Using ATL-optimized fMRI to investigate speech perception challenges

Poster Session B - Sunday, April 14, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC

Jaimy Hannah^1,2 (jhanna62@uwo.ca), Jennifer Rodd³, Ingrid Johnsrude^1,2, Richard Binney⁴; ¹University of Western Ontario, ²Brain and Mind at Western, ³University College London, ⁴Bangor University

Speech perception relies on a network of brain regions known as the core speech network. When faced with challenges such as background noise or semantic ambiguity, we see activation extend into the multiple demand network with different types of challenges leading to different patterns of brain activation. A key region for semantic cognition, the anterior temporal lobe (ATL), is often not shown to be active in fMRI studies, due in part to significant distortions and signal dropout within this region. Recent fMRI techniques, however, have been able to retrieve signal from the ATL. The present study employed ATL-optimized fMRI to investigate acoustic and semantic challenges in the brain. The stimuli consisted of 116 sentences with or without homophones. Half these sentences were presented clearly and half in background noise. Sentence trials were intermixed with trials of silence and signal correlated noise (SCN). We found that the left ATL was active during all speech conditions relative to SCN and silence, however, we found no differences in the ATL across speech conditions. Acoustic ambiguity led to increased activation in bilateral insula, anterior cingulate, and inferior frontal gyrus. Semantic ambiguity led to increased activation in left inferior frontal gyrus. Results for semantic and acoustic ambiguity were consistent with previous findings. The ATL activation we found is consistent with its role as a semantic hub and suggests a non-selective involvement in semantic processing of naturalistic speech. That is, the left ATL appears to be active for semantically rich auditory stimuli, regardless of listening challenges.

Topic Area: LANGUAGE: Semantic