How is time represented in the brain?

Behaviorally relevant sensory signals are not constant in time but vary on many scales. A sudden or gradual change in the patterns of a visual scene or the precise time course of an acoustic communication signal contain information that is of most importance for an organism, not the mean illumination level or mean sound intensity. To process and interpret time-varying signals in a meaningful way, environmental signals have to be integrated over time. As a consequence, temporal relations between and within external stimuli cannot just be encoded in a one-to-one fashion by the nervous system. These observations trigger a general question: “How is time represented in the brain?” Given the large variety and intriguing temporal complexity of many natural pattern sequences, sophisticated neural representations are likely to have been invented during the course of evolution. No simple, universal answer is therefore to be expected to the question posed above. Progress in the understanding of neural coding in the time domain may nevertheless be achieved if one concentrates on a more specific problem: “Which types of representations best support flexible and robust computations of temporal relations?” Of particular importance are algorithms that are compatible with naturally occurring signal variations such as a change of the stimulus intensity or a change of the duration of all signal components, also known as “time warp”. In the following, a collection of basic computing principles is presented. The focus is on algorithms which deal with sensory pattern sequences that vary over time scales from a few to a few hundred milliseconds.