Timing without a timer.

Three separable aspects of the StaddonHiga article may be distinguished: (a) a general critique of accounts of temporal regularities in behavior that appeal to inferred entities such as ‘‘internal clocks’’ or ‘‘pacemakers,’’ (b) a proposal for an alternative approach to temporal control—the multipletime-scale (MTS) model (Staddon & Higa, 1996), and (c) a comparison of MTS with a particular inferred-process theory—scalar expectancy theory (SET) (Gibbon, 1977; Gibbon & Church, 1990). The first and second aspects of the article are addressed in this commentary. (As a preliminary matter, we use the phrase temporal control as shorthand for an orderly relation between behavior and time measured from some event. Clearly, time itself is not a stimulus. There is no environmental or neural event that uniquely and uniformly covaries with time from an arbitrary stimulus and, hence, no possibility that natural selection could have produced a neural system for its detection.)