verb semantically or if two words are antonyms or not. ",2,2,3,2,"Generation control: Required to subvocally change the order of two syllables of a pseudoword.     Decision control: Colour rule task which required discovery of a rule which associated colours with consonants within letter rows and decision as to whether this rule is conformed to or violated by the current set of letter rows. ",4,2,1,1,0,0,0,1,0,"Argued that generation tasks are more challenging than decision tasks as they involve additional proecsses of selection and generation of a word, while the decision tasks simply involve retrieval of one or more specific (provided) words and evaluation of their distance in semantic space. ",2,"Verb generation, antonym generation.","Subvocal syllable manipulation, rest",,,,,,"0.445, 0.678, 0.453",,,,"Strength of lateralisation for single tasks depended greatly on the baseline task used (passive rest versus active control task). ",,,"Not robust against changes in threshold level (normal threshold dependence). ",,"Demonstrated the threshold dependence of LIs from generation tasks. ","Compared effect of baseline on LI; rest as a baseline yielded lower mean LI than the active syllable manipulation task:  VG vs rest = 0.445  VG vs active control = 0.678  AG vs rest = 0.453  AG vs active control not given on own, only as part of conjunction.   ",2,"Verb congruency decision, Antonym decision","Colour rule decision",,,,,,,,,0.536,"LI values not given for semantic decision tasks independently, only within a conjunction (see CTA). Conjunction of verb decision and antonym decision tasks yielded an LI of 0.536, lower than that of the conjunction of generation tasks. ",,,"LIs based on the conjunction of decision tasks were more robust against changes in threshold level (less threshold dependent) than LIs based on the conjunction of generation tasks. ",,"Less threshold dependence seen for conjunction of decision tasks than for conjunction of generation tasks. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Conjunction 1: Generation tasks compared to resting baseline.     Conjunction 2: Generation tasks compared to active control task.     Conjunction 3: Decision tasks compared to active control task. ","Not given. ","Resting,    Syllable manipulation,    Colour rule decision.",,,,,,"0.507, 0.826, 0.536.",,,,"The conjunction of generation tasks versus active control (0.826) yielded stronger LIs than the tasks alone. However, the contrast of verb generation versus active control yielded stronger LIs than conjunction of generation tasks versus baseline (0.507). The conjunction of generation tasks versus active control yielded stronger LIs than the conjunction of decision tasks versus active controls (0.536), which was itself stronger than the conjunction of generation tasks versus resting baseline.","No correlations reported. ",,,"Conjunction analyses did not appear more robust against threshold level than single task analyses, excepting the conjunction of decision tasks which was the least threshold dependent (but single decision task LIs not reported). ",,,"Demonstrated that the use of a higher-level active baseline can be more beneficial in yielding a strong LI than conjunction analysis on two tasks using a low-level resting baseline. When comparing generation and decision conjunctions, decision versus active control was stronger than generation versus resting baseline. ",2,,,,,,,,,"Demonstrated strong effects on laterality of both single versus combined task analysis, and resting baselines versus active control conditions. An active control task may make a STA more strongly lateralised than a CTA using resting baselines. In general, CTA yields the strongest LIs. Generation tasks yield stronger LIs than decision tasks, but the latter may be more robust against threshold dependence. ",2
21,"Doucet, Pustina, Skidmore, Sharan, Sperling,Tracy",2015,"Resting-state functional connectivity predicts the strength of hemispheric lateralization for language processing in temporal lobe epilepsy and normals. ",,23,,4,1,"55 patients with unilateral temporal lobe epilepsy. ",0,,0,0,0,0,0,0,0,0,0,1,"Main focus of paper was on using resting state functional connectivity as a predictor of language dominance. Demonstration of lateralising ability of word generation task. ",2,1,,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used the LI toolbox on thresholded images, citing Wilke and Schmithorst (2006). Specific details of their calculation method not given, but reported that they compared the LIs obtained with this method with those obtained from the bootstrapping method. ",,,2,1,"Anatomically defined mask using the AAL atlas. ",3,3,0,0,0,0,0,0,"Inferior frontal cortex,     Middle and superior temporal cortex. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Verb generation","Covert generation of a verb in response to a visually presented object noun.",1,2,1,1,"Passive fixation. ",2,2,1,1,0,0,0,1,0,,2,"Verb generation","Passive cross fixation",,,,,,0.59,,,,"Strong lateralisation obtained (0.59). ",,"SD of 0.29. ",,,"Compared the LIs obtained with their calculation method with those from a bootstrapping analysis and found high correlations (0.84 and above).  ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Demonstration of strong laterality (0.59) for word generation task. ",2
22,"Drager, Jansen, Bruchmann, Forster, Pleger, P'zwitserlood et al.",2004,"How does the brain accommodate to increased task difficulty in word finding? A functional MRI study. ",,14,"8 male, 6 female.",1,2,,0,,0,0,0,0,0,0,0,1,0,0,"Assessed the effects of task difficulty on the lateralisation of language and other areas. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts with the standard LI equation above a single fixed threshold of p = .001. ","Not given","Not specified. ",2,2,"Defined functionally using the contrast word retrieval tasks - baseline task. Chose the activation cluster corresponding to the left IFG and then reflected it through the midline to obtain its right homologue. ",1,1,0,0,0,0,0,0,"Inferior frontal gyrus. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Word stem completion","Silently generate as many words as possible beginning with the visually presented word beginning. Task was performed at three difficulty levels. ",1,2,1,2,"Silent repetition of a visually presented word. ",4,2,1,0,0,0,0,1,0,,2,"Word stem completion","Silent nonword repetition",,,,,,,,,,"No significant difference in mean LI across the three task difficulty levels was found. Strong laterality found in all conditions. ",,,"LI was robust against changes in task difficulty. ",,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Overall, task difficulty had no significant effect on lateralisation within a frontal ROI for a verbal fluency task. ",2
23,"Fesl, Bruhns, Rau, Wiesmann, Ilmberger, Kegel et al.",2010,"Sensitivity and reliability of language laterality assessment with a free reversed association task- a fMRI study. ",,39,,5,2,,0,,0,1,0,1,1,1,0,0,0,0,"Tested a novel verbal association task as a task for measurement of lateralisation. Assessed strength of LI across multiple regions and reliability of LI measurement across two sessions. ",2,1,1,1,0,0,0,0,0,0,,1,6,1,1,0,0,0,"Single fixed threshold,    Variable threshold. ","Used voxel counts within 42 ROIs to calculate LI using the standard LI equation. Calculated both at a single fixed threshold and using variable thresholds defined according to that threshold level that yielded a criterion number of above threshold voxels (7000). ",0.33,"Calculated LI values such that -1 indicated left dominance and +1 indicated right dominance.   Left = below 0.33  Right = above +0.33  Bilateral = between -0.33 and +0.33",2,1,"Used the AAL atlas to define 42 ROIs associated with language function. Also combined these to form frontal, central, SMA, parietal and temporal ROIs. ",42,5,1,1,0,1,0,0,"Frontal cortex,     Temporal cortex,     Motor cortex,     Parieto-occipital cortex,     SMA.",2,"Mean interval of 1 week between sessions. ",2,1,8,0,0,0,0,0,0,0,0,"Reversed association task. ","Presented a series of words auditorily to a subject, all semantically related to a target word. Subject then required to overtly produce the target word (= reverse association). ",1,1,3,2,"Phoneme sequence listening/detection- listen to unintelligible phoneme sequences and respond afterwards by producing the word 'nothing' (matches number of overt responses required in language task). ",2,1,1,1,0,1,1,0,0,"Task covers a wide range of language processes, including associative semantics, auditory perception and articulation. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Reversed association task","Phoneme sequence listening",30,2,7,,0.7692307692307693,,0.68,0.85,0.64,"Strong laterality obtained across a range of language areas. ","Reliability of LI was increased by 8% when LIs were calculated based on variable rather than fixed thresholds. No significant differences in LI were found between sessions. Highly reliable in frontal ROI, with a correlation of r = 0.94 between testing sessions. Slightly higher reliability found for global LI (r = 0.95). ",,,"Significant correlations were found between LIs from any pair of ROIs. Correlation between frontal and temporal ROI LIs was r = 0.82. Overall, laterality was highly consistent across regions. Global LI was more reliable than all regional LIs. ","Use of variable individually defined thresholds yielded more reliable LIs than the use of a single fixed threshold. ","Argued the results suggest their baseline task was optimal in subtracting out bilateral activation and controlling for articulation. ",2,,,,,,,,,,,,,,,,,,,,,2,,,,"High consistency in degree of lateralisation seen across multiple language regions. ",,,,,"This novel free reversed association task produced strong and reliable lateralisation within multiple ROIs. The most reliable LIs were obtained using a global ROI and variable (as opposed to fixed) thresholds. ",2
24,"Gaillard, Balsamor, Xu, Grandin, Braniecki, Papero et al.",2002,"Language dominance in partial epilepsy patients identified with an fMRI reading task. ",,22,"11 male, 11 female.",1,1,"20 partial epilepsy patients. ",0,,0,0,0,1,1,0,0,0,0,0,"Focus was on identifying temporal areas in fMRI in epilepsy patients. Demonstration of the lateralising ability of a read response naming task across frontal and temporal ROIs. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts within ROIs in the standard LI equation, at 3 different threshold levels (t = 3.0, 4.0 and 5.0). ",0.2,"Standard 0.2 cut-offs. ",2,1,"Drew ROI based on anatomical landmarks while blinded to activation patterns. ",3,3,0,0,0,0,0,0,"IFG,    Middle frontal gyrus,    Wernicke's area.",1,,2,1,5,0,0,0,0,0,0,0,0,"Read response naming","Naming to description sentences were visually presented. Subjects silently read them and then silently named the described object with a single word. ",1,2,3,1,"Passive viewing of dot patterns displayed one at a time- matched the task sentences for sentence length. ",3,2,0,1,0,0,0,1,0,"Task involves sentences, not single words, and so is less 'automatic' Requires decoding written language and reading comprehension, and then semantic retrieval. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Read response naming","Dot pattern viewing",,,,,,0.7,0.65,0.67,,"LI values shown at a threshold of t = 3.0, for MFG, IFG and Wernicke's area. LI strength increased at higher thresholds. IFG LI = 0.65, WA LI = 0.67. ",,,"LI was not robust against changes in threshold level. Wernicke's area showed slightly greater robustness compared to frontal regions.     IFG: 3 = 0.65, 4.0 = 0.83, 5.0 = 0.98.     MFG: 3.0 = 0.7, 4.0 = 0.82, 5.0 = 0.96.    WA: 3.0 = 0.67, 4.0 = 0.8, 5.0 = 0.8.","Highest LIs found in frontal ROIs compared to temporoparietal ROI. At lowest threshold (3.0), MFG showed highest laterality. At highest threshold (5.0), IFG showed highest laterality. WA LI never exceeded 0.8. ","LI was threshold dependent. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"The read response naming task is capable of yielding strong laterality in both frontal and temporal ROIs. This task is thus useful for lateralising temporal receptive language areas involved in reading comprehension. ",2
25,"Gaillard, Sachs, Whitnah, Ahmad, Balsamo, Petrella et al.",2003,"Developmental aspects of language processing: FMRI of verbal fluency in children and adults. ",,22,"15 male, 14 female",1,2,,0,,0,0,0,0,1,0,0,0,0,1,"Main focus of paper was on comparing adults and children. Demonstration of the lateralising power of a semantic fluency task within frontal and temporal regions. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts within ROIs with the standard LI equation at a threshold of t = 4.0. ",0.2,"Standard 0.2 cut-offs. ",2,3,"Regions were selected using historical anatomical data and imaging data implicating these areas in language processing. ",4,3,0,0,0,0,0,0,"Frontal lobe ROI (IFG, MFG, SFG),    IFG,     MFG,     Temporal-parietal ROI (MTG, STG, IPL). ",1,,2,1,1,0,0,0,0,0,0,0,0,"Semantic fluency.","Silent generation of words belonging to the presented category. ",1,2,1,1,"Silent rest. ",2,2,1,1,0,0,0,1,0,"Engages the same language processes as word generation (phonological fluency) tasks e.g. word retrieval and decision processes. Thus these tasks generally yield the same degree of lateralisation. ",2,"Semantic fluency",Rest,,,,,,,0.55,0.52,,"Strong laterality in all regions.  Frontal ROI = 0.55  IFG = 0.57  MFG = 0.7  T-P = 0.52",,"Lowest SD in frontal ROI (0.28), then MFG (0.310, then T-P (0.46), then IFG (0.51). Found that activation in the temporal lobe was more inconsistent- attributed to low demands on comprehension. ",,"Strongest laterality found in MFG, weakest in temporal regions. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Demonstration that semantic fluency can yield strong lateralisation particularly in frontal but also in temporal ROIs. ",2
26,"Haberling, Badzakova-Trajkov, Corballis",2011,"Callosal tracts and patterns of hemispheric dominance: A combined fMRI and DTI study. ",,60,"30 male, 30 female",4,2,,0,,0,0,0,0,0,0,0,0,0,1,"Main focus of paper was on relating structure of the corpus callosum to hemispheric dominance for language. Provides laterality measures for word generation task. ",2,1,1,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used the bootstrapping method (Wilke and Schitthorst, 2007) within the LI toolbox- calculates multiple LIs at multiple thresholds. ",0.1,"Left = > 0.1  Right = < -0.1  Bilateral = between -0.1 and 0.1",2,1,"Defined anatomically within LI toolbox. ",1,1,0,0,0,0,0,0,"Frontal lobe.",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation","Letter cued covert word generation ",1,2,1,1,"Cross fixation",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Cross fixation",51,1,8,,0.85,,0.7,,,"LI values given for different groups according to pattern of laterality across the language and landmark tasks.   Typical (left for lang, right for landmark) = 0.7  Both right = -0.24  Both left = 0.6  Mirror reversal = -0.49",,"Low variability (SD = 0.03 for typical group). ",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Demonstration of strong laterality for word generation within a frontal ROI. ",2
27,"Haberling, Steinemann, Corballis",2016,"Cerebral asymmetry for language: comparing production with comprehension.",,94,"42 male, 52 female",4,2,,0,,0,0,0,1,1,0,0,0,0,0,"Aimed to compare lateralisation for production versus comprehension tasks; no support for the idea that production is more strongly lateralised than comprehension. ",2,1,2,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used the bootstrapping method (Wilke and Schitthorst, 2006) from LI toolbox. ",0.2,"Standard 0.2 cut-offs. ",2,1,"Defined anatomically using the WFU PickAtlas. ",2,3,0,0,0,0,0,0,"Word generation: Broca's area (pars opercularis and pars triangularis)    Synonym decision: Broca's area, combined superior and middle temporal gyri.",1,,2,2,9,1,1,0,0,0,0,0,0,"Word generation,    Synonym decision","Letter cued covert word generation.    Synonym decision- decide if two visually presented words had a similar meaning or were different in meaning. ",1,2,3,3,"Word generation: Passive cross fixation.     Synonym decision- perceptual decision task (decide if two letter strings were identical). ",2,2,1,1,0,0,0,1,0,"These tasks engage quite different language processes. The word generation task involves minimal input and generation of words without reference to meaning. The synonym decision task involves visual input and explicit reference to meaning. ",2,"Letter cued word generation","Cross fixation",,,,,,,0.587,,,"Quite strong lateralisation within Broca's area (LI = 0.587). ",,"Most individuals were left dominant across tasks and ROIs, a few were consistently right dominant and a small number showed inconsistent lateralization patterns. ",,,,,2,"Synonym decision","Perceptual decision",,,,,,,0.598,0.582,,"Relatively strong lateralisation (mean LI around 0.59). ",,"Most subjects were consistently left lateralised in both ROIs, and few were right lateralised and a few inconsistently lateralised. Fairly large spread due to mixed handedness within the sample. ",,"Strength of lateralisation was independent of whether a frontal or temporal ROI was used. The concordance rate for dominance across the two ROIs within the synonym decision task was 93%. Correlation between LIs obtained from frontal and temporal ROIs was .416 for right handers and .615 for left handers. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"High correlations between tasks for both LI value and dominance classification. Concordance rate for language dominance between the two tasks was 87% for the two frontal ROIs and 86% for the frontal and temporal ROI. Correlation between LIs obtained with the two tasks within Broca's area was .516 for right handers and .634 for left handers. ",,,"High concordance between laterality within frontal and temporal ROIs within a synonym decision task. Also high concordance across tasks within both frontal and temporal ROIs. The highest correlation was seen between frontal and temporal LIs for the synonym task (for both right and left handers). ","Reported some cases of crossed lateralisation between frontal and temporal ROIs, including both right and left handers. ",,"Found a significant main effect of handedness, in which right handers showed greater left hemispheric activation than left handers. ",,"Laterality was almost identical across ROIs and tasks, indicating that lateralization is no more pronounced for production than for comprehension. However, cases of crossed laterality were found. ",2
