next up previous
Next: Steve CarcieriJohn Sinclair, Up: No Title Previous: Carlos Brody

Dean Buonomano

Department of Neurobiology, University of California at Los Angeles

How does the brain decode temporal information?

All the information from our environment arrives in the brain in the form of spatial-temporal patterns of action potentials generated at the sensory layers. Depending on the nature of the stimuli and of the task, the defining features of a stimulus may be in the spatial and/or temporal domain. Many complex sensory stimuli, such as speech, rely heavily on temporal information on the scale of tens to hundreds of milliseconds, the time-scale considered here. In my talk I will discuss psychophysical, electrophysiological, and computational data aimed at understanding how we decode temporal information. The psychophysical experiments use a simple temporal task, interval discrimination, to study learning and generalization of temporal stimuli. The electrophysiological and computational studies explore the hypothesis that short-term forms of synaptic plasticity and other time-dependent neuronal properties underlie the decoding of temporal information. Specifically, (1) as a result of short-term forms of plasticity, synaptic strength conveys information about the temporal features of a given stimulus, and (2) in a large network, these time-dependent changes in synaptic strength (and other properties) result in the activation of different populations of neurons, which in turn can form a population code of temporal features. Together these studies support the notion that temporal information may be encoded in a population code, and that specialized timing mechanisms such as delay lines, oscillators or internal clocks may not be necessary for temporal processing on the time scale of tens to hundreds of milliseconds.

Tony Zador
Sat Mar 27 10:58:21 PST 1999