PSU Mark
Eberly College of Science Mathematics Department

Meeting Details

For more information about this meeting, contact Fei Wang, Hope Shaffer, Jinchao Xu, Toan Nguyen.

Title:A neuronal network model to study the context effects on pitch height comparison
Seminar:CCMA PDEs and Numerical Methods Seminar Series
Speaker:Chengcheng Huang, Courant Institute of Mathematical Sciences, New York University
Perception can be strongly influenced by preceding context. Ambiguous stimuli are well suited to study the contextual effect. A sequentially played pair of Shepard tones (comprised of octave-spaced pure tones) can be ambiguously perceived as ascending or descending if they are separated by a half octave (tritone). However, using a preceding sequence of Shepard tones above/below the first tone and between the tritone pair biases the shift perception toward ascending/descending. Paradoxically, a pitch-based decoding algorithm of neurophysiological recordings indicates a repulsive shift of the neural tone representations, raising questions about mechanisms for pitch height comparison. Our network model consists of 3 subpopulations (along tonotopy): two excitatory arrays (UP and DOWN) that drive a common inhibitory array that provides recurrent feedback but with oppositely directed asymmetric footprints. The UP and DOWN units are direction selective and decisions are made by comparing their response levels. Comparisons of Shepard tone pairs using the model, with or without biasing, agree with those in psychophysical studies. Previous tones facilitate inhibitory synaptic strengths, resulting in activity reduction (adaptation) inside the biasing region and enlargement of the detection range for UP and DOWN units at the biasing region.

Room Reservation Information

Room Number:MB315
Date:03 / 21 / 2014
Time:03:30pm - 05:00pm