28,"Harrington, Buonocore, Farias.",2006,"Intrasubject reproducibility of functional MR imaging activation in language tasks. ",,10,"4 male, 6 female",1,2,,0,,0,0,0,1,1,1,0,0,0,0,"Aimed to test the reproducibility of fMRI lateralisation, and whether this might vary according to the ROI and task used. Found that reproducibility within ROIs depended on the task used. ",2,1,3,1,0,0,0,0,0,0,,1,6,1,0,1,0,1,"Single fixed threshold,    Multiple thresholds/unthresholded.","Four methods of LI calculation:  LI volume (voxel count) at threshold of P < .0001  LI volume (voxel count) at threshold of P < .001  LI magnitude of F statistic at threshold of P < .01  LI magnitude of F statistic averaged across multiple thresholds.","Not given","Not specified.",2,1,"Defined anatomically using the Talairach atlas. ",2,3,0,0,0,0,0,0,"IFG,     Temporoparietal cortex (supramarginal, inferior parietal, middle temporal and superior temporal gyri). ",2,"Not specified.",2,6,9,1,1,0,0,1,1,1,0,"Verb generation,     Semantic decision,     Confrontation naming,     Story listening,     Visual and oral sentence comprehension.","Verb generation- generate verbs in response to aurally presented nouns.     Semantic decision- decide covertly whether word represented an abstract or concrete concept.     Naming- covert naming of line drawings.    Story listening- listen to passages.     Visual and oral sentence comprehension- read or listen to sentence and decide if it is semantically and syntactically correct ",2,2,3,3,"Verb generation, story listening, oral sentence comprehension- passive listening to backwards speech.     SD- perceptual decision on letters (upper or lower case).    Naming- attend to visual patterns of lines.     Visual sentence comprehension-  attend to strings of lines (similar length to sentences).",4,3,1,1,1,1,0,1,0,"Wide range of language processes assessed. ",2,"Verb generation","Backwards speech listening",,,,,,,0.91,0.83,,"Strong lateralisation observed in both ROIs. ","Very high test-retest correlation between LIs from the two sessions within both frontal and temporoparietal ROIs for verb generation ( r >= 0.9)","One subject showed strong right lateralisation; their LI values were not included for group averages. ",,"Slightly higher LI within frontal ROI than temporoparietal ROI. ","Method using magnitude of F statistic averaged over multiple thresholds yielded the highest LIs. This method also yielded LIs with higher test-retest correlations across sessions than those yielded by using the volume of activation in each hemisphere in laterality calculations. The same was true of LIs calculated based on F statistics at a single threshold. Thus, overall magnitude measures were more strongly lateralising and more reproducible than extent measures. ",,2,"Semantic decision","Perceptual decision",,,,,,,0.73,0.43,,"Fairly strong lateralisation, particularly in frontal ROI, less so in temporoparietal ROI. ","Fairly high test-rest correlations across sessions, of just over 0.7. Reliability within frontal and temporoparietal ROIs was very comparable. ",,,"Stronger lateralisation found in frontal ROI (about 0.7) compared to temporoparietal ROI (about 0.4). Both ROIs yielded comparable high test-retest correlations. ","Overall magnitude measures were more strongly lateralising and more reproducible than extent measures. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Picture naming","Line pattern viewing",,,,,,,0.55,0.28,,"Moderate to weak laterality obtained across frontal and temporoparietal ROIs respectively. ","Fairly high test-retest correlations across sessions within both ROIs; around 0.75 for frontal ROI and around 0.7 for temporoparietal ROI. Thus, very slightly higher reliability within frontal than temporal ROI. ",,,"Stronger laterality found within frontal ROI; weak asymmetry in temporoparietal ROI (around 0.3). Frontal LIs were also slightly more reliable than temporal LIs (but both were reproducible). ","Overall magnitude measures were more strongly lateralising and more reproducible than extent measures. ",,2,"Story listening","Backwards speech listening",,,,,,,0.3,0.5,,"Weak to moderate laterality found across frontal and temporal ROIs respectively. ","Very high test-retest correlations were found for story listening within the temporal ROI (r >= 0.9). Low reproducibility found within the frontal ROI (r = around 0.5). ",,,"Stronger and more reliable lateralisation found for a temporoparietal ROI compared to a frontal ROI. ","Overall magnitude measures were more strongly lateralising and more reproducible than extent measures. ",,2,"Visual and auditory sentence comprehension","Slash string viewing, backward speech listening",,,,,,,"0.45, 0.66.","0.6, 0.8",,"Stronger LIs found within temporal versus frontal ROIs for both visual and auditory sentence comprehension. Visual comprehension yielded stronger laterality compared to auditory comprehension. ","Very high test-retest correlations across ROIs for the visual sentence comprehension task (r >= 0.9). High test-retest correlations for auditory sentence comprehension within temporal ROI (r close to 0.9), but showed much weaker reproducibility within frontal ROI (r = 0.6). ",,,"Temporoparietal ROIs yielded LIs that were stronger across both auditory and visual sentence comprehension and more reproducible for auditory sentence comprehension. Visual sentence comprehension demonstrated comparably high reproducibility across both ROIs. ","Overall magnitude measures were more strongly lateralising and more reproducible than extent measures. ",,2,,,,,,,,,,,,,,,,,,,2,"Three sets of CTAs tested:  CTAa- Semantic decision, naming, verb generation.   CTAb- story listening, and auditory and visual sentence comprehension.  CTAc- Combination of all 6 tasks. ","Not given. ","Multiple (see single tasks). ",,,,,,,0.7,0.6,,"Highest LI given by the CTA including SD, naming and verb generation (about 0.8). Weakest LI given by CTA including sentence comprehension and story listening tasks. ",,"All task combinations produced high test-rest correlations between LIs across ROIs (r >=0.9). However, the test-retest correlations for the combinations of tasks were no better than those seen for the verb generation task alone. Overall argued that the reproducibility of laterality for CTA was only slightly improved relative to that seen for the best single tasks. ",,,"Stronger LIs found within frontal ROI versus temporal ROI for all task combinations. High reproducibility seen regardless of ROI across combinations. ",,,2,"Not given. ",,,,,,,,"Overall, the reproducibility of LIs within a given ROI depends on the task used; reproducible LIs can be found within temporal as well as frontal ROIs, provided the right task is used (e.g. story listening). Magnitude LIs were more reproducible than extent LIs. Overall, verb generation was superior in terms of strength and reproducibility of LIs across ROIs (comparable to reproducibility of CTA). ",2
29,"Hernandez, Andersson, Edjlali, Hommet, Cottier, Destieux et al.",2013,"Cerebral functional asymmetry and phonological performance in dyslexic adults. ",,16,"16 male, 0 female",1,1,"15 right handed males with a diagnosis of developmental dyslexia. ",0,,0,1,0,0,1,0,0,0,0,1,"Main focus of paper was on correlating functional asymmetry with behavioural performance in developmental dyslexia. Demonstration of lateralisation for a rhyming task using the flip method. ",2,1,2,2,0,0,0,0,0,0,,4,1,0,0,0,0,0,,"Used the flip method (Baciu et al, 2005). First, two sets of images are created for the contrast rhyming vs font matching; right side images (normalised L and R) and a mirror images set (flipped so that LH is on the right). Then did an ANOVA with 2 factors (group and hemisphere) using contrast images (rhyming vs font matching) obtained with flipped and unflipped images for both groups.   Then computed an LI based on the number of activated voxels in ROIs.   ",,,2,2,"ROIs were defined as spheres centred on peaks of activation identified in the interhemispheric analysis (flip method). Direct statistical comparisons done in all 6 ROIs; LI values calculated using voxels within IFG opercularis and occipitotemporal junction only. ",6,5,1,1,0,1,1,0,"Inferior temporal gyrus,     Precentral gyrus,     Superior occipital gyrus,     Opercular part of IFG,     Triangularis part of IFG,     Parietal lobule. ",1,,2,1,3,0,0,0,0,0,0,0,0,"Rhyming task. ","Subjects had to decide if two visually presented words rhymed. Words differed in their levels of orthographic transparency. ",1,2,2,2,"Font matching task- decide if pairs of unreadable fonts were matching or non-matching. ",2,2,1,0,0,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Rhyming decision","Perceptual decision (font matching)",,,,,,,,,,"Hemispheric dominance compared using direct statistical comparisons between hemispheres (flip method). LIs also calculated but not reported. ",,,,"Statistical comparisons (flip method) revealed a significant hemisphere effect in all ROIs. LI analysis revealed strong leftward asymmetry in the IFG. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Rhyming decision task yields significant lateralisation across multiple ROIs using the flip method of interhemispheric comparison. ",2
30,"Hund-Georgiadis, Lex, von Cramon.",2001,"Language dominance assessment by means of fMRI: Contributions from task design, performance, and stimulus modality. ",,14,"7 male, 7 female",1,2,,0,,0,0,0,0,1,0,1,0,0,0,"Compared active and passive baseline tasks and the effect of stimulus modality on lateralisation; the former was found to have a significant effect whereas the latter did not affect laterality. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used number of activated pixels above a fixed threshold of Z = 5.5 within an ROI for the standard LI calculation. ","Not given","Not specified. ",2,1,"Identified anatomically using the Talairach 3D reference system. ",1,1,0,0,0,0,0,0,"Broca's area ",1,,2,3,9,0,1,0,0,0,0,0,1,"Semantic encoding/decision,    Lexical encoding/decision. ","Visual Semantic encoding- decide if visually presented word is abstract or concrete.     Auditory Semantic encoding- decide if aurally presented word is abstract or concrete.    Lexical encoding- decide if word is noun or verb. ",1,2,2,3,"Fixation,    Visual perceptual decision- decide if upper or lower case,    Auditory perceptual decision- decide if speaker is male or female. ",2,3,0,1,1,1,0,1,0,"Suggested that lexical encoding task may engage syntactic processes also (to classify word as noun or verb). ",2,,,,,,,,,,,,,,,,,,,2,"Semantic decision","Fixation, perceptual decision",,,,,,,"0.028, 0.5",,,"LI strength ranged from very weak (0.028) to moderate (0.5) depending on baseline used, fixation versus active perceptual decision respectively. Note these values are based on visual semantic decision. ",,,,,,"Strong effect of baseline; laterality increased from near zero (0.28) to moderately left lateralised (0.5) when baseline was changed from fixation to active perceptual decision. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Lexical decision","Perceptual decision",,,,,,,"-0.047, 0.17",,,"LIs indicated weak lateralisation; lexical encoding evoked bilateral activity. ",,,,,,"Baseline used did affect the LI obtained; increased from slightly negative (-0.047) to slightly positive (0.17) when baseline was changed from fixation to active perceptual decision. ",2,,,,,,,,,,,,,,,,,,,,,2,"A significant correlation was found between semantic and lexical decision tasks regardless of the control task used (r = 0.51). ",,,,,"A higher mean LI was obtained for visual semantic decision (LI = 0.5) versus auditory semantic decision (LI = 0.41). However, correlation between LIs from the auditory and visual task was high (r = 0.91). ","No significant effect of whether the dominant or non-dominant hand was used for responding on LI values. ","No significant effect of task performance on LI values; no correlation found. ","Highest laterality was obtained with semantic decision compared to perceptual decision. Active baseline yielded higher LIs than passive fixation baseline. Weak laterality (i.e. bilateral activation) seen for lexical encoding task, which may be attributable to engagement of syntactic processes. No significant effect of stimulus modality, task performance, nor response hand found. ",2
31,"Hund-Georgiadis, Lex, Friederici, von Cramon.",2002,"Non-invasive regime for language lateralization in right- and left-handers by means of functional MRI and dichotic listening. ",,"34 ","18 male, 16 female",4,2,,0,,0,0,0,1,1,0,0,0,0,0,"Paper compared global and regional LIs, for both semantic encoding and lexical encoding tasks. LI depended on baseline, encoding task and ROI. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used the number of activated pixels above a threshold of Z = 5.5 within ROIs for the standard LI equation. ",0.1,"Left = LI > 0.1  Right = LI < -0.1  Bilateral = -0.1 <= LI <= 0.1",3,1,"Defined anatomically using Talairach coordinates. ",3,3,0,0,0,0,0,0,"Broca's area,     Superior temporal gyrus,     Whole hemisphere (global LI).",1,,2,2,9,0,1,0,0,0,0,0,1,"Semantic encoding/decision,     Lexical encoding/decision. ","Semantic decision- decide if word is abstract or concrete.     Lexical encoding- decide if word is noun or verb. ",1,2,2,3,"Rest,     Perceptual encoding task- decide if word is in upper or lower case letters. ",2,2,0,1,1,0,0,1,0,"Suggested that the lexical encoding task (noun or verb) may engage syntactic processes. ",2,,,,,,,,,,,,,,,,,,,2,"Semantic decision","Fixation, perceptual decision",25,8,1,0,0.7352941176470589,,"0.5, 0.19","0.49, 0.07","0.52, 0.14","LI values given are for semantic decision compared to perceptual decision, for right and left handers separately. This was the most strongly lateralising protocol tested in this study (LI of around 0.5 across ROIs). ",,"Right handers Broca's SD = 0.39.   Left handers Broca's SD = 0.59.     RH global SD = 0.34.   LH global SD = 0.6    RH temporal SD = 0.2.  LH temporal SD = 0.4.",,"Highly comparable LI strength obtained across frontal, temporal and global LIs, for both right and left handers. Significant and strong correlations found between frontal, temporal and global LIs. No cases of crossed dominance across frontal and temporal ROIs.  ",,"Use of fixation as baseline yielded significantly lower laterality than when an active perceptual encoding task was used as a baseline. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Lexical encoding","Fixation, Perceptual decision",,,,,,,,,,"LI values for lexical encoding were weak; approximated from graph at around 0.2 using a perceptual decision baseline, and very close to 0 using a fixation baseline (within Broca's area ROI). ",,,,"LI values for this task only given within Broca's area. ",,"The use of an active versus passive baseline had a strong effect on the laterality obtained; very weak (near zero) mean LI for fixation, moderately weak (around 0.2) mean LI for perceptual decision baseline. ",2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"Strong effect of handedness on LI values (see LI values). ",,"Strong laterality found for a semantic decision task vs perceptual encoding baseline across both frontal, temporal and global LIs (i.e. no effect of region). Weaker laterality found for a lexical encoding task, and for both tasks when a fixation baseline was used. ",2
32,"Hunter, Brysbaert.",2008,"Visual half-field experiments are a good measure of cerebral language dominance if used properly. ",,10,"6 male, 4 female",2,2,,0,,0,0,0,0,1,0,0,0,0,1,"Main focus of paper was on validating the visual half-field method of lateralisation. Demonstration of lateralising ability of word generation task. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used number of voxels active above threshold within ROI for the standard LI equation. ",0.4,"Left = LI > 0.4  Right = LI < -0.4  Bilateral = -0.4 < LI < 0.4",2,1,"ROIs were anatomically predefined (precise method not given). ",1,1,0,0,0,0,0,0,"Inferior frontal cortex. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation.","Covert letter cued word generation. ",1,2,1,2,"Silent repetition of non-word (dada)",2,2,1,0,0,0,0,1,0,"Suggested that semantic fluency likely involved greater working memory demands than letter fluency, since subjects produced a significantly greater number of words in semantic fluency. ",2,"Letter cued word generation","Silent nonword repetition",6,2,2,0,0.6,,"0.36, 0.848",,,"Yielded strong laterality within those left lateralised for language (mean LI of left lat. group = 0.848). Weak laterality found in 2/10 subjects. Those right lateralised for language showed strong rightward lateralisation (LI scores of -0.966 and -0.66).",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Strong laterality was yielded by the word generation task within Broca's area within a sample of left handers; weak laterality (bilaterality) was only found in 2 out of 10 subjects, with the rest showing clear and strong asymmetry. ",2
33,"Jensen-Kondering, Ghobadi, Wolff, Jansen, Ulmer.",2012,"Acoustically presented semantic decision-making tasks provide a robust depiction of the temporo-parietal speech areas. ",,20,"10 male, 10 female",1,2,,0,,0,1,0,1,1,0,1,0,0,0,"Aimed to test the optimal paradigm for inducing lateralised activity within temporo-parietal speech areas. ",2,1,3,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used both voxel number and T-value of the activation within pre-defined ROIs for the standard LI equation. ","Not given","Not specified. ",2,1,"Not described- just defined a priori. ",2,3,0,0,0,0,0,0,"Frontal speech areas- inferior and middle frontal gyrus, anterior insula.     Temporoparietal speech areas- superior and medial temporal gyri, supramarginal and angular gyri. ",1,,2,5,9,1,0,0,0,0,0,1,1,"Word generation,     Verb generation,     Semantic fluency,     Semantic violation/decision,    Lexical decision.","Letter cued word generation,     Verb gen- generate verbs from a noun,     Semantic fluency- generate words within a category,    Semantic violation- decide if sentence contains a semantic violation or not,    Lexical decision- decide if a word is a real word or a pseudoword.",1,2,3,2,"Visual baseline- Perceptual decision on direction of arrow within line of dashes.     Acoustic baseline- Tone decision on whether sequence of two tones was ascending or descending in frequency. ",4,3,1,1,0,1,0,1,0,"All language paradigms were presented both visually and auditorily in separate testing sessions, to assess the effects of modality of presentation. ",2,"Word generation, Verb generation, Semantic fluency.","Perceptual decision (acoustic or visual)",,,,,,,"0.95, 0.95, 0.91","0.87, 0.91, 0.97",,"LI strength depended on task, region and modality.  Strongest LI for both WG and VG seen in frontal ROI during auditory presentation (both 0.98).   Strongest LI for SF seen in temporal ROI during auditory presentation (0.98).   Values given are for visual presentation using voxel number. ",,"Mean LIs calculated using voxel magnitude had higher SDs than LIs calculated using voxel count. ",,"LIs within temporoparietal ROIs tended to be lower than LIs within frontal ROIs, excepting visually presented semantic fluency which yielded a stronger LI within the temporal ROI (0.97) compared to the frontal ROI (0.91).   Which task showed the strongest lateralisation depended on the ROI; for a frontal ROI, strongest lateralising tasks were WG and VG jointly, whereas for temporal ROI SF was the most strongly lateralising task. ","Across both ROIs and all tasks, LIs calculated using voxel counts were higher than LIs calculated using t-values (magnitude of signal change). ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Semantic violation detection","Perceptual decision (visual or auditory)",,,,,,,"0.92 (V), 0.94 (A)","0.88 (V), 0.48 (A)",,"LI strength depended on region and modality.  Strongest LI achieved using auditory presentation and a frontal ROI. Auditory presentation yielded lower laterality within temporal ROI compared to visual presentation; reverse was true in frontal ROI. So, effect of modality depended on region.   ",,"Mean LIs calculated using voxel magnitude had higher SDs than LIs calculated using voxel count. ",,"Frontal ROI always yielded stronger laterality than temporoparietal ROI, regardless of stimulus modality.   ","Across both ROIs and all tasks, LIs calculated using voxel counts were higher than LIs calculated using t-values (magnitude of signal change). ",,2,"Lexical decision","Perceptual decision (auditory or visual)",,,,,,,"0.8 (V), 0.62 (A)","0.95 (V), 0.49 (A)",,"LI strength depended on region and modality.  Strongest LI achieved using visual presentation and a temporal ROI. Auditory presentation yielded weaker LIs within both ROIs.   ",,"Mean LIs calculated using voxel magnitude had higher SDs than LIs calculated using voxel count. ",,"Effect of ROI depended on modality of stimulus presentation; stronger LI for temporal vs frontal ROI when visual, but stronger in frontal vs temporal ROI when auditory. ","Across both ROIs and all tasks, LIs calculated using voxel counts were higher than LIs calculated using t-values (magnitude of signal change). ",,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"The strongest lateralisation within a temporal ROI was found with an auditorily presented word generation task. Strong lateralisation within temporal areas also found using visual semantic fluency and lexical decision. Overall, effect of stimulus modality and ROI was task dependent. However, across tasks and ROIs, LIs based on voxel counts were always higher than LIs based on voxel signal magnitude (t-value).  ",2
34,"Kennan, Kim, Maki, Koizu mi, Constable.",2002,"Non-invasive assessment of language lateralization by transcranial near infrared optimal topography and functional MRI. ",,6,"5 males, 1 female",4,2,,0,,0,0,0,0,1,0,0,0,0,1,"Main focus of task was comparing fMRI to near infrared optical topography. Demonstration of the lateralising power of a sentence comprehension task requiring detection of semantic and syntactic errors. ",2,1,1,1,0,0,0,0,0,0,,2,1,0,0,0,0,0,,"Used the number of active voxels above a threshold of t = 1.5 for the standard LI equation. ","Not given","Not specified. ",1,1,"Whole hemispheres used. ",1,6,0,0,0,0,0,1,"Whole hemispheres",1,,2,1,7,0,0,0,0,0,0,0,0,"Syntactic/semantic violation detection","Subjects had to indicate whether a visually presented sentence was fully correct or either semantically or syntactically incorrect. ",1,2,2,2,"Line decision task.",3,2,0,1,1,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Semantic/syntactic violation detection","Perceptual decision (visual line decision)",,,,,,,,,58,"This task yielded relatively strong lateralisation; note that sample included one left hander with a lower LI than the other subjects of 20. Excluding this participant increased the mean group LI to 65.2.",,"Quite high variability in LI value (note that sample had mixed handedness, but 5 subjects were right handed and only one was left handed). ",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"This sentence comprehension task requiring the detection of syntactic and semantic errors yielded relatively strong lateralisation when a global LI was used. ",2
35,"Kleinhans, Mueller, Cohen, Courchesne.",2008,"Atypical functional lateralization of language in autism spectrum disorders.",,14,"Not specified",4,1,"14 males with a diagnosis of ASD. ",0,,0,0,0,0,1,0,0,0,0,0,"Main focus of paper was on comparing language laterality between controls and individuals with ASD. Demonstration of the lateralising ability of two verbal fluency tasks (letter fluency and semantic fluency). ",2,1,1,1,0,0,0,0,0,0,,4,1,0,0,0,0,0,,"Used the summed volume of significant clusters within an ROI for the following LI equation: (L-R)/ 0.5(L+R). This yielded an LI spanning from -2 to +2. ","Not given","Not specified.",2,1,"ROIs were defined anatomically using the Talairach Daemon programme. ",1,1,0,0,0,0,0,0,"Frontal ROI included the superior, middle and inferior frontal gyri and the insula. ",1,,2,2,9,1,0,0,0,0,0,0,0,"Letter fluency,     Semantic fluency. ","Overt letter cued word generation.    Overt generation of items belonging to a visually presented category. ",1,1,1,2,"Repetition of the word 'nothing'. ",2,2,1,1,0,0,1,1,0,,2,"Letter fluency, Semantic fluency. ","Word repetition",,,,,,,"1.68, .66",,,"Significantly stronger lateralisation found for the letter fluency (1.68) than the semantic fluency task (0.66). ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Found significantly stronger lateralisation for a letter fluency task compared to a semantic fluency task using a frontal ROI; attributed to greater recruitment of right prefrontal cortex during semantic fluency. ",2
