Temporal cognition and neural oscillations

•Temporal cognition requires both temporal interval and phase mapping.•Representing time is a signature of temporal individuation or 'intelligent agent'.•Phase and power of neural oscillations are under endogenous control.•The phase of neural oscillations is a possible metric for internal timing.•Duty cycles bound the duration of accumulation of evidence.

There remains considerable controversy concerning how time is represented in the brain and the lack of consensus has presented a challenge for our understanding of biologically plausible timing mechanisms. In a computational brain, (continuous) time should be discretized. For instance, in Bayesian models of brain functions, the discretization of time raises issues such as what duration or prediction interval should be used to compute the probability that a discrete event (e.g. a neural spike) will occur. Typically, the researcher in the lab defines durations a priori (e.g. the window within which we count spikes) but how the brain sets its own duration or temporal boundary to accumulate evidence for representing time is unclear. When dealing with space, spatiotopic mapping seen throughout cortex intrinsically provides the spatial reference for information processing; by analogy, when dealing with time, neural oscillations are hypothesized to provide a built-in chronoarchitecture for information processing. However, how such structuring principles become conscious and how time becomes intelligible to the mind remains unclear.