Abstract

Our recent study (Montagne and Zhou, JASA 2016) showed that binaural localization cues—interaural time (ITD) and level (ILD) differences—were more variably distributed when stimuli were presented stereophonically instead of from single speakers. We hypothesized that variability in listeners’ responses is directly related to variability in the binaural cues imposed by the stimulus. Here, we investigate the validity of this hypothesis by examining the distribution of ITDs and ILDs using a simulated binaural neural network. The peripheral component of this model includes an updated auditory-nerve model (Zilany et al., 2014) and the central component of this model incorporated binaural correlation and level difference analysis. The decision variable was made based on combined ITD and ILD distributions across frequencies. The modeled data were analyzed and interpreted with regard to results from a parallel behavioral test. The model results suggest that low-frequency ITDs are a major cue for sound source localization even they are ambiguously distributed with multiple peaks. On the other hand, the ILD cue can strongly modulate which ITD peak dominates the perceived source location. The network analysis further investigates potential neural mechanisms by examining decision-making based on an ILD-modulated ITD network vs. separate ITD and ILD networks.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call