36,"Knecht, Jansen, Frank, van Randenborgh, Sommer, Kanowski et al.",2003,"How atypical is atypical language dominance?",,14,"7 male, 7 female",4,2,,0,,0,0,0,0,0,1,0,0,0,1,"Main focus of paper was on investigating the nature of 'atypical' dominance for language. Provides data on the reproducibility of LIs using a phonetic fluency task in typically and atypically lateralised subjects. ",2,1,1,1,0,0,0,0,0,0,,2,2,0,0,0,0,0,,"Used the number of active voxels above an individually defined threshold for the standard LI equation. Threshold was defined at a level yielding a fixed number of active voxels (4000). ",N/A,"Dominance had previously been established based on fTCD data, so that 7 subjects were selected with typical and 7 with atypical dominance. ",1,1,"Whole hemisphere used. ",1,6,0,0,0,0,0,1,"Whole hemispheres",2,"Not specified.",2,1,1,0,0,0,0,0,0,0,0,"Word generation","Letter cued covert word generation.",1,2,1,2,"Silent nonword repetition ('baba')",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Silent nonword repetition",,,,,,,,,"67, -82","Mean LI values for left lateralised subjects were 67 for run 1 and 57 for run 2. Mean LI values for right lateralised subjects were -82 for run 1 and -71 for run 2. No evidence that subjects classified as right lateralised for fTCD show increased bilaterality in fMRI compared to those classified as left lateralised. ","Although LIs were calculated for two independent testing sessions, no statistics were done on comparing LIs across the two sessions in terms of reproducibility. Reported some cases of changes in the degree of laterality across testing sessions, but never in the direction of laterality (i.e. dominance). ","No significant differences in the level of variability in LI values between right and left dominant groups; mean variability index of 0.73 and 0.74 respectively. ",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"For word generation, no differences were found in the extent of variability in LIs between a left dominant and a right dominant group (classified by fTCD). No evidence that one group showed greater bilaterality than the other. Demonstration that some individuals do show right dominance for word generation, in a way concordant with fTCD findings. ",2
37,"Krinik, Lehericy, Duffau, Capelle, Chainay, Cornu, et al.",2003,"Postoperative speech disorder after medial frontal surgery: Role of the supplementary motor area. ",,6,"3 male, 3 female",1,1,"12 patients with medial frontal lobe damage. ",0,,0,0,0,1,0,0,0,0,0,0,"Comparison of global and regional LIs for an auditory cued semantic fluency task used whole hemispheres and the SMA. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used the number of activated voxels surviving a fixed threshold for the standard LI equation. Separate LIs calculated using whole hemispheres and SMA as a ROI. ",0.2,"Standard 0.2 cut-offs. ",3,1,"SMA was defined anatomically using anatomical landmarks. ",2,6,0,0,0,1,0,1,"Whole hemispheres,     SMA. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Semantic fluency.","Covert generation of items belonging to an aurally-presented semantic category. ",1,2,1,1,Rest.,3,1,0,1,0,1,0,0,0,,2,"Semantic fluency",Rest,6,0,0,0,1,0.2,,,0.56,"Relatively strong leftward laterality seen for global LI (all subjects were left lateralised) Very weak asymmetry seen for SMA LI, indicating more bilateral activation. For the SMA, 3 subjects were left dominant, 2 subjects were right dominant, and 1 subject was bilateral. ",,"Global SD = 0.15.  SMA SD = 0.32.  LIs were more variable in the SMA than in whole hemispheres. ",,"Global LIs were stronger and less variable than LIs based on the SMA. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"LIs based on voxel counts within the SMA are generally weaker and more variable between individuals than LIs based on whole hemispheres, using a verbal fluency task. ",2
38,"Lohmann, Deppe, Jansen, Schwindt, Knecht.",2004,"Task repetition can affect functional magnetic resonance imaging-based measures of language lateralization and lead to pseudoincreases in bilaterality. ",,1,"0 males, 1 female",1,2,,0,,1,0,0,0,0,0,0,0,0,1,"Demonstration of the effects of task repetition on measurement of laterality; specifically, a decrease in LI (increase in bilaterality). ",2,1,3,1,0,0,0,0,0,0,,2,1,0,0,0,0,0,,"Calculated two separate LIs using the standard LI equation; one using voxel counts, one using mean signal change. The former was calculated at a fixed threshold of P = 0.001 as this was the P value at which the LI was most reproducible and most robust against moderate changes in P. ","Positive versus negative","Positive LIs taken to indicate left dominance. ",3,3,"Global LI- all voxels in the whole brain (excluding the cerebellum).     Regional LI- used previous imaging experience with task to define ROI as the anterior section of the vascular territory of the MCAs. This was guided by anatomic landmarks. ",2,6,1,0,0,0,0,1,"Global LI- whole hemispheres (excl. cerebellum).     Regional LI- areas surrounding the IFG (including Broca's area). ",5,"10 sessions over 2 months with variable intersessional intervals.",2,1,1,0,0,0,0,0,0,0,0,"Word generation.","Silent, acoustically letter cued word generation. Followed by period in which subject had to report the words aloud.",1,3,1,1,Rest.,2,1,1,0,0,1,1,0,0,"Acquired images during both silent generation period and period in which subjects had to report their generated words out loud- so fMRI LI based on both overt and covert processes.",2,"Letter cued word generation",Rest,,,,,,,,,,"Across all sessions, LIs based on both voxel count and signal change magnitude were positive. ","Over the 10 sessions, LIs based on voxel counts decreased significantly i.e. became more bilateral. Attributable to increase in bilaterally activated voxels; difference between activated voxels in and left and right remained constant. Conversely, LIs based on signal change magnitude were variable without a statistically significant trend. Signal change LIs showed higher variability (variability index = 0.63) than voxel count LIs (VI = 0.35). Thus, both methods did not yield reproducible LIs. ",,,,"Voxel count LIs and signal change magnitude LIs showed different patterns of change over repeated testing; voxel count LIs showed a systematic decrease, whereas signal change LIs were variable without any clear pattern or trend. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Task repetition led to 'pseudoincreases' in bilaterality; this was markedly demonstrated in LIs based on voxel counts. LIs based on signal magnitude showed no consistent trend over repeated testing, but were generally highly variable. Note: Based on a single subject.",2
39,"Mazoyer, Zago, Jobard, Crivello, Joliot, Perchey et al.",2014,"Gaussian mixture modelling of hemispheric lateralization for language in a large sample of healthy individuals balanced for handedness. ",,297,"152 male, 145 female",4,2,,0,,0,0,1,0,0,0,0,0,0,0,"Aimed to investigate the nature of 'atypical' language lateralisation, in terms of whether it is a homogenous or heterogeneous group. Used a Gaussian modelling approach to determine categorisation of individual's hemispheric dominance. ",2,2,1,4,0,0,0,0,0,0,,2,4,0,0,0,0,0,,"Used the bootstrapping method (Wilke and Schitthorst, 2006) within LI toolbox. Computed LIs based on whole hemispheres (both grey and white matter), excluding the cerebellum. ",N/A,"Instead of the use of arbitrary cut-offs for dominance categorisation, used Gaussian model fitting to obtain the optimal number of Gaussian functions for each handedness group and across the whole sample; this was used to obtain thresholds to define different dominance categories. ",1,1,"Defined as whole brain excluding cerebellum, both grey and white matter. ",1,6,0,0,0,0,0,1,"Whole hemispheres",1,,2,1,8,0,0,0,0,0,0,0,0,"Sentence generation.","Subjects had to covertly generate a sentence of a pre-specified standard structure to describe a presented cartoon.",1,2,1,2,"Covertly recite the months of the year when a scrambled image was presented instead of a cartoon. These trials were randomly interspersed in an event-related paradigm. ",3,2,0,1,1,0,0,0,0,"Note the use of an active high level reference task, that involved automatic language. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence generation","Generation of automatic sequence of words. ",250,10,37,,0.8417508417508418,,,,47.8,"Mean LI all participants = 47.8   Mean LI Right handers = 53.5   Mean LI Left handers = 43.2   Numbers in each dominance category recorded are for a three category solution from Gaussian modelling: Typical, ambilateral, and strongly atypical. ",,"Higher variability in LI values in left handed group compared to right handed group. ",,,"Gaussian modelling identified 3 and 4 category solutions for right and left handed groups respectively. Both had two 'typical' groups with positive LI values (G1 and G2). These distributions were overlapping. A third 'ambilateral' group was identified with a mean of around 0 (G3). Unique to the left handed sample was a fourth group (G4) of 'strongly atypical', with strongly negative LI values. This group was very rare (10 left handers). Gaussian modelling over the entire sample yielded 4 groups; latter was identical to G4. ",,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"Found no concordance between handedness and LI beyond chance levels, except for in the strongly atypical group, in which strongly right lateralised language was associated with strong left hand preference. ",,"A gaussian modelling approach can be used to identify dominance groups without the need for arbitrary cut-offs. 'Atypical' language laterality can be split into 'ambilateral' and 'strongly atypical' groups. Strong right hemisphere dominance was only found in left handers; otherwise, no relationship between handedness and laterality. ",2
40,"Miro, Ripolles, Lopez-Barroso, Vila-Ballo, Juncadella, de Diego-Balaguer et al.",2014,"Atypical language organization in temporal lobe epilepsy revealed by a passive semantic paradigm.",,19,"10 male, 9 female.",1,1,"23 temporal lobe epilepsy patients. ",0,,0,0,0,1,1,0,0,0,0,0,"Aimed to test a passive language paradigm to determine if this was capable of producing lateralised activity within temporal regions. ",2,1,2,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used the bootstrapping method using voxel values (signal magnitude). ",0.4,"Left = LI > 0.4  Right = LI < -0.4  Bilateral = -0.4 < LI < + 0.4",2,1,"Defined anatomically using the automated anatomical atlas and the WFU PickAtlas. ",1,2,0,0,0,0,0,0,"Single ROI including:  STG  MTG  ITG  Anterior temporal lobe.",1,,2,2,6,0,0,0,0,0,0,0,0,"Passive sentence listening.    Passive pseudosentence listening. ","Subjects listened to spoken sentences that were either well formed or consisted of pseudowords, generated by taking well formed sentences and substituting phonemes based on phonotactic rules. Instructed to attend carefully as they would be asked questions later. ",1,2,2,1,Rest.,3,1,0,1,1,1,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence listening",Rest,3,1,15,,0.15789473684210525,,,0.03,,"Values shown above for sentence listening. Weak laterality for sentence listening with the majority of subjects showing bilateral temporal activity. Same pattern seen for pseudosentence listening (LI = -0.08), 1 left lateralised, 3 right lateralised, 15 bilateral. ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,"This task produced fairly strong right lateralisation in patients, and so was discussed as a suitable task for identifying atypical language dominance in patients. ",,,,,,"This passive sentence listening task was poor at producing lateralised activity within temporal areas when compared to rest. ",2
41,"Morrison, Churchill, Cuimano, Schweizer, Das, Graham.",2016,"Reliability of task-based fMRI for preoperative planning: A test-retest study in brain tumor patients and healthy controls. ",,12,Unknown,4,1,"12 brain tumour patients. ",0,,0,0,0,0,0,1,0,0,0,0,"Aimed to assess the test-retest reliability of fMRI laterality measures using a rhyming and phonemic fluency task. Overall, reliability of fMRI laterality was task-dependent. ",2,1,3,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used the standard LI equation with both voxel counts and signal magnitude. The former was calculated across multiple thresholds to account for its threshold dependence. ",0.2,"Traditional 0.2 cut-offs. ",2,3,"Looked at t-maps for each task and then identified brain regions using the Talairach-Tournoux Atlas. ",1,3,0,0,0,0,0,0,"Single VOI including:   Inferior frontal gyrus,    Superior temporal gyrus,    Angular gyrus,    Supramarginal gyrus. ",2,"20 minutes (within same fMRI session)",2,2,9,1,0,1,0,0,0,0,0,"Word generation,    Rhyming decision. ","Word generation- subjects had to write down words generated in response to visually provided letter using a tablet and stylus within the scanner.     Rhyming decision- decide if pairs of words rhymed.",1,2,3,2,"Word gen- wrote varying lengths of symbol strings (e.g. 888, 8888, 88).     Rhyming- perceptual decision on whether pairs of lines were matching in orientation or not.",3,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Symbol writing",,,,,,,,,,"LI values not given. ","Measured the reliability of both dominance classification and of LI values. Reproducibility for phonemic fluency was lower than that seen for rhyming in both LI extent (66.7%) and LI magnitude (75%). Reduced correlations between LI values across sessions relative to rhyming were found of r = 0.63 for LI extent and r = 0.66 for LI magnitude.",,"Reproducibility of word generation LIs was less robust against changes in the choice of activity measure for LI calculation than that of rhyming decision LIs. ",,"Which method was more reproducible depended on the task used; LI extent and LI magnitude showed fairly comparable levels of reproducibility for rhyming decision (both 100% dominance agreement, and r = 0.91 and 0.9 respectively). For word generation, LI magnitude was slightly more reproducible for both dominance classification (75% versus 66.7% for extent) and LI values correlation (r = 0.66 versus 0.63 for extent). ",,2,,,,,,,,,,,,,,,,,,,2,"Rhyming decision","Perceptual decision",,,,,,,,,,"LI values not given.","Dominance was reproducible in 100% of subjects for the rhyming task with both LI extent and LI magnitude. High correlations seen between LI values across sessions for both LI extent (r = 0.91) and LI magnitude (r = 0.9). ",,"Reproducibility of LIs calculated from rhyming decision contrasts were more robust against choice of activity measure for LI calculation than that of LIs based on word generation contrasts. ",,"LI extent and LI magnitude showed fairly comparable levels of reproducibility in both dominance classification (both reproducible in 100% of subjects), and LI value test-retest correlation (r = 0.91 and 0.9 respectively). ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Reliability of fMRI LI was task dependent. Rhyming decision laterality was more reproducible than that produced by word generation, in terms of concordance of language dominance and correlation between LI values across two testing runs.   ",2
42,"Nadkarni, Andreoli, Nair, Yin, Young, Kundu et al.",2015,"Usage of fMRI for pre-surgical planning in brain tumor and vascular lesion patients: Task and statistical threshold effects on language lateralization.",,25,"13 male, 12 female",1,1,"92 brain tumour patients. ",0,,0,1,0,0,0,0,0,0,0,0,"Demonstration of the threshold dependence of LIs obtained with the word generation task. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at four different threshold levels (t < 2.0, 2.67, 3.5, 4.0). ",0.2,"Traditional 0.2 cut-offs.",2,3,"ROIs were created based on MNI coordinates reported in Vigneau et al's (2006) meta-analysis; generated homologous ROIs in the right hemisphere for these left hemisphere language areas. ",Multiple,5,1,1,0,1,0,0,"Broca's area,     Wernicke's area,     Other associated motor and language areas.     (See Vigneau et al's, 2006 meta-analysis). ",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation","Visually letter cued covert word generation. ",1,2,1,1,Rest.,2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation",Rest,,,,,,,,,,"Left lateralisation found in the inferior, middle and superior frontal gyri across all threshold values. Varied from about 0.55 to 0.73.",,,"Laterality dominance was robust against changes in threshold value. LI values however were not robust, showing typical increases with increasing threshold level, from about 0.55 to 0.73. ",,"Threshold level chosen affected the mean LI value obtained, but not the dominance classification of individual subjects. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Threshold level did not affect dominance categorisation but resulted in increasing LI values with increasing threshold levels. ",2
43,"Niskanen, Kononen, Villberg, Nissi, Ranta-aho, Saisanen, et al.",2012,"The effect of fMRI task combinations on determining the hemispheric dominance of language functions.",,20,"9 male, 11 female",5,2,,0,,0,0,0,1,1,0,0,0,0,0,"Aimed to compare the laterality obtained using different language task combinations, to assess the combination of tasks that yielded optimal lateralisation measurement. ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"The number of voxels surviving a variable threshold were used for the standard LI equation. Threshold was defined adaptively for each subject as 80% of the maximum t-value within the ROI. ",0.1,"Left = LI > 0.1  Right = LI < -0.1  Bilateral = -0.1 <= LI <= 0.1",2,1,"Defined using the WFU PickAtlas.",3,3,0,0,0,0,0,0,"Broca's area,     Wernicke's area,    Combined ROI consisting of both Broca's and Wernicke's areas, as well as BA 46, Heschl's gyrus and the hippocampus.",1,,2,5,9,1,0,1,0,1,0,1,1,"Word generation (WG),     Word pair category task (WP),    Phoneme decision task (PD),    Responsive naming (RN),     Sentence comprehension (SC).","Word gen- covert generation to visually presented letter.    Word pair- covertly generate the category to which a pair of words both belong.     Phoneme decision- decide if the name of an object within a line drawing contains the phoneme /i/ or not.    Responsive naming- covertly generate the name of an object being described (written description).     Sentence comprehension- decide if a spoken sentence is semantically correct or not.",2,2,3,3,"WG- fixation and finger tapping task.   WP- tone decision task (which of a pair was higher).   PD- perceptual decision task (line drawings).  RN- view varying rows of hashes (#).  SC- tone decision task (did sequence contain 2 high pitch tones).   ",4,3,1,1,1,1,0,1,0,,2,"Letter cued word generation","Fixation and finger tapping",19,0,1,,0.95,0.88,0.94,0.71,,"Relatively strong laterality across LIs, weakest for Wernicke's area, strongest for Broca's area. Classification based on combined ROI LIs. ",,,,"Mean LI for Broca's area was higher (0.94) than for Wernicke's area (0.71) or for the combination of language areas (0.88). ",,,2,,,,,,,,,,,,,,,,,,,2,"Phoneme decision","Perceptual decision",18,2,0,,0.9,0.65,0.66,0.54,,"Moderate laterality observed in all ROIs (around 0.6). Classification based on combined ROI LIs. ",,"This task yielded the largest deviation in language dominance/variability in LI values between subjects. ",,"Stronger laterality observed in frontal versus temporal ROI (0.66 versus 0.54). ",,,2,,,,,,,,,,,,,,,,,,,2,"Read response naming","Hash pattern viewing",19,1,0,,0.95,0.94,0.96,0.93,,"Very strong laterality across all ROIs. Classification based on combined ROI LIs. ",,,,"Similarly high laterality found across ROIs (all > 0.9). Strongest in frontal ROI (0.96). ",,,2,,,,,,,,,,,,,,,,,,,2,"Auditory Sentence comprehension","Tone decision",18,2,0,,0.9,0.69,0.88,0.69,,"Moderately strong laterality seen across ROIs; particularly strong in Broca's area. Classification based on combined ROI LIs. ",,,,"Stronger laterality within Broca's area (0.88) than within Wernicke's area (0.69), or within the combined ROI (0.69). ",,,2,"Word pair task","Tone decision",18,2,,,0.9,0.78,0.74,0.76,,"Reasonably high LI strength across ROIs. Classification based on combined ROI LIs. ",,,,"Very similar LI values across frontal, temporal and combined ROIs. Only differ by 0.02. ",,,2,"CTA_1: All tasks  CTA_2: Visual tasks (WG, PD, RN)  CTA_3: Auditory tasks (SC, WP)  CTA_4: Optimal combination of auditory and visual tasks (WG, RN, SC). ",,"Multiple (see language paradigms). ",19,0,1,,0.95,0.87,0.99,0.85,,"LI values shown for combination of all tasks.   Visual tasks: Com = 0.94, W = 0.91, B = 0.95.  Auditory tasks : Com = 0.77, W = 0.76, B = 0.867.  Optimal: Com = 0.99, W = 0.92, B = 0.97.  Strongest LI values shown for optimal combination of visual and auditory tasks. ",,,"Variability between subjects and within subjects was decreased for the task combinations compared to the single task analyses. Variability was particularly decreased when both auditory and visual tasks were included in the combined task analysis. However, when the PD and WP tasks were included, this increased variability- these tasks had relatively poor activation power. ",,"Generally Wernicke's area mean LIs are lower than Broca's area LIs for all task combinations. LIs using combined language area ROI gives higher laterality for the visual task combination and the optimal visual and auditory combination. ",,,2,,,,,,"Auditory tasks produced much greater variability in LI values between subjects than visual tasks. Argued this may be due to the choice of baseline task; the tones used are acoustically simpler than words. ",,,"Overall, concluded that the optimal protocol for measuring language lateralisation with fMRI used a combined task analysis including word generation, read response naming and sentence comprehension. This combination includes both visual and auditory tasks and produced strong and consistent lateralisation. ",2
