Modelling timing performance on the peak procedure

Expectancy in humans in multisecond peak-interval timing with gaps

“…Moderate or weak negative correlations between start and spread times also suggested that some variability was related to response threshold variability. These conclusions were similar to those in animal studies (e.g., Cheng & Miceli, 1996;Cheng & Westwood, 1993;Church, Meck, & Gibbon, 1994;Matell, Bateson, & Meck, 2006), revealing notable similarities in timing with humans, birds, and rodents (see Penney, Gibbon, & Meck, 2008). Correlations were also analyzed to examine the contribution of memory and threshold processes in the present study.…”

Scalar Properties in Animal Timing: Conformity and Violations

“…Scalar variance can be tested by calculating a standard deviation or coefficient of variation for the response function and plotting this against the FI value, or by superimposing the response functions on the same relative scale (e.g., elapsed time in the interval scaled by the FI value in force). For example, Church et al (1994) used rats and measures derived from individual trials on PI and found that both the means and standard deviations of these measures increased linearly with FI values ranging from 15 to 60 s. This was confirmed later on with pigeons and FI values of 5, 10, and 20 s (Cheng & Miceli, 1996; see also Cheng & Roberts, 1991). Superimposition was found using rats (Church et al, 1998), pigeons (Gibbon, Fairhurst, & Goldberg, 1997a), and starlings (Rodriguez-Girones & Kacelnik, 1999).…”

A Model of Interval Timing by Neural Integration

“…This strong prediction of the model might seem to conflict with a general consensus in the timing literature that timed response time distributions are approximately Gaussian (implying a skewness of nearly 0). For a small enough CV, however, a skewness-to-CV ratio of 3 implies only a very small positive skewness, and in any case, it is well known that a heavier right tail often arises in empirical response time distributions in timing tasks (e.g., Cheng and Miceli, 1996; Cheng and Westwood, 1993; Rakitin et al, 1998; Roberts, 1981). In the next section, we show that data from rats in the peak-interval task and humans in a time-reproduction task are clearly consistent with this prediction.…”

“…In calculating these values, we adopted exclusion criteria commonly used in the timing literature: specifically, we excluded trials in which the start time exceeded the FI, the stop time was less than the FI, or the stop time exceeded three times the FI (cf. Balci et al, 2008a; Cheng and Miceli, 1996). We compared fits of Gaussian, gamma and inverse Gaussian distributions to the data, and we further estimated the coefficient of variation (CV) and skewness of these middle times (computed from the empirical data) to test whether the skewness-to-CV ratio equalled the predicted value of 3.…”