Midbrain band-enhanced neurons encode vowels via robust across-frequency contrast

Evidence

Figure 4 shows model responses to the vowel/æ/for the three MTF types for healthy and impaired model AN inputs. The healthy band-enhanced responses (Fig. 4a , blue) reflect the patterns of F0-synchronized rate shown in Fig. 2 ; responses are decreased near formant frequencies (orange arrows), because synchrony capture in the healthy AN reduces the low-frequency fluctuations that ultimately drive this IC type. Band-enhanced neurons tuned to frequencies between formants (Fig. 4a , green arrow) have strong responses due to the strong F0-synchronized rate in the healthy AN for these frequency channels (Fig. 2d , green arrow). The strong contrast in rate across frequency channels for the band-enhanced responses is robust across SPLs and in background noise (Carney et al. 2015 ). — Dijk, Pim van, Physiology, Psychoacoustics and…, 5488