44,"Ocklenburg, Hugdahl, Westerhausen",2013,"Structural white matter asymmetries in relation to functional asymmetries during speech perception and production.",,29,"14 male, 15 female",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Compared laterality for a word generation and a passive listening task. Stronger laterality was found for word generation. The two tasks were significantly correlated within frontal but not temporal ROIs. ",2,1,3,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Compared both voxel count and voxel value LIs using the standard LI equation (using LI toolbox). The midline (+/-5mm) was excluded from the analysis and the threshold intensity level set to three. ","Not given","Procedure not specified, although numbers in each group given. ",2,1,"Used the LI-toolbox for SPM8. ",2,3,0,0,0,0,0,0,"Frontal lobe,     Temporal lobe.",1,,2,2,9,1,0,0,0,0,1,0,0,"Word generation,     Passive speech listening.","Silent auditory letter cued word generation,    Passive listening to words beginning with the same letter. ",1,2,3,1,"Rest (subjects told to focus on the sound of the fMRI scanner).",3,1,1,0,0,1,0,0,0,,2,"Letter cued word generation",Rest,23,3,3,,0.7931034482758621,,"0.49 - 0.51","0.29 - 0.3",,"LI strength depended greatly on region used, but was similar across voxel count and voxel value measures. Highest mean LI achieved within a frontal ROI using voxel values. Numbers within each dominance category shown for frontal ROI. ",,,,"Stronger laterality found within frontal ROI. For temporal ROI, only 15/29 subjects showed significant asymmetries (13 leftward, 2 rightward). ","Very similar LI strength found with both voxel count and voxel value measures:  Frontal ROI: voxel count = 0.49, voxel value = 0.51.  Temporal ROI: voxel count = 0.29, voxel value = 0.3.",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Passive speech listening",Rest,17,7,5,,0.5862068965517241,,"0.19 - 0.2","-0.03 - -0.06",,"LI strength was generally weak and region dependent. Stronger lateralisation found with a frontal ROI (about 0.2). When temporal mean LI values were considered for each dominance group separately, the mean for left dominant subjects was 0.15/0.15, and for right dominant subjects was -0.2/-0.22. ",,,,"Frontal LIs were stronger than temporal LIs. Direction of asymmetry within temporal ROI was much more mixed than within frontal ROI; in the temporal ROI, of those 25 subjects showing asymmetry, 12 were leftward and 13 were rightward. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"For frontal LIs, both voxel count LIs and voxel value LIs were significantly correlated between the tasks (r = 0.4). No correlation seen between tasks for temporal LIs. May be attributable to the fact that only half of the subjects showed asymmetries for word generation, and/or the increased incidence of right sided asymmetry in the passive listening task. ",,,,,,,,"Strength of laterality depended largely on task and region. Strongest laterality found for word generation within frontal ROI. Reduced LI values were seen for passive speech listening compared to word generation and for temporal compared to frontal ROIs. Particularly high level of rightward asymmetry seen for passive speech listening within the temporal lobe. Correlation between the two tasks was seen for frontal but not for temporal LIs.  ",2
45,"Orellana, Visch-Brink, Vernooji, Kalloe, Satoer, Vincent, et al.",2015,"Crossed cerebrocerebellar language lateralization: An additional diagnostic feature for assessing atypical language representation in presurgical functional MR imaging. ",,20,"9 male, 11 female",4,1,"19 left handed patients who had undergone preoperative fMRI. ",0,,0,0,0,1,1,0,0,0,0,1,"Main focus of paper was on comparing lateralisation within the cerebral hemispheres versus the cerebellum. Found that crossed cerebrocerebellar lateralization was found in all subjects regardless of whether language was typically or atypically lateralised. ",2,1,1,3,0,0,0,0,0,0,,1,5,0,0,0,0,0,,"The number of voxels in each ROI was determined using a threshold-independent method (Branco's t-weighting method), and then used for the standard LI equation. ",0.1,"Left = LI > 0.1  Right = LI < -0.1  Bilateral = -0.1 <= LI <= +0.1 ",2,1,"Method not specified. ",2,5,1,1,0,0,0,1,"Two ROIs:  One cerebral ROI consisting of the IFG, superior and middle temporal gyri, and angular and supramarginal gyri.   One cerebellar ROI consisting of the cerebellum. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Verb generation task.","Covertly generate a verb to an aurally presented noun. ",1,2,1,1,"Passive listening to high and low tones. ",2,1,0,1,1,1,0,0,0,"Used a tone listening task as a baseline for auditory verb generation. ",2,"Verb generation","Passive tone listening",13,5,2,,0.65,-0.16,,,0.14,"Mean group LI strength appears relatively weak; however there was a large overrepresentation of left handers, and so a greater number of right lateralised subjects. When analysing left dominant subjects only, mean group LI is 0.295. Majority of subjects were right lateralised within the cerebellum (12/20), 6/20 were bilateral, 2/20 were left lateralised. ",,,,"Found a negative correlation between cerebral and cerebellar LIs; found a crossed cerebrocerebellar activation in 12/20 (60%) of subjects. None of the participants showed the same language dominance in both the cerebrum and cerebellum. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,"Found similar inverse relationship between lateralisation in cerebral versus cerebellar ROIs as in healthy subjects. ",,"Main finding was a crossed lateralization across cerebral and cerebellar LIs. ",,,,"Demonstrated crossed lateralisation for cerebral versus cerebellar LIs, irrespective of whether subjects were typically or atypically lateralised for language. ",2
46,"Partovi, Jacobi, Rapps, Zipp, Karimi, Rengier et al.",2012b,"Clinical standardized fMRI reveals altered language lateralization in patients with brain tumor. ",,14,"7 male, 7 female",1,1,"57 right handed brain tumour patients. ",0,,0,0,0,1,1,0,0,0,0,0,"Demonstration of laterality obtained for a word generation and sentence generation task within both frontal and temporal ROIs. ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used the number of active voxels for the standard Li equation. Threshold set (variably?) at the level of the peak activation of the nondominant hemisphere. ","Not given","Not specified. ",2,1,"Defined anatomically within BrainVoyager pre-processing software. ",2,3,0,0,0,0,0,0,"Broca's area,     Wernicke's area (including superior temporal sulcus, middle temporal gyrus). ",1,,2,2,9,1,0,0,0,0,0,0,1,"Semantic fluency,     Sentence generation.","Semantic fluency- silently generate words belonging to a visually presented category.     Sentence generation- silently repeat a standardized sentence to describe a picture that had been practised before the scan. ",1,2,1,1,"Not specified- assume rest used as baseline",2,2,0,1,0,0,0,0,0,,2,"Semantic fluency",Rest,,,,,,,0.825,0.578,,"Relatively strong laterality seen in both ROIs, but strongest in Broca's area. ",,,,"Weaker laterality found within Wernicke's area (0.58) than Broca's area (0.83). ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence generation",Rest,,,,,,,0.644,0.653,,"Relatively high laterality found in both ROIs; strength of laterality was practically the same for both Broca's area and Wernicke's area. ",,,,"Effectively same extent of lateralisation found in both frontal and temporal ROIs (0.64 versus 0.65). ",,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Strong laterality found for word generation and sentence generation tasks within frontal and temporal ROIs. Stronger laterality found in frontal than temporal ROI for word generation, but almost identical for sentence generation across ROIs. ",2
47,"Partovi, Konrad, Karimi, Rengier, Lyo, Zipp et al.",2012a,"Effects of covert and overt paradigms in clinical language fMRI. ",,14,"7 male, 7 female",1,2,,0,,0,0,0,0,1,1,0,0,0,0,"Compared covert and overt paradigms for lateralisation. Found that covert paradigms were more strongly lateralising than overt paradigms. ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used voxel counts for the standard LI equation. A variable threshold was used, defined as the threshold of the peak activation of the nondominant hemisphere. ","Not given","Not specified.",2,1,"Defined anatomically (not otherwise detailed). ",2,3,0,0,0,0,0,0,"Broca's area (IFG),     Wernicke's area (STG, supramarginal and angular gyri).",3,"Two sessions on the same day, one on another day (interval not specified).",2,4,9,1,0,0,0,0,0,0,1,"Covert semantic fluency,    Overt semantic fluency,     Covert sentence generation,     Overt sentence generation.","Semantic fluency- generate words belonging to visually presented category.     Sentence generation- repeat standardized previously defined sentence corresponding to the picture presented. ",1,3,1,1,"Cross fixation.",2,2,0,1,0,0,1,0,0,,2,"Semantic fluency","Cross fixation",,,,,,,0.83,0.58,,"LIs were significantly stronger for covert word generation than for overt word generation, across both ROIs. LI values shown above are for covert WG.   Covert frontal = 0.83  Overt frontal = 0.66  Covert temporal = 0.58  Overt temporal = 0.06","Reproducibility only assessed for covert paradigms. Found no significant differences between LIs across the three sessions for either Broca's area or Wernicke's area. Suggested that LIs for Wernicke's area showed slightly more variability across testing sessions than Broca's area. ","SDs suggest slightly greater between subject variability for LIs within Wernicke's area compared to Broca's area.  Variability may be slightly higher for overt measures in Wernicke's area.   BA SD (covert) = 0.37  BA SD (overt) = 0.37  WA SD (covert) = 0.55  WA SD (overt) = 0.65",,"LIs were stronger within frontal than within temporal ROIs, for both overt and covert paradigms. Some evidence that variability in LI values was also higher in temporal ROI compared to frontal ROI. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence generation",Rest,,,,,,,0.65,0.65,,"LIs were significantly stronger for covert word generation than for overt word generation, in Wernicke's area only (difference was not statistically significant in Broca's area). LI values shown above are for covert SG.   Covert frontal = 0.65  Overt frontal = 0.24  Covert temporal = 0.65  Overt temporal = -0.06","Reproducibility only assessed for covert paradigms. Found no significant differences between LIs across the three sessions for either Broca's area or Wernicke's area. Suggested that LIs for Wernicke's area showed slightly more variability across testing sessions than Broca's area. ","SDs suggest slightly greater between subject variability for covert LIs within Wernicke's area compared to Broca's area.  Variability appears higher for overt measures in Broca's area.   BA SD (covert) = 0.44  BA SD (overt) = 0.81  WA SD (covert) = 0.72  WA SD (overt) = 0.76 ",,"LI strength was highly comparable between frontal and temporal ROIs for both overt and covert paradigms (identical mean LI for covert). Some evidence that variability in LI values was higher in temporal ROI compared to frontal ROI. ",,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Covert paradigms yield stronger laterality than overt paradigms. Covert sentence and word generation tasks yield strong and highly reproducible LIs. Covert sentence generation yielded identical mean LI values within frontal and temporal ROIs, whereas covert word generation yielded stronger laterality in frontal than temporal ROIs. ",2
48,"Perlaki, Horvath, Orsi, Aradi, Auer, Varga et al.",2013,"White-matter microstructure and language lateralization in left-handers: A whole-brain MRI analysis. ",,16,"0 male, 16 female",2,2,,0,,0,0,0,0,0,0,0,0,0,1,"Main focus was on structure-function relationships in brain asymmetry. Demonstration of the lateralising ability of word generation in the frontal lobe in a sample of left handers. ",2,1,,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used the bootstrapping method in LI toolbox to calculate weighted LIs averaged over multiple thresholds. Not specified whether voxel count or voxel value was used. ",0.2,"Traditional 0.2 cut-offs used. ",2,1,"ROI was generated in standard space by combining the frontal lobe defined by the 25% probability thresholded MNI structural Atlas with its mirror image. ",1,1,0,0,0,0,0,0,"Frontal lobe. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation","Aurally letter cued silent word generation. ",1,2,1,1,Rest.,2,1,1,0,0,1,0,0,0,,2,"Letter cued word generation",Rest,10,5,1,,0.625,,0.24,,,"Low mean LI value reflects exclusive left handed sample. Majority of subjects still left lateralised for language, but a significant proportion were strongly right lateralised. ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Demonstration of both strong left and right lateralisation in a sample of left handers using the word generation task and a frontal ROI. 5/16 subjects were right dominant, 10/16 left dominant, 1/16 bilateral. ",2
49,"Pravata, Sestieri, Mantini, Briganti, Colicchio, Marra et al.",2011,"Functional connectivity MR imaging of the language network in patients with drug-resistant epilepsy. ",,12,"Not specified",1,1,"11 epileptic patients.",0,,0,0,0,0,1,0,0,0,0,1,"Main focus of paper was on resting connectivity in the language network of epilepsy patients. Demonstration of the lateralising ability of the verb generation task across anterior and posterior language areas. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts for the standard LI equation, computed across multiple thresholds between P < .164 and the least stringent P value. Computed two LIs, one global, one regional. ",0.2,"Traditional 0.2 cut-offs.",3,2,"Four regional ROIs (anterior and posterior) were obtained by grouping the task-evoked BOLD activation of controls (and patients).  ",2,6,1,1,0,0,0,1,"Two ROIs:    Global LI,    Regional LI- made up of anterior and posterior language areas. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Verb generation.","Silent generation of a verb to go with a visually presented noun. ",1,2,1,1,"Fixation/rest. ",2,2,0,1,1,0,0,1,0,,2,"Verb generation","Cross fixation",11,0,1,0,0.9166666666666666,,0.7,,0.43,"Moderate to strong laterality found, depending on whether a regional or global LI was used. Strongest found for regional LI (frontal and temporal areas). ",,"Similar variability for both global and regional LI:  Global SD = 0.2.  Regional SD = 0.24.",,"Stronger LI found for regional approach (ROI of both frontal and temporal regions) than a global LI using whole hemispheres. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Stronger lateralisation found for verb generation task when using a regional rather than a global LI. ",2
50,"Propper, O'Donnell, Whalen, Tie, Norton, Suarez et al.",2010,"A combined fMRI and DTI examination of functional language lateralization and arcuate fasciculus structure: Effects of degree versus direction of hand preference.",,25,"9 male, 16 female",4,2,,0,,0,0,0,1,0,0,0,0,0,1,"Main focus of paper was on relating functional and structural asymmetries to handedness. Demonstration of the lateralising power of an antonym generation task within Broca's and Wernicke's areas. ",2,1,1,3,0,0,0,0,0,0,,1,5,0,0,0,0,0,,"Used a threshold-independent histogram/t-weighting method (Branco et al, 2006). Involved creating histograms of number of active voxels across t thresholds, multiplying by a linear weighting function, and then integrating the area under the curve for the right and left hemispheres separately. These values were then compared in a standard LI equation format. ","Not given","Not specified/used. ",2,1,"Generated atlas-based structural masks using WFU PickAtlas software, for Broca's and Wernicke's areas. ",2,3,0,0,0,0,0,0,"Broca's area- BA 44 and 45.  Wernicke's area- STG, MTG and supramarginal and angular gyri. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Antonym generation","Silent generation of antonym for visually presented word. ",1,2,1,1,Fixation/rest.,2,2,0,1,0,0,0,1,0,,2,"Antonym generation","Cross fixation",21,4,,,0.84,,0.47,0.26,,"LI strength depended on region and handedness.   Consistent right handed: BA LI = 0.42, WA LI = 0.36.  Consistent left handed: BA LI = 0.52, WA LI = 0.29.  Inconsistently handed: BA LI = 0.47, WA LI = 0.17.  4 subjects were classed as right lateralised (2 right lateralised in Broca's area, 2 right lateralised in both Broca's and Wernicke's areas, 0 right lateralised in Wernicke's area only). ",,"Comparable variability across ROIs:  Broca's area SD = 0.7.  Wernicke's area SD = 0.6.",,"Stronger mean group LI within frontal than temporal ROI. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,"Two subjects were right lateralised in Broca's area who were not classed as right lateralised in Wernicke's area (but not specified whether they were bilateral or left lateralised in Wernicke's area). No subjects were right lateralised only in Wernicke's area. ",,"Effect of handedness on LI depended on the ROI. For BA, right handed subjects showed a lower mean LI than left handers. For WA, the opposite was true.   Consistent right handed: BA LI = 0.42, WA LI = 0.36.  Consistent left handed: BA LI = 0.52, WA LI = 0.29.  Inconsistently handed: BA LI = 0.47, WA LI = 0.17.  ",,"Demonstrated moderate laterality for an antonym generation task using a histogram method of LI calculation. Greater laterality seen for frontal (0.47) than for temporal (0.26) ROI. LI depended on handedness (but inconsistent pattern across ROIs). ",2
51,"Ramsey, Sommer, Rutten, Kahn.",2001,"Combined analysis of language tasks in fMRI improves assessment of hemispheric dominance for language functions in individual subjects. ",,16,"16 male, 0 female",4,2,,0,,0,0,0,1,1,0,0,0,0,0,"Compared combined task analysis to single task analysis in terms of the strength of lateralisation obtained. Also combined laterality across frontal and temporal language regions. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at two threshold levels (t = 3.0 and 4.5). ","Not given","Not specified.",2,1,"Defined anatomically and manually using Brodmann's mapping system; blind to the activity maps. ",3,3,0,0,0,0,0,0,"Frontal ROI- BA 44 and 45.   Temporal ROI- MTG, STG, angular and supramarginal gyri.   Frontal and temporal ROI- above areas combined. ",1,,2,4,9,1,1,0,0,0,0,0,1,"Verb generation,     Antonym generation,    Semantic decision,    Reversed spelling semantic decision.","VG- covert generation of verb to visually presented noun.  AG- covert generation of antonym to visually presented word.   SD- decide if noun is an animal.   RSD- same as SD, but words were presented with reversed spelling (requires extra phonological processing before semantic decision can be made). ",2,2,3,3,"Generation tasks- passive cross fixation.   Decision tasks- button presses cued by dots on screen.",4,2,1,1,0,0,0,1,0,,2,"Verb generation, antonym generation.","Cross fixation",,,,,,"*0.59, 0.65",,,,"LI strength depended on task and threshold level. LI values given for t = 4.5 (estimated from graph). At the higher threshold, AG gave a stronger LI. At the lower threshold, VG gave the stronger LI.   VG 3.0 = 0.72.  VG 4.5 = 0.59.   AG 3.0 = 0.57.   AG 4.5 = 0.65.  ","Compared LI values across two testing sessions for the verb generation task. LIs were less reproducible with a low versus a high threshold. ",,"LI not robust against changes in threshold level. ","No significant effect of frontal versus temporal ROI, so these were combined into a single ROI for all analyses. ","LIs were not significantly correlated across threshold levels. Across tasks, LIs values at the 3.0 threshold tended to be lower than those at the 4.0 threshold (but difference not significant). For verbal fluency tasks, LIs were weaker for 4.5 then 3.0 for verb generation; the reverse was true of antonym generation. Thus, which task produced the highest laterality depended on the threshold used. High threshold produced more reproducible LIs for VG than low threshold. ",,2,"Semantic decision","Button presses.",,,,,,*0.47,,,,"LI strength depended on threshold level; lower for low threshold. Note that LI values are estimated from graph.   3.0 LI = 0.3  4.5 LI = 0.47",,"LIs became more variable at a low threshold (3.0) compared to a high threshold (4.5).","LI not robust against changes in threshold level, in terms of both strength and variability. ","No significant effect of frontal versus temporal ROI, so these were combined into a single ROI for all analyses.","Across tasks, LIs values at the 3.0 threshold tended to be lower than those at the 4.0 threshold (but difference not significant). For SD, stronger LI found for higher threshold. LIs were not significantly correlated across threshold levels. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Reversed spelling semantic decision","Button presses",,,,,,*0.51,,,,"LI strength depended on threshold level; lower for low threshold. Note that LI values are estimated from graph. LI was significantly lower than VG LI.   3.0 LI = 0.21  4.5 LI = 0.51",,,"LI not robust against changes in threshold level.","No significant effect of frontal versus temporal ROI, so these were combined into a single ROI for all analyses.","Across tasks, LIs values at the 3.0 threshold tended to be lower than those at the 4.0 threshold (but difference not significant). For reversed SD, stronger LI found for higher threshold. ",,2,"Two CTA task combinations:  CTA1: Verb generation, antonym generation, semantic decision.   CTA2: Verb generation, reversed semantic decision. ","Used conjoint analysis of scan data across different language tasks; analysed the whole set of data for the individual tasks combined as if they were a single task. ","Cross fixation, button pressing. ",,,,,,0.88,,,,"CTA1 yielded the strongest laterality at the high threshold compared to single tasks. Not true at low threshold (verb generation was strongest, CTA second strongest). Thus, LI strength was threshold dependent.   CTA1 3.0 = 0.58 (*estimated from graph)  CTA1 4.5 = 0.88  Similar results for CTA2- produced stronger and more consistent laterality. ","All tasks were significantly correlated with CTA1; strongest correlation seen with antonym generation (0.9), then verb generation (0.88), then SD (0.62). ",,"CTA LI had the smallest SD out of all task analyses; so LIs were more consistent across subjects for CTA (both 1 and 2).  ","CTA1 LI was highly correlated across thresholds (r = 0.93); only task analysis to show a significant correlation across thresholds. Thus, CTA was more robust against changes in threshold level. ",,"Higher LI found at high compared to low threshold; however, LIs were highly correlated across threshold levels (unique to CTA). ",,2,"Calculated correlations between tasks using the low threshold data. Found that only VG and AG showed a significant correlation (0.89). SD was only weakly correlated with VG and AG (0.43/0.46). Reversed SD and VG were not correlated. ",,,,,,,,"LI was task and threshold dependent, but not ROI dependent. Overall, CTA provided the best lateralisation in terms of strength, variability/consistency and robustness against changes in threshold level. Generation tasks yielded stronger laterality than decision tasks. VG LIs were reproducible at a high but not a low threshold. ",2
