Normal older participants (aged 60-79 years), with known scores on the Culture Fair Intelligence Test, were tested on 4 timing tasks (i.e., temporal generalization, bisection, differential threshold, and interval production). The data were related to the theoretical framework of scalar timing theory and ideas about information processing and aging. In general, increasing age and decreasing IQ tended to be associated with increasing variability of judgments of duration, although in all groups events could be timed on average accurately. In some eases (e.g., bisection), performance differences between the older participants and students nearly 50 years younger used in other studies were negligible. Anecdotes and some experimental findings suggest that time experience and behavior may change with increasing age from adulthood. As usual, in the psychology of time an initial problem is to decide on interrelations between diverse findings that may have no common theoretical basis (Wearden, 1994). For example, various authors have reported an apparent speeding up of phenomenological time experience with age (e.g., "Christmas seems to come round quicker every year"), a change claimed to be sufficiently orderly to warrant mathematical description and reification into "laws" (e.g., Doob, 1971; Lejeune & Pouthas, 1991; Lemlich, 1975), but it is unclear what relation this effect may have to judgments that people of different ages make about the duration of events in laboratory experiments (e.g., those that require the production of time intervals with specific durations). Laboratory studies are few and inconsistent. For example, Feifel (1957) required older and younger participants to produce intervals of 30, 60, 180, and 300 s and found that although productions from both groups underestimated the real-time duration, older participants underestimated it more, a finding not replicated in a similar study by Surwillo (1964), in which no clear age differences were found. McGrath and O'Hanlon (1968) studied the production of longer durations (1-8 minutes) and again found that older participants produced shorter durations, particularly at the shorter time values, and Kline, Holleran, and Orme-Rogers (1980) found the shortest durations produced by older participants when they counted to produce intervals from 30 to 90 s. In one of the few studies to report the variability of