The neural basis of indirect speech act comprehension: Evidence from behavioral variant frontotemporal degeneration (P4.197)

Objective: To evaluate the ability of patients with frontotemporal degeneration, who have focal dementia due to atrophy in frontal and temporal regions, to comprehend indirect speech acts. Background: Historically, the majority of language research has focused on the phonetic, semantic, and syntactic components of speech. This work suggested two main processing hubs located in left peri-Sylvian cortex: the inferior frontal gyrus and posterior-superior temporal gyrus. However, this classic model fails to account for the complexities of naturalistic communication. To fully appreciate speaker meaning, listeners often engage additional inferential processes, which may require additional brain regions. Patients diagnosed with behavioral variant frontotemporal degeneration, who show social and executive limitations due to focal atrophy in frontal and anterior temporal regions, allow us to test this hypothesis. Design/Methods: We assess inferential processing during language comprehension using a novel indirect speech task. Participants were asked to judge the meaning of a short dialogue (i.e. yes, no) between two interlocutors. Dialogues consisted of a polar question and a reply, and conditions varied according to inferential demand. Direct replies involved a syntactic rearrangement of the question into statement format. Indirect replies provided additional elaborative information explaining how or why, without giving a direct reply. Results: The data demonstrate that control subjects perform at ceiling in both direct (mean accuracy= 98.3%) and indirect (mean accuracy=97.5%) conditions. Conversely, in patients with bvFTD, we observed impaired performance in the indirect (mean accuracy=84.76%) relative to the direct condition (mean accuracy=92.86%) (Z=1.826, p=0.06). Finally, regression analyses using structural MRI images demonstrate that indirect speech comprehension in bvFTD is related to atrophy in inferior frontal gyrus, as well as medial, ventral, and dorsolateral prefrontal cortices. Conclusions: Together, our data support the hypothesis that indirect speech comprehension involves extralinguistic regions beyond the traditional peri-Sylvian language network. Study Supported by: NIH (AG043503, AG017586, AG010124, AG032953, NS063111), Dana Foundation, Wyncote Foundation, and the Hearst Foundation Fellowship. Disclosure: Dr. Healey has nothing to disclose. Dr. Spotorno has nothing to disclose. Dr. Serva has nothing to disclose. Dr. Grossman has received personal compensation for activities with Avid. Dr. Grossman has received personal compensation in an editorial capacity for Neurology. Dr. Grossman has received research support from BMS.