53,"Razafimandimby, Maiza, Herve, Lecardeur, Delamillieure, Brazo et al.",2007,"Stability of functional language lateralization over time in schizophrenia patients. ",,10,"8 male, 2 female",1,1,"10 schizophrenic patients. ",0,,0,0,0,0,0,1,0,0,0,0,"Provides measure of reliability of LIs obtained using a story listening task over two sessions. ",2,1,1,5,0,0,0,0,0,0,,4,1,0,0,0,0,0,,"Calculated using right minus left fractional signal variation values. LI range unknown. ","Positive versus negative","Not precisely specified, but seem to be classed on the basis of which side of 0 their LIs were across the two testing sessions. ",2,1,"Anatomical method of definition not specified. ",1,3,0,0,0,0,0,0,"One combined ROI consisting of:  IFG pars triangularis,  MTG,   Angular gyrus.",2,"26 months on average",2,1,6,0,0,0,0,0,0,0,0,"Story listening.","Passive listening to a factual story (in native French). Same story used for both testing sessions.",1,2,2,1,"Passive listening to a factual story in a non-native unknown language (Tamil). ",3,1,1,1,0,1,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Story listening","Unknown language story listening",8,,2,,0.8,,,,,"Precise LI values not given, but reported that 7/10 subjects showed strong left lateralisation across testing sessions. Two showed bilateral LIs across testing sessions. One subject switched from strongly left lateralised to bilateral across testing sessions (attributed to better audio delivery/comprehension at second session).  ","LIs were strongly correlated across testing sessions. When both controls and patients were combined, r = 0.81. Slope of the linear regression was close to 1 and the intercept close to 0, indicating no change in lateralisation between the first and second session. ",,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Reported reproducible leftward lateralisation for passive story listening within a combined frontal and temporal ROI. ",2
54,"Ruff, Brennan, Peck, Hou, Tabar, Brennan, et al.",2008,"Assessment of the language laterality index in patients with brain tumor using functional MR imgaging: Effects of threshold, task selection, and prior surgery. ",,7,"2 male, 5 female",1,1,"13 brain tumour patients. ",0,,0,0,0,0,1,0,0,0,0,0,"Comparison of laterality obtained with phonemic fluency, verb generation and semantic fluency tasks. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts for the standard LI equation. Computed LIs at 15 different statistical thresholds. Calculated the median LI value obtained at all thresholds between P < .008 and the least stringent P value. ",0.2,"Traditional 0.2 cut-offs.",2,1,"Anatomically defined using AFNI software. ",1,1,0,0,0,0,0,0,"IFG ",1,,2,3,9,1,0,0,0,0,0,0,0,"Verb generation,     Phonemic fluency,     Semantic fluency.","VG- silently generate verb to go with visually presented noun.     PF- silently generate words beginning with visually presented letter.     SF- silently generate words belonging to visually presented category. ",1,2,1,1,"Cross fixation.",2,2,1,1,0,0,0,1,0,,2,"Word generation, Verb generation, Semantic fluency.","Cross fixation",,,,,,,0.84,,,"Strongest LI demonstration for verb generation task (MLI = 0.84). ",,"Reduced variability in LIs (as measured by median absolute deviation) seen for verb generation (MAD = 0.06) as compared to word generation and semantic fluency. Greatest variability seen for semantic fluency task (MAD = 0.19). ","No subjects showed a switch in hemispheric dominance across changes in threshold level for the verb generation task. Switches in dominance with changes in threshold were seen however for the semantic fluency task. ",,"Demonstrated the classic threshold dependence of LIs across 15 threshold levels. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"A verb generation task was found to be optimal for lateralisation in terms of strength and consistency of lateralization across threshold levels. Greater variability across thresholds and weaker laterality was found for semantic fluency (in some cases, dominance classification was threshold dependent). ",2
55,"Rutten, Ramsey, van Rijen, van Veelen",2002,"Reproducibility of fMRI-determined language lateralization in individual subjects. ",,9,"6 male, 3 female",4,2,,0,,0,1,0,1,1,1,0,0,0,0,"Compared reproducibility and strength of lateralisation across different tasks, CTA versus STA, and regional versus global LIs. ",2,1,1,1,0,0,0,0,0,0,,2,3,0,0,0,0,0,,"Used voxel counts within the standard LI equation at 6 different threshold levels (from T = 2.0 to 4.5, increment 0.5). ","Not given","Not specified/used. ",3,1,"Anatomically defined using standard Brodmann anatomical criteria, blind to brain activity maps. ",2,6,1,1,0,0,0,1,"Language ROI: IFG, IFS, posterior part of STG and MTG, supramarginal gyrus, and angular gyrus.     Global ROI: Whole hemispheres ",2,"Mean interval of 5 months between sessions.",2,3,9,1,0,0,0,1,0,0,0,"Verb generation,     Antonym generation,     Picture naming.","VG- covert generation of verb to visually presented noun.     AG- covert generation of antonym to visually presented word.     PN- covertly name line drawing of object. ",2,2,1,1,"Cross/blank screen fixation.",2,2,0,1,1,0,0,1,0,,2,"Verb generation, antonym generation.","Cross fixation",,,,,,,,,,"LI strength depended on ROI- LIs were significantly higher when using a regional LI as compared to a global (whole hemispheres) LI, across all thresholds. Higher LIs were found for higher thresholds. No significant difference between LIs for VG and AG at any threshold level for regional LIs. ","Reproducibility of LI was dependent on task, ROI and threshold. For regional LIs, VG showed significant between session correlations across thresholds; correlations for AG were weak and only significant at t = 2.0 and 4.0. For global LIs, VG was again significantly correlated across sessions; AG showed improved correlations (significant across ts 2.0-4.0).  The mean absolute difference between sessions was smaller for VG (18) than for AG (22). ",,"Neither VG nor AG LIs were robust against changes in threshold. AG reached a plateau around T = 3.5, whereas VG carried on increasing up to the top threshold of 4.5. ","Weaker between session correlations for global versus regional LIs.","LI increased with increasing t threshold level. For VG, increases were seen across the full range of threshold levels, whereas for AG a plateau was reached at about T = 3.5. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Picture naming",Rest,,,,,,,,,,"LI strength depended on ROI and threshold. Significantly higher LIs were found for regional versus global ROIs. LIs increased with increasing threshold. ","For both regional and global LIs, PN was not reproducible; LIs were not significantly correlated between the sessions at any threshold level. PN showed the biggest mean absolute difference between sessions out of all tasks used (31). ",,"LI not robust against changes in threshold level; but typically reached a plateau around T = 3.5. ","No effect of ROI on reproducibility of LIs; correlations between sessions were not significant for either regional or global LIs. ROI had an effect on strength of LI; regional LIs were significantly stronger than global LIs. ","LI increased with increasing t threshold level, reaching a plateau typically at t = 3.5. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Verb generation,     Antonym generation,     Picture naming. ","Scanner data from all three tasks were appended and conjointly analysed. Data from all analyses were incorporated in the multiple regression matrix and statistical maps were obtained from the appended time series of the three individual tasks.","Passive cross fixation/rest.",,,,,,,,,,"LI strength depended on ROI. Significantly higher LIs were found for regional versus global ROIs. LI strength increased with increasing threshold. ","Correlations not calculated, but found no significant differences in strength of regional LI between any tasks. ","Both regional and global LIs were significantly correlated between sessions for CTA across all threshold levels. Reproducibility of LI for CTA appeared more stable over threshold levels than the single tasks. CTA LI had the smallest mean absolute difference between sessions out of all tasks (14). ",,"CTA LI reproducibility appeared possibly more robust against changes in threshold level compared to other tasks. LI strength was not robust against changes in threshold level; increased with increasing threshold. ","ROI had an effect on strength of LI; regional LIs were significantly stronger than global LIs. ROI however did not have an effect on reproducibility of LI; correlations between sessions were significant across all thresholds for both global and regional LIs. ","LIs increased as a function of threshold.",,2,"Correlations not calculated, but found no significant differences in strength of regional LI between any tasks. ",,,,,,,,"Reproducibility of LIs depended on task and region, but was relatively independent of threshold. CTA produced the most robust and reliable lateralisation. Significantly stronger laterality was found for regional versus global LIs. Verb generation was reliable, whereas picture naming and antonym generation were not. However, strength of laterality did not differ significantly between tasks. ",2
56,"Sanjuan, Bustamante, Forn, Ventura-Campos, Barros-Loscertales, Martinez et al",2010,"Comparison of two fMRI tasks for the evaluation of expressive language function.",,18,"9 male, 9 female",1,1,"58 patients with epilepsy/brain lesions. ",0,,0,0,0,0,1,0,0,0,0,0,"Compared laterality for a verb generation and a verbal fluency task, using a frontal ROI.",2,1,1,1,0,0,0,0,0,0,,2,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at a single fixed threshold. ",20,"Traditional +/-20 cut-offs.",2,1,"Defined using anatomical criteria (Brodmann areas). ",1,1,0,0,0,0,0,0,"Frontal ROI consisting of the inferior frontal gyrus and inferior frontal sulcus.",1,,2,2,9,1,0,0,0,0,0,0,0,"Verb generation,     Word generation.","VG- covert generation of verbs in response to aurally presented nouns.     WG- covert generation of words beginning with aurally presented letter. ",1,2,1,2,"VG- silent repetition of letters.     WG- silent repetition of word 'casa' (Spanish for house).",2,1,1,1,1,1,0,0,0,,2,"Verb generation, word generation",,,,,,,,"82.1, 65.8.",,,"Mean LI was stronger for the verb generation task (82.1) than the word generation task (65.8). 41% of the participants had a VG LI that was over 30 units higher than the WG LI.",,"LIs were slightly more variable for word generation (+/- 33.18) than for verb generation (+/- 28.21). Not know if significantly different. ",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"LIs for the word generation and verb generation tasks were significantly correlated (r = .58). ",,,,,,,,"Stronger lateralisation was found for a verb generation task over a word generation task within a frontal ROI; however LIs for the tasks were significantly correlated. ",2
57,"Sanjuan, Forn, Ventura-Campos, Rodriguez-Pujadas, Garcia-Porcar, Belloch et al.",2010,"The sentence verification task: A reliable fMRI protocol for mapping receptive language in individual subjects",,22,"12 male, 10 female",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Compared lateralisation within frontal and temporoparietal areas for a sentence verification task, as a procedure for measuring lateralisation of receptive language function. ",2,1,1,1,0,0,0,0,0,0,,2,3,0,0,0,0,0,,"Used voxel counts for the standard LI equation. Used a two-threshold (TT) correlation analysis to determine those voxels that were significantly activated within ROIs (see Auer and Frahm, 2009). ",20,"Left = LI > 20  Right = LI < -20  BL = -20>= LI <= +20",2,1,"Anatomically defined using the Toolbox WPU PickAtlas using standard anatomical criteria according to Rutten et a (2002). ",4,3,0,0,0,0,0,0,"Frontal ROI: inferior frontal gyrus and sulcus.     Temporal ROI: posterior STG and MTG.     Parietal ROI: angular and supramarginal gyri.     Temporoparietal ROI: sum of the temporal and parietal ROIs. ",1,,2,1,7,0,0,0,0,0,0,0,0,"Sentence verification task.","Decide if an auditorily presented sentence is true or false, and report their answer by either raising their right (true) or left (false) hand. ",1,2,2,2,"Phoneme decision task- decide if there is a letter 'A' within a pair of letters presented auditorily, by raising their right (true) or left (false) hand.",3,1,0,1,1,1,0,0,0,"Involves syntactic, semantic and lexical processes. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence verification","Phoneme decision",22,0,0,0,1,79.32,54.84,69.27,,"Strongest mean LI was found within the parietal ROI (79.32).   Frontal: 54.84 (17 left dominant, 5 bilateral).  Temporal: 61.07 (20 left dominant, 2 bilateral).  Parietal: 79.32 (20 left dominant, 1 bilateral, 1 right dominant).  Temporoparietal: 69.27 (22 left dominant). ",,"Greatest variability in LI shown for frontal ROI (SD = 35.44), then parietal ROI (SD = 30.69), then temporal ROI (SD = 26.56). Least variability shown in combined temporoparietal ROI (SD = 21.57). ",,"Stronger laterality was found within temporoparietal than frontal ROIs. Strongest within parietal ROI (79.32), than combined temporoparietal ROI (69.27) than temporal ROI (61.07) then frontal (54.84).",,"Note that a relatively high level language baseline was used (phoneme decision). This yielded strong laterality for the sentence verification task. ",2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,"LIs obtained with the temporal and the parietal ROIs were only just not significantly correlated (r = 0.35, p = 0.055). ",,,,,"A sentence verification task using an active phoneme decision baseline yielded strong laterality in temporoparietal areas (weaker in frontal areas). All subjects were left dominant within temporoparietal ROI (right handed sample). This is thus a suitable task for assessing lateralisation of receptive language function. ",2
58,"Seghier, Lazeyras, Pegna, Annoni, Zimine, Mayer, et al. ",2004,"Variability of fMRI activation during a phonological and semantic language task in healthy subjects.",,26,"20 male, 10 female (in original sample of 30)",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Compared laterality for a phonological and a semantic task within frontal and temporal language areas. ",2,1,1,1,0,0,0,0,0,0,,1,3,0,0,0,0,0,,"Used voxel counts for the standard LI equation. Main LIs used a single fixed threshold of P < 0.005, but also calculated LIs across a range of threshold levels to plot TDLCs. Calculated using voxels across whole hemispheres. Also calculated a frontal dominance index, to compare laterality in frontal vs posterior ROIs: FDI = (left frontal voxels - left posterior voxels) / (left frontal voxels + left posterior voxels). ","mean - 2 standard deviations","Threshold for left and atypical language lateralisation defined adaptively as the group mean minus two standard deviations. ",3,1,"LIs were calculated using whole hemispheres. FDI used Talairach coordinates to delimite anterior and posterior cubic boxes ",3,6,1,1,0,0,0,1,"Global LI- whole hemispheres.     FDI- frontal and posterior ROIs in the left hemisphere. ",1,,2,2,9,0,1,1,0,0,0,0,0,"Rhyming decision,     Semantic categorization. ","RD- decide if two visually presented words rhymed or not, in a go-nogo task paradigm (only respond with right hand if words rhymed).     SD- decide if two visually presented words were from the same semantic category or not, in a go-nogo task (only respond with right hand if words in the same semantic category). ",1,2,2,2,"Perceptual decision task- decide if pairs of Greek letter strings were visually identical or different, in a go/nogo task (only respond if strings are visually identical). ",4,2,1,1,0,0,0,1,0,"Both tasks require translation from orthography to phonology, but the semantic categorization task additionally requires activation of word representations. ",2,,,,,,,,,,,,,,,,,,,2,"Sematic categorisation","Perceptual decision",,,,,,,,,0.76,"Semantic categorisation yielded a stronger mean LI (0.76) than the rhyming decision task. ",,"SD = 0.19 (slightly lower than that of rhyming decision task). Reported lower variability in the semantic categorisation task than in rhyming decision. ","LIs were not robust against changes in threshold level (as shown in threshold dependent laterality curves). ","Mean FDI for semantic categorisation was 0.29, significantly higher than that of the rhyme task. Thus, laterality in the SC task involves a greater contribution from frontal versus posterior asymmetries compared to the rhyming task. ","Increasing LI seen with increases in threshold level used. ",,2,"Rhyming decision","Perceptual decision",,,,,,,,,0.66,"Relatively strong lateralisation seen for rhyming decision (0.66); but this was weaker than that seen for semantic categorization. ",,"Higher variability seen in the rhyming decision task (SD = 0.27) than in the semantic categorisation task (SD = 0.19). ","LIs were not robust against changes in threshold level (as shown in threshold dependent laterality curves). ","Mean FDI for rhyming decision was 0.09, significantly lower than that of the semantic categorization task. Thus, laterality in the SC task involves a greater contribution from frontal versus posterior asymmetries compared to the rhyming task. Again, the rhyming task showed higher variability in FDI than semantic categorization. ","Increasing LI seen with increases in threshold level used. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Reported stronger and less variable laterality indices for a semantic categorisation versus a rhyming decision task. Frontal dominance indices suggested that this might be explained by a stronger frontal activation in the SC versus the RD task i.e. frontal asymmetries may be stronger for the SC versus the RD task. ",2
