Dynamic attending and responses to time.

A temporally based theory of attending is proposed that assumes that the structure of world events affords different attending modes.Future-oriented attending supports anticipatory behaviors and occurs with highly coherent temporal events.Time judgments, given this attending mode, axe influenced by the way an event's ending confirms or violates temporal expectancies.Analytic attending supports other activities (e.g., grouping, counting), and if it occurs with events of low temporal coherence, then time judgments depend on the attending levels involved.A weighted contrast model describes over-and underestimations of event durations.The model applies to comparative duration judgments of equal and unequal time intervals; its rationale extends to temporal productions/extrapolations.Two experiments compare predictions of the contrast model with those derived from other traditional approaches.One characteristic of modern society is a preoccupation with fixed time schedules and standardized timekeepers.We maintain appointments at hourly intervals, rush to meet the 5:00 p.m. bus, and dine at predetermined hours.Yet our natural ability to judge time remains poorly understood.How often do we estimate the time elapsed since last glancing at a clock and discover with surprise that we were fairly accurate?Surprise is understandable because at least as often we lose track of time and err.The validity of these impressions is confirmed by laboratory research showing that duration judgments depend not only on actual physical duration but also on a variety of nontemporal factors.These include the spatial layout and complexity of an event as well as the attentional set, skill, affect, and constitutional state of the judge (Allan, 1979;Fraisse, 1984; Kristofferson, 1984).Researchers have addressed many of these issues that include both psychophysical problems (e.g., Weber's Law for time discrimination) and organismic variables (e.g., age, drugs, and arousal effects).Of recent interest is the influence of nontemporal information on time judgments, due largely to a fascination with such problems as the filled interval effect.This phenomenon reveals that two equivalent time intervals may not be judged as such because of the nontemporal information that fills them.Although the most popular models of judged duration attempt to explain this effect (e.g., Block, 1978;Ornstein, 1969), the effect itself raises problems for a general theory of time estimation (Allan, 1979).