Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Individual differences in behavioral and neural sensitivity to phonetic contrasts
Poster D71 in Poster Session D, Wednesday, October 25, 4:45 - 6:30 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Sara Guediche1, Rachel Theodore, Hannah Olson, Matt Phillips, Emily Myers; 1University of Connecticut
Decades of research have pointed to a central role for the superior temporal gyrus in the perception of phonetic category structure. Historical proposals have suggested that the left and right hemispheres may exhibit differential sensitivity to acoustic properties and thus sound category distinctions, with left-hemisphere sensitivity for sounds distinguished primarily by short-duration timing differences (e.g. stop consonants) and right-hemisphere sensitivity for sounds distinguished primarily by longer-duration spectral differences (e.g., vowels). Yet stop consonants and vowels also differ in the degree to which they are perceived categorically; specifically, listeners have been reported to perceive stop sounds more categorically than vowels. This raises the possibility that gradience of perception, rather than the type of contrast itself, could drive differences in hemispheric sensitivity. Individuals also vary in their perception of speech sounds, with some listeners perceiving more gradient distinctions within and between categories while others show a more classic “categorical perception” pattern with diminished sensitivity to contrasts within a category. The current study uses a within-subject design to examine behavioral and neural individual differences in the perception of stimuli that vary along both stop voicing (/g/ vs. /k/) and vowel (/æ/ vs. /ɛ/) dimensions. Pilot experiments were used to select stimuli for four 6-step continua that collectively spanned both dimensions (i.e., /gæ/ - /gɛ/, /kæ - /kɛ/, /gæ/ - /kæ/, /gɛ/ - /kɛ/ resulting in a stimulus matrix consisting of 36 items. Thus far, thirty subjects across different age groups participated in two sessions, separated by 1-3 weeks. At the first session, participants performed a visual analogue scale task (rating stimuli on a scale of 1-6) and a discrimination task (same-different decisions) for each of the 4 continua. They also completed hearing assessments and a battery of cognitive and language assessments. At the second functional magnetic resonance imaging (fMRI) session, participants passively listened to the full matrix of 36 stimuli in the scanner. On 10% of the trials, participants heard a different syllable (catch trial) and were asked to press a button. Of interest, performance on the visual analogue scale and discrimination tasks did not qualitatively differ for stop and vowel continua; however, extensive individual variability in gradient sensitivity across all continua was observed. Further analyses will assess the relationship between individual differences in gradiency for stop and vowel continua and the relationship between behavioral measures and graded vs. categorical neural representations. In addition, the current study will use representational similarity analysis (RSA) to compare whether a theoretical categorical or graded model better predicts the similarity matrix for patterns of brain activity in predefined regions of interest within the superior temporal gyrus, in both the left and right hemisphere. Results will speak to potential left versus right hemisphere differences in sensitivity to the acoustic differences between phonetic contrasts and perceptual gradiency.
Topic Areas: Speech Perception,