59,"Seghier, Kherif, Josse, Price",2011,"Regional and hemispheric determinants of language laterality: Implications for preoperative fMRI.",,82,"39 male, 43 female",5,2,,0,,0,0,0,1,0,0,0,0,0,0,"Argue against the use of a global LI in light of regional dissociations in laterality i.e. spatial heterogeneity in language lateralisation. ",2,1,1,6,0,1,0,0,0,1,"Flip method,     Threshold independent method (Nagata et al, 2001). ",1,6,1,0,0,0,1,"Unthresholded method,       Flip method (single threshold?)","Calculated LIs using a threshold independent approach (Nagata et al, 2001). Plots the number of L and R voxels activated across a range of threshold levels. Then calculates a non-linear regression of the curve to provide a constant term that can be used within the standard LI equation.     Also calculated voxel based laterality maps using the flip method to assess regional lateralisation. These maps code the interaction between task and hemisphere at every voxel. ",Positive/negative.,"Classification based on whether LI was positive or negative. ",3,3,"Global ROI- whole hemispheres excluding the cerebellum and the mesial part of the brain (around the inter-hemispheric fissure).     Regional ROI- reduced voxel based laterality map to 50 compact clusters, using voxels showing greater left than right activity for the task. ",50,6,1,1,0,1,1,1,"Global LI- whole hemispheres.     Regional LI- 50 clusters across the whole brain. ",1,,2,1,2,0,0,0,0,0,0,0,0,"Semantic word matching task. ","Categorize three visually presented words by matching one of two words to a target word. ",1,2,2,2,"Perceptual matching task- match one of two strings of unfamiliar greek letters to target.     Perceptual matching on photos of unfamiliar non-objects. ",2,2,0,1,0,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,"Semantic matching task","Perceptual matching task",67,,,,0.8170731707317073,,,,,"Median LI = 0.26.   ",,"LI values ranged from -0.5 to +0.6. ",,"Found a strong negative correlation between laterality in clusters on the angular gyrus and laterality in ventral precentral gyrus for the same task i.e. a regional dissociation in laterality. The strongest interaction between task (language vs baseline) and hemisphere was found in left occipito-temporal cortex. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,"Crossed lateralisation between the angular gyrus and the ventral precentral gyrus; a strong negative correlation was found, that could not be explained by handedness or gender.",,"Right handed subjects were more left lateralized than left handed subjects. ","Found a strong correlation between global LI and right hemisphere activation, indicating that inter-subject variability in language lateralization is primarily driven by right rather than left hemisphere activation. Thus concluded that right lateralization does not imply that the left hemisphere has a redundant role in language function. ","The use of a global laterality index is inappropriate given regional heterogeneity in lateralisation, particularly in light of dissociations in laterality across regions, found here between the angular gyrus and ventral precentral gyrus. Indivudal variability in laterality was driven by differences in right rather than left hemisphere activation. ",2
60,"Sepeta, Berl, Wilke, You, Mehta, Xu, et al.",2016,"Age-dependent mesial temporal lobe lateralization in language fMRI.",,57,"28 male, 29 female",6,1,"77 epilepsy patients. ",0,,0,0,0,1,0,0,0,0,0,0,"Main focus of paper was on comparing lateralisation within the mesial temporal lobe between paediatric and adults epilepsy samples. Demonstration of lateralisation in mesial temporal lobe in healthy adults for an auditory sentence comprehension task.  ",2,1,1,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used the LI toolbox bootstrapping method (Wilke and Schitthorst, 2006). ",0.2,"Left = LI >= 0.2  Right LI <= -0.2  Bilateral -0.2 < LI < 0.2",2,1,"MTL and language areas were defined using the Wake Forest PickAtlas. ",3,5,1,1,1,0,0,0,"MTL ROI- included hippocampi and parahippocampal gyri.     Broca's area.     Wernicke's area. ",1,,2,1,7,0,0,0,0,0,0,0,0,"Auditory description decision task.","Participant hears a sentence describing and naming an object; must decide if description accurately matches the object (press button if accurate). ",1,2,2,2,"Reverse speech with tone identification- press a button when tone follows reverse speech. ",3,1,0,1,1,1,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Auditory description decision task.","Tone decision.",,,,,,0.42,,,,"Moderately strong laterality found for sentence comprehension task within MTL. Mean values not given for language areas. ",,,"Reported that MTL LIs were similar regardless of activation thresholds (value reported at the most lenient threshold of 0.05). ","Subjects who were left lateralised for Broca's and Wernicke's areas were mostly left lateralised for MTL. Those with right/bilateral language laterality had variable MTL activation asymmetry. Linear regression found that laterality in Wernicke's and Broca's areas significantly predicted laterality in MTL. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,"Language ROI laterality significantly predicted MTL laterality for the same language task. ",,,,,"Demonstrated moderate laterality within the MTL for a sentence comprehension task, with most subjects showing left lateralisation. MTL laterality was predicted by laterality within Broca's and Wernicke's areas for the same task. ",2
61,"Somers, Neggers, Diederen, Boks, Kahn, Sommer.",2011,"The measurement of language lateralization with functional transcranial doppler and functional MRI: A critical evaluation",,22,"14 male, 8 female",4,2,,0,,0,0,0,0,1,0,0,0,0,1,"Main focus of paper was on correlated fTCD and fMRI measures of laterality. Demonstration of laterality for word generation task. ",2,1,2,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used signal intensity change (measures by summing beta values) in suprathreshold voxels for the standard LI equation. Thresholds were set at the individual t values at p = 0.001. ",,"Paper did not classify dominance, but individual LI values given. Note that I have used the traditional 0.2 cut-offs to classify individuals myself and reported on form. ",2,1,"Created a mask using the automated anatomic labelling (AAL) atlas (Tzourio-Mazoyer et al, 2002). ","1 large VOI",3,0,0,0,0,0,0,"Single mask covering the following areas:  IFG (pars triangularis),  the insula,  MTG,   STG,   supramarginal gyrus,  angular gyrus. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation. ","Silent generation of word beginning with a visually presented letter (cued by an auditory tone). ",1,2,1,1,"Rest (blank screen, imagine non-verbal items e.g. a starry sky). ",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation",Rest,17,2,3,,0.7727272727272727,0.47,,,,"Paper gave individual LI values. Average given above calculated using all subjects; for only those showing left lateraisation, mean LI = 0.69. Numbers in each dominance category calculated by using traditional 0.2 cut-offs. ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Demonstrated moderate laterality for the word generation task in a mixed handedness sample using a single language ROI composed of both frontal and temporal areas. ",2
62,"Sommer, Ramsey, Mandl, Kahn. ",2003,"Language lateralization in female patients with schizophrenia: An fMRI study.",,"12 ","0 male, 12 female",1,1,"12 female schizophrenic patients. ",0,,0,0,0,0,1,0,0,0,0,1,"Main focus of paper was on comparing lateralisation between schizophrenic patients and healthy controls. Demonstration of lateralisation for a combined task analysis (CTA) using verb generation and a reverse-reading semantic decision task. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation. Single fixed threshold used of p = 0.05 (amounts to a t value of approx. 4.5, depending on the number of voxels for each individual). ","Not given. ","Not specified.",2,1,"VOIs were manuall delineated using the DISPLAY tool from the Montreal Neurological Institute (MNI). ","1 large VOI",3,0,0,0,0,0,0,"One large VOI comprising:  Broca's area (BA 44 and 45),  MTG,   STG,   Supramarginal gyrus,  Angular gyrus. ",1,,2,2,9,1,1,0,0,0,0,0,0,"Verb generation,     Reverse-read semantic decision. ","VG- silently generate a verb to go with a visually presented noun.     Reverse-read SD- word visually presented spelt backwards. Covertly vocalize the correct word and press a button if the word is an animal. ",2,2,3,1,Rest/fixation.,4,2,1,1,0,0,0,1,0,"The revers-read SD task prevents direct orthographic word recognition and forces engagement in phonological decoding. ",2,"Verb generation",Rest,,,,,,,,,,"LI values not given for verb generation task on its own (single task analysis), only for CTA. ",,,,,,,2,"Reverse-read semantic decision",Rest,,,,,,,,,,"LIs not given for task alone, only for CTA. ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Verb generation,  Reverse-read semantic decision. ","Obtained brain activation maps by analysing the fMRI scans acquired during both tasks together. ",Rest.,,,,,,0.75,,,,"Strong laterality obtained for this task combination (0.75). ",,,"SD = 0.2.",,,,,2,,,,,,,,,"Demonstration of strong left lateralisation obtained for an expressive-receptive combined task analysis in a sample of female right handers. ",2
63,"Stippich, Mohammed, Kress, Hahnel, Gunther, Konrad, et al.",2003,"Robust localization and lateralization of human language function: An optimized clinical functional magnetic resonance imaging protocol.",,14,"7 male, 7 female",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Demonstration of lateralisation for sentence generation and word generation tasks within Broca's and Wernicke's areas, in a sample of right handers. ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used cluster size (voxel counts) for the standard LI equation. Threshold adaptively set at the level of the peak activation in the non-dominant hemisphere. ","Not given. ","Not specified/used. ",2,3,"Used functional maps to identify Broca's and Wernicke's areas in the left hemisphere, helped by typical landmarks. Then generated their anatomical homologues in the right hemisphere. ",2,3,0,0,0,0,0,0,"Broca's area,     Wernicke's area. ",1,,2,2,9,1,0,0,0,0,0,0,1,"Sematic fluency,     Sentence generation.","SF- covertly generate words belonging to visually presented category.     SG- covertly generate standard predefined (learnt) sentence in response to a picture of an object. Which sentence went with each picture was learnt outside the scanner. ",1,2,1,1,Fixation.,4,2,0,1,1,0,0,1,0,,2,"Semantic fluency",Fixation,,,,,,,0.83,0.78,,"Strong laterality found across both frontal and temporal ROIs. All subjects would have been classed as left dominant within both ROIs, if traditional 0.2 cut-offs were used. ",,"Broca's area SD = 0.24,  Wernicke's area SD = 0.2.   Thus, intersubject variability in LI seems comparable across ROIs. ",,"Stronger LI found within Broca's area (0.83) than within Wernicke's area (0.78), but not statistically compared, so may not be a statistically significant difference. Both ROIs yielded strong mean LIs. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence generation",Fixation.,,,,,,,0.89,0.85,,"Very strong laterality found across both ROIs. All subjects would have been classed as left dominant across both ROIs if traditional 0.2 cut-offs were used. ",,"Broca's SD = 0.2.  Wernicke's SD = 0.15.   Thus, intersubject variability in LI seems comparable across ROIs. ",,"Stronger LI found within Broca's area (0.89) than within Wernicke's area (0.85), but not statistically compared, so may not be a statistically significant difference. Both ROIs yielded strong mean LIs. ",,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Demonstration of strong laterality obtained for both a semantic fluency and a sentence generation task across both frontal and temporoparietal ROIs in a sample of right handers. ",2
64,"Suarez, Whalen, O'Shea, Golby",2007,"A Surgical Planning Method for Functional MRI Assessment of Language Dominance: Influences from Threshold, Region-of-Interest, and Stimulus Mode. ",,13,"8 male, 5 female",1,2,,0,,0,1,0,1,0,0,0,0,0,0,"Compared the traditional method of LI equation with a new threshold-independent method (T-weighting) across multiple ROIs for an antonym generation task. ",2,2,1,6,1,0,1,0,0,0,"Standard LI equation,     t-weighted LI.",1,6,0,0,1,1,1,"Multiple thresholds (TDLC),     Unthresholded t-weighting method.","Compared two methods of LI calculation:  Standard LI equation- using voxel counts across a range of threshold values.     T-weighted LI- Used Branco et al's (2006) method. First plot a histogram of number of active voxels against T-score. Multiple this distribution by a weighting function that assigns higher weight to higher T-scores. Then use integrated areas for right and left hemispheres for a standard LI equation. ",0.1,"Left = LI > 0.1  Right = LI < -0.1  BL = -0.1 <= LI <= 0.1",3,1,"Defined anatomically using standard Human Atlas segmentations and Brodmann areas. ",6,5,1,1,0,1,1,1,"Whole hemispheres (global LI),  IFG,   Supramarginal gyrus,  Temporoparietal gyrus,  Precetral gyrus,  Middle occipital gyrus,  Transverse temporal gyrus.",1,,2,1,1,0,0,0,0,0,0,0,0,"Antonym generation","Overtly generate an antonym for a visually or auditorily presented word. Vocalized responses overtly without moving their lips (asked to speak like a ventriloquist). ",1,1,3,1,Fixation.,2,3,1,1,0,1,1,1,0,"Task requires receptive decoding, expressive encoding and vocalization, and perception of language in both visual and auditory modalities. ",2,"Antonym generation",Fixation,,,,,,,0.29,0.36,0.07,"LI strength depended on ROI. Strongest mean LI found for supramarginal gyrus. LI values given above are for threshold independent method, using visual stimulus presentation. ",,"For standard LI method, within and across subject variability was high for global and non-language LIs, but lower within language ROIs. Across 3 language ROIs, highest level of variability was found in the temporoparietal gyrus, then STG, and the least amount of variability in the IFG. ","Standard LI was highly threshold dependent- in most cases, LI switched from right to left dominant with changes in threshold level i.e. there was a zero crossing in threshold dependent laterality curve. ","Non-language ROIs- no clear asymmetry for either standard LI or t-weighted LI.   Hemispheric ROIs- no clear asymmetry for either standard LI or t-weighted LI.   Language ROIs- significant leftward asymmetry found for SMG (V = 0.36, A = 0.38) and IFG (V = 0.29, A = 0.23), but no asymmetry in TPG (V = 0.05, A = 0.06).  Statistical analyses confirmed that across both stimulus modes, mean LI in IFG and SMG were significantly more leftward than hemispheric LI, which was not significantly different from non-language ROIs LIs.  ","Standard LI method was extremely threshold dependent and yielded high variability in LI across and within subjects. Threshold independent method was thus argued to be favourable, yielding clear left lateralisation in IFG and SMG. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,"Found no significant differences across stimulus mode in the three language ROIs. However, for auditory mode SMG LI was significantly higher than IFG LI, whereas for visual mode these regional LIs did not differ significantly from one another. Thus, SMG laterality is enhanced beyond that of IFG by the use of auditory stimulus presentation. SMG also showed less variability for auditory versus visual mode. ",,,"Laterality depended on method of LI calculation and region. The effect of stimulus modality was region dependent (but generally had no effect). Hemispheric and non-language ROI LIs were highly variable and not significantly lateralised. Significant lateralisation found in IFG and supramarginal gyrus but not in temporoparietal gyrus. SMG LI was less variable and significantly higher than IFG LI for auditory but not visual stimulus presentation. ",2
65,"Sveller, Briellmann, Saling, Lillywhite, Abbott, Masterton et al.",2006,"Relationship between language lateralization and handedness in left-hemispheric partial epilepsy",,70,"24 male, 46 female",4,1,"74 epilepsy patients. ",0,,0,0,0,0,1,0,0,0,0,0,"Main focus of paper was relating lateralisation to handedness in epilepsy sample. Demonstration of lateralisation for verb generation task in mixed handedness sample. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for standard LI equation, using a single fixed threshold of p < 0.001. ",0.2,"Participants classified as either 'typical' or 'atypical' using a single threshold of 0.2. ",2,1,"Manually drawn using anatomical landmarks. ","1 large VOI",3,0,0,0,0,0,0,"Broca's area (IFG),   Wernicke's area (posterior STG),   Angular gyrus,   Middle frontal gyrus. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Verb generation.","Covertly generate a verb to go with a visually presented noun. ",1,2,1,1,"Fixation. ",2,2,0,1,1,0,0,1,0,,2,"Verb generation",Fixation,62,,,,0.8857142857142857,0.57,,,,"Moderately strong laterality found for verb generation in sample of mixed handedness. ",,"SD = 0.34.",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"A regression analysis found no relationship between LI and handedness in the control group. 8/70 subjects showed right handedness and atypical lateralisation, while 3/70 subjects showed left handedness and typical lateralisation. ",,"Moderately strong laterality found for verb generation task in a large sample with mixed handedness using a combined language ROI. No significant relationship between handedness quotient and LI was found; however a combination of left handedness with typical lateralisation was rare (3/70 subjects). ",2
66,"Szaflarski, Holland, Jacola, Lindsell, Privitera, Szaflarsk.",2008,"Comprehensive presurgical functional MRI language evaluation in adult patients with epilepsy.",,49,"37 male, 12 female",6,1,"38 epilepsy patients. ",0,,0,0,0,1,1,0,0,0,0,0,"Compared lateralisation for word generation and semantic decision tasks across frontal, temporal and global ROIs, to determine the optimal protocol for lateralising activity. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation. Used a single fixed threshold of Z score >= 2.58.",0.1,"Left = LI > 0.1  Right = LI < -0.1  BL = -0.1<= LI <= 0.1",2,2,"Defined ROIs based on activation maps, and then mirrored functionally defined regions from both left and right hemispheres. ",3,3,0,0,0,0,0,0,"Broca's area,     Wernicke's area,    Global ROI- combined frontal and temporal ROIs. ",1,,2,2,9,1,1,0,0,0,0,0,0,"Verb generation,     Semantic decision. ","VG- Covert generation of verbs to go with auditorily presented noun.     SD- decide using button press whether auditorily presented animal word is both native to the US and commonly used by humans. ",1,2,3,2,"VG- Bilateral finger tapping in response to tones.     SD- Tone decision task, decide if tone sequence contained two tones of a particular frequency. ",4,1,0,1,1,1,0,0,0,,2,"Verb generation","Finger tapping",41,3,5,,0.8367346938775511,0.35,0.42,0.11,,"Strongest LI found for frontal ROI. ",,"Frontal SD = 0.31,  Temporal SD = 0.44,  Combined/global SD = 0.29.  ",,"Weak laterality found in Wernicke's area (0.11), moderate laterality found in Broca's area (0.42) and combined ROI (0.35). ROI didn't affect significance of correlation between tasks (significant for all ROIs). ",,,2,"Semantic decision","Perceptual decision",47,1,1,,0.9591836734693877,0.49,0.54,0.36,,"Moderate laterality found across all ROIs. Strongest in Broca's area (0.54). ",,"Frontal SD = 0.36,  Temporal SD = 0.46,  Combined/global SD = 0.3.  Temporal LIs seem to show increased variability compared to frontal LIs. ",,"Stronger mean LI found in Broca's area (0.54) than Wernicke's area (0.36). Combined frontal and temporal ROI had a mean LI of 0.49. ROI didn't affect significance of correlation between tasks (significant for all ROIs). ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"Found significant correlations between the two tasks (verb generation and semantic decision) across all three ROIs (frontal, temporal, combined frontal and temporal). ",,,,,,,,"LIs for a semantic decision task were found to be more left lateralised than those of a verb generation task. May be attributable to the use of an active tone decision baseline for SD (finger tapping used for VG). However, LIs were highly correlated between tasks, across both frontal and temporal ROIs. Frontal LIs were higher than temporal LIs. ",2
67,"Tailby, Weintrob, Saling, Fitzgerald, Jackson",2014,"Reading difficulty is associated with failure to lateralize temporooccipital function",,42,"24 male, 18 female",6,1,"10 epilepsy patients with reading difficulties, 34 epilepsy patients without reading difficulties. ",0,,0,0,0,0,1,0,0,0,0,1,"Compared laterality for letter cued word generation and the verb generation task in a sample of epilepsy patients and healthy controls. ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Method of LI calculation not specified- simply cites Abbott et al (2010).   Emailed author: used same method as Abbott et al.  Generated TDLCs for each subject plotting LI as a function of number of active voxels. Threshold of chosen LI was then defined adaptively for each subject using a set of criteria. ","Not given. ","Not specified. ",2,2,"Used activation maps from the WG task to define language ROIs; constructed via a random effects analysis. ",4,5,1,1,1,0,0,0,"Inferior frontal gyrus,  Lingual gyrus,  Temporo-occipital cortex (MTG, MTS, lateral occipital cortex),  Medial cerebellum.   (Not clear if LIs calculated across all areas, or within temporo-occipital cortex only). ",1,,2,2,9,1,0,0,0,0,0,0,0,"Word generation,     Verb generation.","WG- covert generation of words in response to visually presented letter.     VG- covert generation of verb in response to visually presented noun. ",1,2,1,1,Rest/fixation,2,2,1,1,1,0,0,1,0,,2,"Word generation, verb generation.",Rest,,,,,,"0.82, 0.78.",,,,"Strong mean LI for both word generation (0.82) and verb generation (0.78). Strength of laterality appears fairly comparable between tasks. ",,"WG SD = 0.16.   VG SD = 0.26.   Some suggestion of greater variability in LI for VG compared to WG. ",,"Unknown whether LI values are calculated using all ROIs, or only the temporo-occipital ROI. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Strong and comparable laterality found for verb generation and word generation tasks; however precise ROI used not known. ",2
