Author's School

Graduate School of Arts & Sciences

Author's Department/Program

Biology and Biomedical Sciences: Neurosciences


English (en)

Date of Award

Spring 3-6-2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Bruce A Carlson


Perception of sensory cues requires peripheral encoding followed by extraction of behaviorally relevant signal components by central neurons. Some sensory systems can detect temporal information with submillisecond accuracy, despite these signals occurring faster than the approximately 1 ms timescale of neuronal firing. In sound localization, the best studied example of this phenomenon, there are at least two distinct mechanisms for detecting submillisecond timing differences, indicating that multiple solutions to this fundamental problem exist. I investigated mechanisms for processing submillisecond timing differences by studying electrosensory processing in a time coding expert, mormyrid weakly electric fish, which can detect submillisecond differences in the duration of electric signals.

First, I measured responses of peripheral receptors to stimuli of different durations. I found that each unit responded preferentially to longer stimuli, but with response thresholds that varied among units within the behaviorally relevant range of durations. This variability establishes a population code operating at near threshold intensities in which the number and identity of responding receptors represents duration. At higher stimulus intensities all units respond independent of duration, rendering the population code obsolete. Importantly, peripheral receptors respond either to the start or end of a signal. Thus, stimulus duration is also represented by a temporal code, as a difference in spike times between receptors.

Next, I investigated the central mechanism for detection of submillisecond spike time differences by recording from time comparator neurons (Small Cells) in the midbrain. Recording from Small Cells is challenging because their somas are small and relatively inaccessible. I therefore designed a novel method using retrograde labeling to directly visualize and record from Small Cells in vivo. I showed that patterns of duration tuning vary among Small Cells due to a combination of blanking inhibition corresponding to one edge of a stimulus and variably delayed excitation corresponding to one or both edges of a stimulus. Other circuits that detect submillisecond timing differences rely either on precisely-timed inhibition or delay-line coincidence detection. I demonstrate a novel mechanism by which mormyrids combine delay-line coincidence detection with precisely-timed blanking inhibition to establish diverse patterns of duration tuning among a population of time comparators.



Permanent URL: