Breadcrumb Navigation


Spatio-temporal contrast enhancement and feature extraction in the bat auditory midbrain and cortex

J Neurophysiol 110: 1257-68. 10.1152/jn.00226.2013

Authors/Editors: Hoffmann S
Warmbold A
Wiegrebe L
Firzlaff U
Publication Date: 2013
Type of Publication: Journal Articles 2001 - 2017


Navigating on the wing in complete darkness is a challenging task for echolocating bats. It requires the detailed analysis of spatial and temporal information gained through echolocation. Thus, neural encoding of spatio-temporal echo information is a major function in the bat auditory system. Here we presented echoes in virtual acoustic space and used a reverse-correlation technique to investigate the spatio-temporal response characteristics of units in the inferior colliculus (IC) and the auditory cortex (AC) of the bat Phyllostomus discolor. Spatio-temporal response maps (STRMs) of IC units revealed an organization of suppressive and excitatory regions that provided pronounced contrast enhancement both along the time- and azimuth axes: Most IC units showed either spatially centralized short-latency excitation spatio-temporally imbedded in strong suppression, or the opposite, i.e., central short-latency suppression imbedded in excitation. This complementary arrangement of excitation and suppression was very rarely seen in AC units. In contrast, STRMs in the AC revealed much less suppression, sharper spatio-temporal tuning, and often a special spatio-temporal arrangement of two excitatory regions. Temporal separation of excitatory regions ranged up to 25 ms, and was thus in the range of temporal delays occurring in target ranging in bats in a natural situation. Our data indicate that spatio-temporal processing of echo information in the bat auditory midbrain and cortex serves very different purposes: While the spatio-temporal contrast enhancement provided by the IC contributes to echo-feature extraction, the AC reflects the result of this processing in terms of a high selectivity and task-oriented recombination of the extracted features.

Related Links