A Neuronal Network Model for Context-Dependent Perceptual Decision on Ambiguous Sound Comparison

Friday, February 27, 2015 - 12:00 to 13:00
Benedum G28
Speaker Information
Chengcheng Huang
New York University, Courant Institute

Abstract or Additional Information

Many natural stimuli contain perceptual ambiguities that can be cognitively resolved
by the surrounding context, e.g. preceding stimuli. In audition, preceding context can bias the
perception of speech and non-speech stimuli. Here, we developed neuronal network model to
account for how preceding auditory stimuli affects our perception of pitch change direction. Our
model draws inspiration from a recent psychophysical experiment where listeners experienced
opposite percepts (either ascending or descending) of an ambiguous tone pair depending on the
spectral positions of the preceding tones. Our recurrent firing-rate network model can detect
frequency change of successively played stimuli due to asymmetric inhibition. We propose a
novel adaptation mechanism, facilitation of inhibitory synapses, which successfully accounts for
the context-dependent perception demonstrated in behavioral experiments.

Research Area