68,"Thivard, Hombrouck, du Montcel, Delmaire, Cohen, Samson",2005,"Productive and perceptive language reorganization in temporal lobe epilepsy.",,17,"7 male, 10 female",6,1,"18 right TLE (temporal lobe epilepsy) and 18 left TLE patients. ",0,,0,0,0,1,1,0,0,0,0,0,"Compared LIs across frontal and temporal ROIs for sematic fluency, sentence repetition and story listening.",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at a single fixed threshold of P < 0.001. ",0.2,"Traditional 0.2 cut-offs. ",2,1,"Not specified.",2,3,0,0,0,0,0,0,"Frontal ROI,     Temporal ROI. ",1,,2,3,9,1,0,0,0,0,1,0,1,"Semantic fluency,     Sentence repetition,    Story listening.","SF- covertly generate words belonging to an aurally presented category.     SR- listen to and silently mentally repeat short sentence.     SL- listen to audio story. ",1,2,3,3,"SF and SR- rest.     SL- listen to the same story played backwards. ",4,1,0,1,1,1,0,0,0,"Semantic fluency engages expressive component of language, sentence repetition is both expressive and receptive, and story listening engages only receptive component of language. ",2,"Semantic fluency",Rest,16,,1,,0.9411764705882353,,0.51,0.57,,"Moderately strong mean LIs for both frontal and temporal ROIs. Numbers in each dominance category shown for frontal LIs. ",,"Frontal SD = 0.27,  Temporal SD = 0.64.   Suggests greater variability in LI for temporal ROI. ",,"Slightly stronger mean LI found in temporal (0.57) versus frontal (0.51) ROI; but not statistically compared so may not differ significantly. Temporal ROI may show greater variability in LI (reported that the temporal lobes were inconsistently activated during the semantic fluency task). ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Story listening","Backwards speech listening",10,1,4,,0.5882352941176471,,0.72,0.43,,"Strong laterality found in frontal ROI; moderate laterality found in temporal ROI. Numbers in each dominance category are shown for temporal ROI. ",,"Frontal SD = 0.42,  Temporal SD = 0.41.  Very comparable levels of variability in LI across ROIs. ",,"Stronger laterality found in frontal (0.72) compared to temporal (0.43) ROI. No effect of ROI on level of variability in LI. Reported however that the frontal lobes were inconsistently activated during the story listening tasks. ",,,2,,,,,,,,,,,,,,,,,,,2,"Sentence repetition",Rest,7,1,9,,0.4117647058823529,,0.26,0.15,,"Weak laterality found in both frontal and temporal ROIs. Numbers in each dominance category shown using temporal ROI. ",,"Frontal SD = 0.56,  Temporal SD = 0.26.  Suggests greater variability in LI for frontal ROI. ",,"Sentence repetition was poorly lateralised in both frontal (0.26) and temporal (0.15) ROIs; not statistically compared so may not be significantly different. Between subject variability in LI appeared to be greater for frontal ROI compared to temporal ROI. ",,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,"One control subject showed slightly right sided temporal lobe activation in story listening, but normal left-sided frontal activation during semantic fluency. ",,,,"LI depended on task and ROI. Strongest lateralisation found for story listening in the frontal lobes; but reported that the frontal lobes were inconsistently activated for story listening. The next strongest laterality was found for the fluency task in both frontal and temporal lobes, then story listening in temporal lobes. Sentence repetition was poorly lateralised across both ROIs.  ",2
69,"Tie, Suarez, Whalen, Radmanesh, Norton, Golby",2009,"Comparison of blocked and event-related fMRI designs for pre-surgical language mapping.",,6,"3 male, 3 females",1,1,"8 brain tumour patients. ",0,,1,0,0,0,0,0,0,0,0,0,"Compared laterality within temporal and frontal ROIs for blocked versus event-related designs, using an antonym generation task. ",2,4,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at a single fixed threshold of p < 0.05.",Positive/negative.,"Positive LI indicated left dominance, negative LI indicated right dominance. ",2,1,"Anatomically defined using the WFU PickAtlas software. ",2,3,0,0,0,0,0,0,"Inferior frontal gyrus,    Posterior part of the superior temporal gyrus. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Antonym generation.","Overtly generate antonym for visually presented word, with minimal movement of the head, jaw or lips. ",1,1,1,1,Rest/fixation.,2,2,0,1,0,0,1,1,0,,2,"Antonym generation",Rest,,,,,,,"0.31, 0.83.","0.55, 0.16",,"Mean LI values:  Frontal BD: 0.83,  Frontal ERD: 0.31,  Temporal BD: 0.16,  Temporal ERD: 0.55.  ",,"Large degree of variability in LI values; one subject reported with an LI value of -1 (based on only 1 voxel in the RH, 0 in the LH), and another with an LI value of +1 (based on only 6 voxels in the LH, 0 in the RH). ",,,"For three controls, the event-related design showed more left lateralised activity than the blocked design. For the remaining three controls, the blocked design appeared to show more robust activation in left language areas than the event-related design. However, argued that it is possible that the two designs require different optimal threshold levels.  ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"High degree of discordance between the statistical maps generated by event-related and block designs. Generally, the event-related paradigm yielded more activation within language areas. Difficult to draw conclusions about which gave strongest laterality due to high level of individual variability. Noted that the optimal threshold required by each design may be different. ",2
70,"Tzourio-Mazoyer, Marie, Zago, Jobard, Perchey, Leroux.",2015,"Heschl's gyrification pattern is related to speech-listening hemispheric lateralization: FMRI investigation in 281 healthy volunteers. ",,281,"139 male, 142 female",4,2,,0,,0,0,0,1,1,0,0,0,0,1,"Main focus of paper was on relating structural asymmetries in Heschl's gyrus to functional language laterality. Demonstration of laterality for speech listening task within whole hemispheres using an event-related design. ",2,2,1,4,0,0,0,0,0,0,,2,4,0,0,0,0,0,,"Used the bootstrapping method (Wilke and Schitthorst, 2006) to calculate weighted hemispheric lateralization indices (HFLIs). ","Not given. ","Not specified. ",1,1,"Anatomically defined using template used for fMRI data normalization, excluding the cerebellum. ","1 ",6,0,0,0,0,0,1,"Whole hemispheres. ",1,,2,1,6,0,0,0,0,0,0,0,0,"Speech listening.","Subjects listened to a list of overlearned familiar words e.g. months, days of the week and/or seasons while fixating on a central cross. ",1,2,2,2,"After listening to speech, subjects had to detect when the central fixation cross changed to a square using a button press response. ",3,1,0,0,0,1,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Speech list listening","Visual detection",,,,,,,,,-4.88,"Mean LI was significantly right lateralised (-4.88+/-33).",,"High level of inter-individual variability in LI, with values varying from -70 to 80. ",,"Global LI reported. Variability in global LI was explained by positive correlation with BOLD variations in left Heschl's gyrus, supramarginal gyrus and anterior insular, and negative correlation with BOLD variations in the right STS. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"No significant effect of handedness on LI. ",,"Significant right lateralisation found for group mean LI using a speech listening task, global ROI and an event-related design. Explained by high level of individual variability in LI (range: -70 to 80). Lateralisation depended on activity within phonological areas that had strong and opposite asymmetries",2
72,"Van der Haegen, Cai, Brysbaert.",2012,"Colateralization of broca's area and the visual word form area in left-handers: FMRI evidence.",,57,"16 male, 41 female",2,2,,0,,0,0,0,1,1,0,0,0,0,0,"Assessed the colateralization of the IFG during speech and the ventral occipital cortex during reading. Overall correlation between these LIs was strongly significant, but there were 3 cases of cross-lateralisation.",2,4,1,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used Wilke and Schmithorst's (2006) bootstrapping method within LI toolbox, that calculated LI values iteratively (100 samples) at 20 threshold levels. A weighted overall mean LI was then calculated by assigning higher weight to higher thresholds. Note that a blocked design was used for word generation, but an event-related design was used for lexical decision.",0.5,"Left = LI > 0.5  Right = LI < -0.5  ",2,1,"Defined using the AAL template.",2,4,1,0,0,0,1,0,"IFG (pars opercularis and pars triangularis) for word generation.     Ventral occipital cortex for lexical decision task. ",1,,2,2,9,1,0,0,0,0,0,0,1,"Word generation,     Lexical decision.","WG- covert generation of words beginning with visually presented letter. A blocked design was used.     LD- decide if visually presented word is a real word or a pseudoword using a button press (yes or no). An event-related design was used.  ",1,2,3,2,"For WG: Silent repetition of visually presented non-word (baba)",4,2,1,1,0,0,0,1,0,,2,"Letter cued word generation","Silent nonword repetition",20,30,7,,0.3508771929824561,,"0.73 (left lateralised subjects) and -0.75 (right lateralised subjects) ",,,"Yielded strong lateralisation, either to the left hemisphere or to the right hemisphere- average LI values were 0.73 for left lateralised subjects and -0.75 for right lateralised subjects. ",,"Higher variability in LI value, since used a sample of left handers to increase variability in laterality.     Left lateralised group range: 0.51 to 0.94.  Right lateralised group range: -0.62 to -0.94.  Bilateral group range: -0.5 to 0.29.",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Lexical decision","Event-related design used.",18,6,31,,0.3157894736842105,"0.48 (left lateralised subjects) and -0.41 (right lateralised subjects) ",,,,"Moderately strong laterality in left dominant (0.48) and right dominant (-0.41) groups. The laterality pattern was much less clear/much more bilateral for lexical decision than for word generation. ",,"Group range: -0.49 to 0.44.   Left lateralised group range: 0.52 to 0.89.  Right lateralised group range: -0.78 to -0.51.",,"The laterality pattern was much less clear for vOT lexical decision than for word generation in IFGthere was a higher percentage of bilateral language representation in vOT than in IFG. ",,,2,,,,,,,,,,,,,,,,,,,,,2,"Correlation analysis between tasks yielded evidence of colateralization of IFG (for word generation) and vOT (for lexical decision) in the majority of left handed participants; LIs from the two tasks/regions were highly correlated (r = 0.59, p < .001). ",,,"Correlation analysis revealed colateralization of IFG and vOT in the majority of participants; LIs from the two regions were highly correlated (r = 0.59, p < .001). ","Reported 3/57 cases of crossed dominance for vOT and IFG. There was a higher incidence of crossed LI values for vOT and IFG in RH speech (IFG) dominant subjects. May reflect reduced anatomic frontotemporal connections in the RH. ",,,,"Demonstration of colateralization between IFG (during word generation) and vOT (during lexical decision) in the majority of subjects in a left handed sample. However, reported a small number of cases (3/57) with crossed dominance. ",2
73,"Van der Haegen, Cai, Seurinck, Brysbaert",2011,"Further fMRI validation of the visual half field technique as an indicator of language laterality: A large-group analysis. ",,50,Unknown,2,2,,0,,0,0,0,0,1,0,0,0,0,1,"Main focus of paper was on comparing fMRI and the visual half-field technique for measuring hemispheric dominance. Demonstration of lateralisation for word generation in a frontal ROI. ",2,1,1,4,0,0,0,0,0,0,,1,4,0,0,0,0,0,,"Used the bootstrapping method (Wilke and Schitthorst, 2006) within LI toolbox, to calculate LIs at 20 different threshold levels with 100 bootstrap resamples (resample ratio k = 0.25). Used this to calculated a weighted mean LI for each subject. ",0.6,"Left = LI > 0.6,  Right = LI < -0.6,  BL = -0.6 <= LI <= 0.6.",2,1,"Defined anatomically using the AAL template. ",6,1,0,0,0,0,0,0,"6 different frontal ROIs:  IFG pars triangularis and pars opercularis together,  Pars triangularis,  Pars opercularis,  Pars orbitalis,  Insula,  Precentral cortex",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation.","Covert generation of words beginning with visually presented letter. ",1,2,1,2,"Silent non-word repetition (baba). ",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Silent nonword repetition",25,20,5,,0.5,,"0.82 or -0.85",,,"LI values shown for IFG (combined pars triangularis and pars opercularis), for left dominant (0.82) and right dominant (-0.85) subjects separately. ",,"High level of inter-subject variability in LI due to exclusively left handed sample. LIs ranged from -0.97 to 0.96. ",,"Strongest laterality found in combined IFG ROI. Values listed below shown for left dominant subjects:    Pars triangularis = 0.79,  Pars opercularis = 0.72,  Pars orbitalis = 0.47,  Insula = 0.43,  Precentral cortex = 0.75.  ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Strongest laterality found for a word generation task within IFG ROI (both pars opercularis and pars triangularis). Weakest laterality found within the insula. Equal numbers of typical (left) and atypical (right and bilateral) lateralised subjects in this left handed sample (50:50 ratio). ",2
74,"van Oers, Vink, van Zandvoort, van der Worp, de Haan, Kappelle.",2010,"Contribution of the left and right inferior frontal gyrus in recovery from aphasia. A functional MRI study in stroke patients with preserved hemodynamic responsiveness. ",,13,"4 male, 9 female",1,1,"13 stroke patients (all right handed, assuming control group are also all right handed). ",0,,0,0,0,1,1,0,0,0,0,1,"Main focus of study was on patients (relating lateralisation to recovery). Provides demonstration of lateralisation for picture word matching, word generation and semantic decision tasks across frontal and posterior language areas. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation. ","Not given. ","Not specified/used.",2,1,"Defined anatomically using the WFU PickAtlas. ",3,3,0,0,0,0,0,0,"IFG,     Posterior ROI (angular, supramarginal, superior and middle temporal gyri),    Combined frontal and posterior ROI. ",1,,2,3,9,1,1,0,0,1,0,0,0,"Verb generation,     Semantic decision,    Picture-word matching.","VG- covertly generate a verb to go with a visually presented noun.    SD- decide whether a visually presented noun was an animal or not using a button press.     PWM- decide if a word matched a simultaneously presented picture using a button press.",1,2,3,3,"VG- fixation.     SD and PWM- press the left or right button when an arrow pointing left or right was presented. ",4,2,0,1,1,0,0,1,0,,2,"Verb generation",Fixation,,,,,,0.44,0.47,0.41,,"Moderately strong laterality obtained across all three ROIs; strongest in the IFG. ",,"Frontal SD = 0.06,  Temporal SD = 0.08,  Combined SD = 0.06.  Similar levels of low inter-subject variability found.   ",,"Similar strength of laterality found across all ROIs. Slightly higher in IFG (0.47) compared to temporal ROI (0.41), but not statistically compared. ",,,2,"Semantic decision","Perceptual decision",,,,,,0.43,0.43,0.38,,"Moderately strong laterality obtained across all three ROIs; strongest in the IFG and combined frontal-temporal ROI (both 0.43). ",,"Frontal SD = 0.07,  Temporal SD = 0.14,  Combined SD = 0.06.  Similar levels of low inter-subject variability found, slightly more variable in temporal ROI.  ",,"Similar strength of laterality found across all ROIs. Slightly higher in IFG (0.43) compared to temporal ROI (0.38), but not statistically compared. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Picture word matching","Perceptual decision",,,,,,0.47,0.47,0.48,,"Moderately strong laterality obtained across all three ROIs; strongest in the temporal ROI (0.48). ",,"Frontal SD = 0.08,  Temporal SD = 0.12,  Combined SD = 0.08.  Similar levels of low inter-subject variability found, slightly more variable in temporal ROI.  ",,"Almost identical strength of laterality found across all ROIs. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Similar moderately strong laterality found across all tasks and ROIs. Strongest laterality found for picture-word matching in temporal ROI; weakest found for semantic decision in temporal ROI. Greatest difference between posterior and frontal LIs found for verb generation task (0.41 versus 0.47). ",2
75,"van Rijn, Aleman, Swaab, Vink, Sommer, Kahn. ",2008,"Effects of an extra X chromosome on language lateralization: An fMRI study with klinefelter men (47,XXY). ",,14,"14 male, 0 female",1,1,"14 men with Klinefelter syndrome. ",0,,0,0,0,1,1,0,0,0,0,1,"Main focus of study was on comparing patients and controls. Provides comparison of LIs obtained across different frontal and posterior language ROIs for a combined task analysis approach. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at a threshold of p < 0.001. Also used cluster thresholding, in which only clusters of 5 voxels or more were included in the voxel counts for the LI equation. ","Not given. ","Not specified/used.",2,1,"Anatomically defined using WFU PickAtlas. ",5,3,0,0,0,0,0,0,"Broca's area,     Superior temporal gyrus,    Middle temporal gyrus,    Angular gyrus,    Supramarginal gyrus.",1,,2,3,9,1,1,0,0,0,0,0,0,"Verb generation,     Antonym generation,     Semantic decision. ","VG- covertly generate verb to go with visually presented noun.     AG- covertly generate antonym in response to visually presented word.     SD- decide if visually presented word was an animal using button press. ",2,2,3,3,"For VG and AG- passive dot pattern viewing.    For SD- dot cued button presses. ",4,2,0,1,1,0,0,1,0,,2,"Verb generation, antonym generation.","Passive dot pattern viewing.",,,,,,,,,,"Individual task LIs not given- only used CTA. ",,,,,,,2,"Semantic decision","Dot cued button presses",,,,,,,,,,"Individual task LIs not given- only used CTA. ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Verb generation,     Antonym generation,    Semantic decision.","Analysed the fMRI scans during all three tasks conjointly. ","VG and AG- Dot pattern viewing.  SD- dot cued button presses.  ",,,,,,0.82,0.83,0.77,,"Strong laterality found across all ROIs; strongest mean LI found for angular gyrus (mean LI = 1.00). ",,,"Highest variability seen in MTG, lowest in angular gyrus (SD = 0).     All language regions SD = 0.26,  STG SD = 0.41,  MTG SD = 0.51,  Angular gyrus SD = 0.00,  Supramarginal gyrus SD = 0.15,  Broca's area SD = 0.32. ",,"Strong laterality found across all ROIs, but varied across ROIs. Strongest in angular gyrus, weakest in MTG.   All language regions = 0.82,  STG = 0.77,  MTG = 0.69,  Angular gyrus = 1.00,  Supramarginal gyrus = 0.91,  Broca's area = 0.83. ",,,2,,,,,,,,,"Demonstration of strong laterality found across anterior and posterior language areas for a CTA consisting of verb and antonym generation and semantic decision. Strongest and least variable laterality found in angular gyrus. ",2
76,"van Veelen, Vink, Ramsey, Sommer, van Buuren, Hoogendam.",2011,"Reduced language lateralization in first-episode medication-naive schizophrenia.",,43,"39 male, 4 female",1,1,"35 schizophrenic patients. ",0,,0,0,0,1,1,0,0,0,0,1,"Main focus of paper was on comparing patients and controls. Demonstration of lateralisation for word generation within frontal and posterior ROIs. ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used voxel counts for the standard LI equation. Threshold was defined on an individual basis using half the mean maximum t-value (Fernandez et al, 2001; Jansen et al, 2006). Calculated the mean activation value in those 5% of voxels showing the highest level of activation in that ROI; threshold set at 50% of that average maximum activation value. ","Not given. ","Not specified/used. ",2,1,"Anatomically defined using the AAL atlas. ",7,3,0,0,0,0,0,0,"Insula,   MTG,   *STG,   Supramarginal gyrus,  Angular gyrus,  *IFG (pars triangularis)    *LI values only reported for STG and IFG. ",1,,2,3,9,1,1,0,0,0,0,0,0,"Verb generation,     Antonym generation,     Semantic decision. ","VG- covertly generate verb to go with visually presented noun.     AG- covertly generate antonym for visually presented word.     SD- decide if visually presented word is an animal using a button press. ",2,2,3,3,"WG and AG- passive viewing of squares on screen.     SD- perceptual decision (decide if asterisks presented on a screen corresponded to a cue pattern). ",4,2,0,1,1,0,0,1,0,,2,"Verb generation, antonym generation.","Visual pattern viewing. ",,,,,,,,,,"Individual task LIs not given- only reported CTA. ",,,,,,,2,"Semantic decision","Perceptual decision",,,,,,,,,,"Individual task LIs not given- only reported CTA. ",,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Verb generation,     Antonym generation,     Semantic decision.","Three language tasks were modelled as separate sessions within the same model. ","VG and AG- Passive square viewing,   SD- perceptual decision task. ",,,,,,,0.24,0.23,,"LI values only given for IFG and STG. Similarly weak/moderate laterality found across both LIs. ",,,"IFG SD = 0.20,  STG SD = 0.16. ",,"Very similar almost identical mean LI values found across both IFG and STG (0.24 and 0.23 respectively). ",,,2,,,,,,,,,"Found weak laterality across both a frontal and temporal ROI for a CTA consisting of verb generation, antonym generation and semantic decision in a sample of right handers, using standard LI equation with a variable threshold. ",2
