The classically conditioned nictitating membrane response: The CS-US interval function with one trial per day

The CS-US interval (ISI) function in rabbit nictitating membrane response conditioning with very long intertriai intervals

“…The results confirm earlier findings (Levinthal, 1973;Levinthal & Papsdorf, 1970) that show an optimal lSI considerably longer than the 2oo-250-msec value commonly found for this response system in the classical conditioning literature. In addition, it has been shown that, with an lSI of 1,100-2,200 msec, an extended ill of up to 48 h in length can support conditioning.…”

“…Administering 3 trials per day, Ferrara and Guisto (1973) found no significant lSI effect with values of 200, 700, or 1,200 msec. Moreover, with 1 trial per day (intertrial interval-ITIof 24 h), a 1,250-msec group shows faster CR acquisition than a 250-msec group (Levinthal, 1973;Levinthal & Papsdorf, 1970), just the opposite of the traditional lSI function reported with massed-trial training. Data from control groups for pseudoconditioning or sensitization in the Levinthal (1973) CR performance to be due to genuine conditioning processes rather than to procedural artifacts.…”

Classical conditioning of the rabbit nictitating membrane response can be fast or slow: Implications for Lennartz and Weinberger’s (1992) two-factor theory

“…In the case of the NM response, Kehoe and Gormezano's (1974) finding is by no means the fastest known. Levinthal and his associates have consistently observed CR acquisition in the range of 2 to 8 trials under 1 trial per session (Levinthal, 1973;Levinthal & Papsdorf, 1970;Levinthal et al, 1985). In any event, the difference in maximum rates between the nonspecific and specific responses amounts to only a handful of trials-a far smaller difference than the dozens of trials that are the case when the debatable practice of averaging across studies was used.…”

“…Figure 1 summarizes those [mdings. It shows the mean number ofCS-US trials to the first CR as a function of the number of trials per session for each group in the available studies (Kehoe, Cool, & Gormezano, 1991;Kehoe & Gormezano, 1974;Levinthal, 1973;Levinthal & Papsdorf, 1970;Levinthal, Tartell, Margolin, & Fishman, 1985;Salafia, Daston, & Martino, 1976;Salafia, Terry, & Daston, 1975). The data were plotted by using Lennartz and Weinberger's (1992) method to approximate the trial of the first CR.…”

“…For example, if mean percentage CR exceeded the 10% criterion in the third block and each block contained five trials, then the value used was 12.5-the midpoint between 11 and 15. (The data from the earliest examination of trials per session [Hupka, Massaro, & Moore, 1968] tracted for each group in the available studies (Kehoe et a1., 1991;Kehoe & Gonnezano, 1974;Levinthal, 1973;Levinthal & Papsdorf, 1970;Levinthal et a1., 1985;Salafia et a1., 1976;Salafia et a1., 1975). be used, because the size of the blocks were too large for a useful approximation.)…”