Info

Short Duration (s)

FIGURE 3.1 Sensitivity to time is characterized by local maxima at 12 and 24 sec (A), 12 sec (B), and 0.2 and 1.2 sec (C). Unfilled symbols: average across rats. Filled symbols: a running median was performed on each rat's data, and the smoothed data were averaged across rats to identify the most representative local maxima in sensitivity. (A) Rats discriminated short and long noise durations with the duration adjusted to maintain accuracy at 75% correct. Short durations were tested in ascending order with a step size of 1 sec (n = 5) and 2 sec (n = 5). Sensitivity was similar across step sizes (r(15) = .701, P < .01), departed from zero (binomial z = 5.51, P < .001), and was nonrandom (r(14)lag1 = .710, P < .01). SEM = 0.03. (B) Methods are as described in panel (A), except short durations were tested in random order (n = 13) or with each rat receiving a single interval condition (n = 7); results from these conditions did not differ. Sensitivity departed from zero (binomial z = 6.04, P < .001) and was nonrandom (r(7)lag1 = .860, P < .01). SEM = 0.02. (C) Methods are as described in panel (A), except intervals were defined by gaps between 50-msec noise pulses, and short durations were tested in descending order with a step size of 0.1 sec (n = 6). Sensitivity departed from zero (binomial z = 4.84, P < .001) and was nonrandom (r(18)lag1 = .736, P < .001). SEM = 0.04. Sensitivity was measured using d' from the signal detection theory. d' = z[p(short response | short stimulus)] - z[p(short response | long stimulus)]. Relative sensitivity is d' -mean d'. (Adapted from Crystal, J.D., J. Exp. Psychol. Anim. Behav. Process, 25, 3-17, 1999; and Crystal, J.D., Behav. Process., 55, 35-49, 2001b).

0 0

Post a comment