Back

Echoes as Signal, Not Noise: Reverberation Sharpens Sensitivity to Temporal Structure

Song, P.; Zhang, L.; Huang, Y.; Xu, H.; Zhai, Y.; Bao, X.; Ye, H.; Mehmood, I.; Pandit, N. S.; Wang, Y.; Tu, Z.; Chen, P.; Zhang, T.; Zhao, X.; Yu, X.

2025-12-17 neuroscience
10.64898/2025.12.15.693365 bioRxiv
Show abstract

Sensitivity to temporal structure is fundamental to hearing, yet how natural reverberation shapes this sensitivity is unclear. Here we identify echo-facilitated temporal sensitivity (EFTS), a principle whereby the auditory system exploits echoes to selectively enhance responses to rapid temporal changes. Using transitional click trains with subtle inter-click-interval (ICI) shifts, we first show that increasing echo strength systematically enlarges the sound-level step at ICI transitions in both recorded and simulated stimuli, sharpening the physical boundary between segments of distinct temporal structure. In humans, psychophysical detection of ICI changes improves monotonically with echo level in both simulated and real free-field environments, while a late EEG change response scales with echo strength as onset responses remain comparatively stable. In awake rats, electrocorticography over auditory cortex reveals parallel echo-dependent enhancement of transition-evoked activity. Single units exhibit echo-level-dependent amplification of change-related firing and improved neurometric discriminability. Laminar local field potential and current source density analyses further show that echo-dependent divergence emerges first in granular and infragranular layers before propagating to supragranular cortex, consistent with thalamocortical drive followed by intracortical amplification. Together, these findings establish EFTS as a cross-species mechanism that reframes echoes as structured signals the brain uses to sharpen temporal integration in everyday listening.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.