Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Beyond homologous lesions: A comparative analysis of behavioral and neuroanatomical features of crossed aphasia and lesion-matched left-hemisphere aphasia
Poster Session B, Wednesday, September 30, 4:30 - 6:30 pm, Wangari Maathai
Cong Du1, Nina F. Dronkers1, Maria V. Ivanova1; 1University of California, Berkeley
Crossed aphasia (CA) – aphasia following right-hemisphere (RH) stroke in right-handed individuals – presents a fundamental challenge to classical models of left-hemisphere language dominance. Prior literature has focused largely on cortical lesion localization and language symptoms in cases of CA (Marien et al., 2001; Ishizaki et al., 2012). However, several questions remain. Do cases of CA exhibit behavioral and neuroanatomical patterns comparable to typical aphasia after left-hemisphere (LH) stroke? The present study systematically compared RH stroke-induced aphasia with anatomically matched homologous LH lesion aphasia groups to characterize behavioral similarities and differences, as well as cortical and white matter neuroanatomical features. We report 6 chronic CA cases and a cohort of 113 chronic LH aphasia cases serving as the comparison pool. 2 CA cases were diagnosed with Broca’s aphasia, 1 with Wernicke’s aphasia, and 3 with anomic aphasia, as classified by the Western Aphasia Battery (WAB; Kertesz, 2007); the mean WAB Aphasia Quotient (AQ) across all CA cases was 74.94. To identify homologous LH comparison groups, RH lesions were flipped into the left hemisphere in MNI space and compared against LH stroke lesions using the Dice coefficient. LH cases yielding a Dice coefficient ≥ 0.5 were selected and further restricted to matching aphasia subtypes. For each RH case and matched LH group, Crawford & Howell’s modified t-test was used to compare behavioral performance. Neuroanatomical comparisons included language-related cortical regions of interest (ROIs) and white matter tract lesion loads. To further investigate white matter disconnection patterns, disconnectome maps were generated for each RH case using BCBToolkit (Foulon et al., 2018). Compared with homologous and subtype-matched LH groups, CA cases generally exhibited relatively preserved language performance despite clear RH-specific visuospatial impairments (e.g., decreased block design performance). Specifically, the RH Broca’s cases showed relatively spared reading and writing, while the RH Wernicke’s case showed relatively preserved repetition and comprehension. The RH anomic cases demonstrated less pronounced differences, possibly reflecting ceiling effects associated with milder language deficits. Overall, RH cases had higher AQ scores and outperformed matched LH groups across multiple language measures. RH lesions involved cortical ROIs and white matter tracts homologous to LH areas commonly associated with post-stroke aphasia. Across matched LH groups, white matter tract lesion load profiles showed greater consistency than cortical ROI lesion load profiles. All CA cases showed involvement of the Corpus Callosum, while the Anterior Commissure was implicated in 3 cases. Disconnectome analyses revealed disconnection patterns extending into the left hemisphere in all RH cases, possibly reflecting disruption of interhemispheric commissural pathways and indirect effects on left language networks. Notably, one RH case demonstrated aphasia despite near-zero involvement of canonical language-related cortical ROIs, highlighting the potential contribution of white matter pathways to crossed aphasia. These findings suggest that CA may not represent a simple homologous counterpart of typical LH aphasia. Homologous lesion anatomy did not consistently correspond to similar behavioral phenotypes across hemispheres. Systematic examination of white matter disconnection may provide complementary insights into language impairments following RH stroke and help refine neuroanatomical models of crossed aphasia.
Topic Areas: Disorders: Acquired,