Why Hemispheric Lateralization of Language is Only Part of the Story in Epilepsy Surgery Candidates.

Commentary Using their large sample of epilepsy surgery candidates, Berl et al. have made an important contribution regarding the complexity of the lateralization patterns during a semantic decision task of aurally presented word definitions. They have demonstrated the need for assessing lateralization of language in both Wernicke's area (WA) and Broca's area (inferior frontal gyrus [IFG]) rather than relying solely on hemispheric lateralization. Almost three quarters of the patients had left lateralization of both IFG and WA in contrast to 97.5 percent of the healthy controls. However, they identified four left (L) dominant (L IFG/L WA, L IFG/bilateral WA, L IFG/bilateral WA, L IFG only), three right (R) dominant (R IFG/R WA, R WA/ bilateral IFG, R IFG/bilateral WA), one bilateral (bilateral IFG/ bilateral WA), and two crossed-dominant patterns (L IFG/R WA, R IFG/ L WA) in the patients. In contrast, the control group had three left dominant (L IFG/L WA, bilateral IFG/ L WA, L IFG/ bilateral WA), one bilateral (bilateral IFG/bilateral WA), and one crossed-dominant pattern (R IFG/L WA). The L IFG, three right dominant, and LIFG/R WA patterns of the epilepsy patients not found in the control group imply epilepsy-related regional reorganization. The clinical and linguistic implications of these atypical patterns, as well as their age-related development need to be further studied in surgical outcome and prospective studies, respectively. By comparing a priori categorization of language patterns with data-driven hierarchical cluster methods, Berl et al. also highlighted the problems involved in using the traditional cutoff points for the lateralization index ([LI] –0.2–0.2) for assessing laterality strength. Their data driven classification demonstrates a LI cutoff of +0.5 for left and –0.5 for right laterality rather than the traditional >0.2 and <0.2 respective cut points. Their findings also indicate that bilateral language is found in epilepsy patients with an LI from +0.4 to –0.4 instead of the traditionally used LI range between +0.2 and –0.2. The findings of this study dovetails with those of Seghier et al. (1) who concluded from their findings in 82 healthy volunteers the importance of both identifying within hemisphere as well as hemispheric fMRI regional activation prior to neuro-surgical intervention. Using a semantic matching task, Seghier et al. (1) reported that individuals vary in the degree of their OBJECTIVE: Functional MRI is sensitive to the variation in language network patterns. Large populations are needed to rigorously assess atypical patterns, which, even in neurological populations, …