Therapist perspectives on the clinical utility of hand performance information from at-home egocentric video in outpatient neurorehabilitation: a multi-methods evaluation study.

Humans recognize voices rapidly and automatically, supporting communication and social inference. Despite the robustness of this ability, its neural basis remains incompletely understood. Neuroimaging studies have consistently shown preferential activation to voices across anterior, middle, and posterior superior temporal gyrus (pSTG) and sulcus (pSTS), in responses to vocal sounds (; ; ; ; ). However, these findings remain largely correlational, leaving unresolved which auditory cortical representations are necessary for categorical voice perception (CVP), and whether apparent voice sensitivity reflects categorical processing per se or sensitivity to stimulus properties that covary with vocal sounds. A central unresolved issue concerns the organization of voice processing within auditory cortex. Serial hierarchical models posit that CVP emerges through stepwise transformations from primary auditory cortex to higher-order temporal regions (; ), predicting that early auditory areas provide prerequisite input to downstream voice-selective sites. In contrast, parallel-distributed models propose that primary and nonprimary auditory regions contribute jointly to categorical perception via partially independent, reciprocally connected pathways (). Distinguishing the relative contributions of these accounts requires causal perturbation of specific auditory nodes while behavior is measured. We tested whether CVP depends on a distributed posterior auditory circuit in which primary and nonprimary regions are jointly required. We present a patient with extensive resection of left STG and STS, with preservation of posteromedial Heschl’s gyrus (pmHG) and pSTG. This lesion pattern constrains auditory processing in the left hemisphere eliminating large portions of canonical "temporal voice areas”. We combined quantitative lesion mapping, intracranial recordings, focal electrical stimulation, and connectivity analyses to evaluate the causal contributions of pmHG and pSTG to CVP. If CVP depends on a distributed encoding in a posterior auditory circuit, then perturbation of either node would impair CVP.