77,"Vassal, Schneider, Boutet, Jean, Sontheimer, Lemaire.",2016,"Combined DTI tractography and functional MRI study of the language connectome in healthy volunteers: Extensive mapping of white matter fascicles and cortical activations.",,20,"20 male, 0 female",1,2,,0,,0,0,0,1,1,0,0,0,0,1,"Main focus of paper was on correlating structural and functional asymmetries. Demonstration of global laterality for a sentence comprehension task. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at a single fixed threshold of p < 0.05 (with minimum cluster size of 2). ","Not given. ","Not specified. ",1,,"Not specified- simply used voxel counts across the whole brain. ",1,6,0,0,0,0,0,1,"Whole hemispheres. ",1,,2,1,7,0,0,0,0,0,0,0,0,"Sentence comprehension decision task.","Decide if a pair of visually presented sentences with the same words (different grammatical constructions) had the same or different meanings. ",1,2,2,2,"Letter decision task- decide if pairs of consonant strings were identical or if they differed by just one consonant. ",1,2,0,1,1,0,0,1,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence decision","Letter string decision",,,,,,,,,0.71,"Strong leftward lateralisation found across whole hemispheres (global LI = 0.71). ",,"SD = 0.25. ",,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,,,"Found strong lateralisation for a sentence comprehension task involving both semantic and syntactic processing using a global LI. ",2
78,"Vikingstad, George, Johnson, Cao.",2000,"Cortical language lateralization in right handed normal subjects using functional magnetic resonance imaging.",,23,"17 male, 20 female",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Looked at laterality for picture naming and word generation tasks across anterior and posterior language areas. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts (volume of activated tissue) for the standard LI equation at a single fixed threshold of P < .0006 (reflects both voxel thresholding and cluster thresholding). ",0.5,"Left = LI > 0.5,  Right = LI < 0.5,  BL = -0.5 < LI < 0.5.",2,,"Not specified- assume anatomically defined as done a priori. ",2,3,0,0,0,0,0,0,"Frontal ROI (inferior and middle frontal gyri),    Temporal ROI (supramarginal, angular and superior temporal gyri). ",1,,2,2,9,1,0,0,0,1,0,0,0,"Verb generation,     Picture naming.","VG- silent generation of verb in response to visually presented noun.     PN- silent naming of line drawings of common objects. ",1,2,1,1,"VG- passive viewing of five forward slashes.     PN- passive viewing of nonsense line drawings. ",4,2,0,1,1,0,0,1,0,,2,"Verb generation","Passive slash viewing",16,0,7,,0.6956521739130435,,0.58,0.48,,"Moderately strong laterality found in both ROIs, but strongest in frontal ROI. Values in each dominance category reported based on LIs within frontal ROI. ",,"More variability among female subjects than male subjects; male subjects all had LIs of 0.6 and above, whereas some female subjects showed LIs of 0 up to 1 (frontal ROI). ",,"Stronger mean LI in frontal (0.58) than temporoparietal ROI (0.48)- but these were not statistically compared. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Picture naming","Nonsense line drawing viewing",,,,,,,0.51,0.45,,"Moderate laterality found across both ROIs, but strongest in frontal ROI. ",,"Some considerable variability in LI values across subjects. Based on temporal LIs, 2 subjects were right lateralised, about 16 were bilateral and 18 were left lateralised (note larger sample for picture naming than verb generation). Numbers estimated from frequency plot. ",,"Higher mean LI found for frontal (0.51) than temporoparietal (0.45) ROI, but not statistically compared. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,"Laterality data for both tasks was available in 10 female subjects. 8 subjects showed the same trend of lateralization for temporal ROI in naming and frontal ROI in verb generation (3 bilateral, 5 left lateralised). 2 subjects showed dissociation in their dominance across the two tasks; were left lateralised in one task and bilateral in the other (different way around for each subject). ",,,,"Evidence of dissociation in dominance classification across temporal ROI for naming and frontal ROI for verb generation in two subjects; each was left lateralised in one task (different tasks in each case) but bilateral in the other.   Case 1: -0.13 (temporal naming), 0.73 (frontal VG).  Case 2: 1.00 (temporal naming), 0.29 (frontal VG). ",,"No significant correlation between handedness index and LI. ",,"Moderately strong lateralisation found for verb generation and picture naming tasks. Strongest lateralisation found for verb generation in frontal ROI. Biggest difference between frontal and posterior LIs found for VG. Reported two cases of 'dissociated dominance' involving left lateralisation in one task but bilateral activity in the other (using temporal ROI for naming but frontal ROI for VG). ",2
79,"Vernooji, Smits, Wielopolski, Houston, Krestin, van der Lugt.",2007,"Fiber density asymmetry of the arcuate fasciculus in relation to functional hemispheric language lateralization in both right- and left-handed healthy subjects: A combined fMRI and DTI study. ",,20,"9 male, 11 female",5,2,,0,,0,0,0,0,1,0,0,0,0,1,"Main focus of study was on relating structural and functional asymmetries. Demonstration of lateralisation for verb generation task within a combined frontal and temporoparietal ROI. Note that this study used the same participants as Orellana et al (2014), but seems to be different functional data. ",2,1,1,1,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Used voxel counts for the standard LI equation, at a fixed threshold of p < 0.05.",0.1,"Left = LI >= 0.1,  Right = LI <= -0.1.",2,1,"Defined anatomically using an electronic atlas. ",1,3,0,0,0,0,0,0,"Combined frontal and temporoparietal ROI consisting of:  IFG,   Middle and superior temporal gyri,  Supramarginal and angular gyri. ",1,,2,1,1,0,0,0,0,0,0,0,0,"Verb generation.","Covertly generate a verb to go with an auditorily presented noun. ",1,2,1,1,"Passive tone listening. ",2,1,0,1,1,1,0,0,0,,2,"Verb generation","Passive tone listening",13,5,2,,0.65,0.22,0.49,0.02,,"Strength of LI depended on ROI and handedness. Values above shown for whole sample.     Frontal LI: RH = 0.87, LH = 0.28.  Temporoparietal LI: RH = 0.35, LH = -0.16. ",,"High level of intersubject variability in LI seen, due to mixed handedness sample with a higher proportion of left than right handers (higher variability seen for left than right handers). Variability appears to greater for frontal ROI compared to temporoparietal ROI across whole sample, but this did not hold for right handers only.     Combined ROI SD = 0.53  Frontal SD = 0.66  Temporoparietal SD = 0.52 ",,"Stronger mean LI for frontal versus temporoparietal ROIs across the whole sample and for both right and left handed groups individually. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"Mean LIs were significantly higher in right than left handed subjects, both for the combined ROI and for frontal and temporoparietal ROIs separately. ",,"Strength of laterality for verb generation task depended on ROI and handedness; stronger laterality seen for frontal ROIs and for right handers (LI = 0.87). Higher variability in LI values seen for left handers compared to right handers.  ",2
80,"Vingerhoets, Alderweireldt, Vandemaele, Cai, Van der Haegen, Brysbaert et al.",2013,"Praxis and language are linked: Evidence from co-lateralization in individuals, with atypical language dominance. ",,20,"4 male, 16 female",5,2,,0,,0,0,0,1,1,0,0,0,0,1,"Main focus of task was on relating language laterality to manual praxis. Demonstration of laterality for a word generation task in a group of participants with known atypical and typical language dominance, across multiple ROIs (both frontal and parietal).  ",2,1,1,1,0,0,0,0,0,0,,1,2,0,0,0,0,0,,"Used voxel counts for LI equation, in which R and L were the other way around i.e. (R - L)/(R + L). Thus, +1 indicated total right dominance. Used variable threshold adaptively defined at that level to yield the 10% most active voxels within that ROI for each participant. ","Dominance previously defined.","Subjects were selected from a larger group study (Van der Haegen et al, 2011), in which their language dominance had been established using word generation, and a 0.6 cut-off (Left = LI > 0.6, right = LI < -0.6, BL = -0.6 <= LI <= 0.6). ",2,2,"A functional ROI was defined based on the word generation-control contrast. Yielded a symmetrical ROI encompasses BA 44 and 45 (IFG). Also functionally defined a further 5 ROIs, as areas in which the word generation and tool pantomiming tasks overlapped (latter task used for praxis LI). ",6,5,1,1,0,1,0,0,"IFG,   Dorsolateral prefrontal cortex,  Ventral premotor cortex,  Dorsal premotor cortex,  SMA,  Posterior parietal cortex.",1,,2,1,1,0,0,0,0,0,0,0,0,"Word generation.","Covert generation of words beginning with visually presented letter. ",1,2,1,2,"Silent repetition of the non-word 'baba'. ",2,2,1,0,0,0,0,1,0,,2,"Letter cued word generation","Silent nonword repetition",,,,,,,"0.68 / -0.65",,,"LI values reported for IFG, for right dominant (0.68) and left dominant (-0.65) subject groups separately. Note the reversal of the sign, due to reversal of R and L in the LI equation (R - L). Strongest laterality found for dorsal premotor cortex, weakest laterality found in posterior parietal cortex (see effect of ROI). ",,,,"An ANOVA found a main effect of ROI, across the following ROI LI values (left dominant/right dominant):  dlPFC = -0.75/0.91,  vPMC = -0.59, 0.62,  dPMC = -0.93, 0.94,  SMA = -0.67, 0.63,  PPC = -0.48, 0.81.  Lateralisation for each of these 5 regions was positively and significantly correlated with every other region.     ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,,,"In most participants, all 5 ROIs lateralized to the same hemisphere for the word generation task. LIs across the 5 ROIs were positively and significantly correlated with one another. ","There were some cases in which individuals showed incongruous dominance in one area as compared to all the other areas; two subjects showed dissimilar dominance in the posterior parietal cortex, one in the ventral premotor cortex, and two in the SMA. ",,,,"Strongest laterality found for a word generation task within a dorsal premotor cortex region, lowest found within posterior parietal cortex. Found some cases in which subjects showed discrepant/incongruous dominance in one ROI as compared to the other four. Overall, laterality across the 5 ROIs (frontal and parietal) was significantly and positively correlated.",2
81,"Wilke, Lidzba.",2007,"LI-tool: A new toolbox to assess lateralization in functional MR-data ",,"12 ","5 male, 7 female",1,2,,1,"Generated synthetic imaging data using Matlab scripts. This was generated as 'null data' containing only normally distributed noise, without any form of lateralisation. ",0,1,0,0,0,0,0,0,0,0,"Aimed to compare multiple approaches to LI calculation, including voxel count vs value and global versus regional LIs, in terms of their robustness against issues such as threshold level, data sparsity, and vulnerability to outliers. ",2,1,3,1,0,0,0,0,0,0,,1,6,0,1,1,1,0,"Multiple thresholds (plotted TDLCs),    Variable/adaptively defined thresholds.","Compared multiple methods of LI calculation all based around the standard LI equation. Plotted lateralisation curves for each subject using both voxel count and voxel value. Also compared global and regional LIs. Lastly, compared the fixed threshold approach with an adaptive thresholding approach, in which the threshold is set at the mean intensity of the voxels in the image. Also looked at the effects of clustering and variance weighting, in which the variability in the value of a given voxel is taken into account when calculating the LI (highly variable voxels are devalued and so contribute less to the LI).","Not given. ","Not specified/used.",3,1,"Selected anatomically defined ROIs using masks. ",2,6,1,0,0,0,0,1,"Frontal ROI,    Global ROI (excluding the midline).",1,,2,1,1,0,0,0,0,0,0,0,0,"Word chain task.","Silent generation of word chains, in which each new word has to begin with the last letter of the previous word. First letter given auditorily by the researcher.",1,2,1,1,Rest.,2,1,1,0,0,0,0,0,0,,2,"Word chain task.",Rest.,,,,,,,,,,,,,"Demonstrated the strong threshold dependency of the LI, and some cases of switches in dominance with changes in threshold. Laterality curves allow one to assess trends towards laterality. Adaptive thresholding was reported to perform reasonably well (possibly less threshold dependent). Clustering and variance weighting greatly increased robustness against changes in threshold level (curves were flattened and stabilised). ","The global LI was more vulnerable to being affected by activation far from the ROI. For one subject, global LI showed a switch in dominance across iterations. Thus concluded that regional LIs are superior. ","Very similar lateralisation curves seen for both voxel count and voxel value LIs; these were almost perfectly correlated. Neither approach was superior in terms of robustness against statistical outliers/data sparsity at high thresholds. ",,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,,,2,,"High thresholds yielded an artificial impression of laterality in completely random data. However, clustering (minimum cluster size required) was robust against this effect of high thresholding. Neither smoothing nor using a minimum voxel count by themselves allowed resistance against this artificial skewing of laterality. ",,,,,,,"Demonstrated no differences between voxel count and voxel value LIs in terms of LI strength or robustness against threshold/statistical outliers. Clustering and variance weighting greatly improved robustness of LI against these. Global LIs were unstable; regional LIs were argued to be more sensitive. Adaptive thresholding may be beneficial. ",2
82,"Tzourio-Mazoyer, Joliot, Marie, Mazoyer.",2016,"Variation in homotopic areas' activity and inter-hemispheric intrinsic connectivity with type of language lateralization: An FMRI study of covert sentence generation in 297 healthy volunteers.",,297,"152 male, 145 female",4,2,,0,,0,0,0,1,0,0,0,0,0,0,"Compared laterality for a sentence generation task across multiple ROIs, to look at patterns of regional asymmetry across the brain in typically and atypically lateralised subjects. ",2,2,2,5,0,0,0,0,0,0,,1,1,0,0,0,0,0,,"Created contrast images for each subjects (language task versus control task). Then compared these contrast map results in left and right homotopic ROIs, by calculating the left minus the right difference in BOLD variation, to give an asymmetry index. Also computed HFLIs (global LIs) as in Mazoyer et al (2014)- but already reported in that previous paper. Then used an ANOVA for each ROI to identify the regional correlates of the differences in hemispheric lateralisation between the three dominance groups. ","Gaussian modelling.","Dominance categories were obtained using Mazoyer et al's (2014) Gaussian mixture modelling approach; classed as typical, ambilateral and strong atypical. This was based on their HFLIs (global LIs). ",3,1,"Used the AICHA atlas (an atlas of homotopic functional ROIs) to define 192 homotopic pairs of ROIs. Calculated an asymmetry index for each of these 192, then did ANOVAs to look for group differences in asymmetry across dominance groups for each region. From this, selected 58 homotopic ROIs that showed differences in asymmetry values across dominance groups. ",58,5,1,1,1,1,1,1,"58 regions across frontal, temporal, parietal, occipital, subcortical and insular regions. ",1,,2,1,8,0,0,0,0,0,0,0,0,"Sentence generation","Subjects were required to covertly generate sentences in response to presented line drawings of scenes. These sentences had to be of the same structure: the sentences were to start with a subject and its adjective, followed by a verb that described the action taking place, and were to end with another adverbial phrase of place. ",1,2,1,2,"Covert recitation of the months of the year, when presented with a scrambled line drawing (randomly interleaved with non-scrambled drawings for sentence generation). ",2,2,0,1,1,0,0,0,0,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence generation","Generation of automatic sequence of words. ",,,,,,,"0.74 / -0.53","0.28 / -0.35",,"Paper gives mean LI values for typical, ambilateral and strongly atypical groups across 58 ROIs. All these 58 ROIs show significant differences between these groups in LI in ANOVA. 50 ROIs were found to be leftward in typicals but rightward in atypicals. ",,"Typicals and atypicals were found to demonstrate mirroring patterns of asymmetry across almost all ROIs. In contrast, typicals and ambilaterals showed comparable patterns of asymmetry; both showed leftward asymmetry in 37/50 ROIs. ",,"All 58 ROIs showed a significant negative correlation between LIs for the typical group and LIs for the atypical group. Full table of ROI LIs given in paper. Some selected values (typical/atypical):  IFG (triangularis) = 0.74 / -0.53,  Supramarginal gyrus = 0.66 / 0.1,  Angular gyrus = 0.29 / -0.26,  Superior temporal sulcus = 0.28 / -0.35,  Middle temporal gyrus = 0.22 / -0.31,  Middle occipital gyrus = 0.01 / -0.51.  ",,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,,"Handedness had no effect on LI in typicals. In contrast right and left handed ambilaterals showed significantly different patterns of asymmetry (left handers showed reduced leftward asymmetry compared to right handers). ",,"Found that asymmetries at a regional level were consistent with those at a global level in terms of language dominance. Typicals and strongly atypicals showed mirroring patterns of regional asymmetries; across 50 ROIs, LI was negatively correlated between groups. Differences in asymmetry between the groups were found in both language and non-language ROIs. Demonstration of strong frontal LI for sentence generation task. ",2
83,"Zaca, Jarso, Pillai.",2013,"Role of semantic paradigms for optimization of language mapping in clinical fMRI studies.",,"12 ","6 male, 6 female.",1,2,,0,,0,0,0,1,1,0,0,0,0,0,"Compared laterality for 5 language tasks within different ROIs to establish the optimal paradigm for assessing language laterality. ",2,1,2,3,0,0,0,0,0,0,,1,5,0,0,0,0,0,,"Used Branco et al's (2006) unthresholded method, in which the weighted sum of all voxel t values in the ROI are used for the standard LI equation. ","Not given. ","Not specified/used.",2,1,"Anatomically defined using the MNI atlas. ",5,3,0,0,0,0,0,0,"IFG,  SFG,  MFG,  MTG,  STG.",1,,2,5,9,1,1,0,0,0,0,0,1,"Word generation,    Sentence completion,    Visual antonym pair,    Auditory antonym pair,    Noun-verb association.","WG- covert letter cued word generation.    SC- covert generation of word to complete visually presented sentence.    vAP- decide if a visually presented pair of words are antonyms or not.     aAP- decide if auditorily presented pair of words are antonyms or not.     NV- Decide if verb on bottom right of noun is more semantically related to that noun than verb presented on bottom left. ",1,2,3,3,"WG- Passive viewing of nonsense drawings.     SC- scan through a series of scrambled letters that resemble words in a sentence.     vAP- view drawing and decide if simultaneously presented cross is presented in the top or bottom right hand corners of the screen.     aAP- decide if pair of tones are identical.     NV- same as vAP. ",4,3,1,1,1,1,0,1,0,,2,"Letter cued word generation",Rest,,,,,,,,,,"This was the strongest lateralising task in the expressive ROIs (IFG, MFG, SFG). ",,,,"WG LI was significantly greater than the other task LIs for the three expressive/frontal ROIs, but did not differ significantly from the other tasks in the MTG or STG (posterior receptive ROIs). ",,,2,"Noun-verb association, Antonym pair","Perceptual decision",,,,,,,,,,"In general, noun-verb association was more strongly lateralising than visual and auditory antonym pair tasks; the reverse was never true in any ROI. Auditory antonym pair generally appeared the weakest lateralising task. No significant differences between auditory and visual AP, apart from higher laterality for visual AP than auditory AP in the MTG. ",,,,"Noun-verb association was more strongly lateralising than both auditory and visual antonym pair tasks in the IFG. It was also significantly stronger than auditory antonym pair in the MTG. Auditory AP was significantly less lateralising than WG across all ROIs apart from the STG. Noun-verb association was significantly less lateralising than WG in MFG and SFG. Auditory AP was the least lateralising task in both the frontal ROIs and the MTG. No significant differences between these tasks in the STG. ",,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,,,,,,,,,,,,,,,,,,,2,"Sentence completion","Nonword letter string scanning",,,,,,,,,,"Sentence completion was significantly less lateralising than WG in the IFG and MFG. SC did not in general outperform noun-verb association or visual antonym pair tasks, except noun-verb association in the SFG. SC was more strongly lateralising than auditory AP in the MTG. ",,,,"ROI affected whether SC was significantly less lateralising than WG or significantly more lateralising than noun-verb association. It was significantly less lateralising than WG in the IFG and MFG, and significantly more lateralising than noun-verb association in the SFG only. SC was more strongly lateralising than auditory AP in the MTG. ",,,2,,,,,,,,,,,,,,,,,,,,,2,,,,,,"Compared auditory and visual antonym decision tasks. No significant differences were found between the two modalities, apart from higher laterality for visual AP than auditory AP in the MTG. ",,,"Word generation was the most strongly lateralising task in the frontal ROIs. An auditory antonym pair decision task was particularly poorly lateralising in the STG and MTG. ",2