Intertrial interval

Passive avoidance procedures generally use intertrial intervals of 24 18 hours, and thereby measure longer term memory. It is possible to shorten the intertrial interval to, for example, 1 hour to assess drug effects on shorter term memory (Bartus Dean 1985). On the other hand, passive avoidance procedures are not frequently used for comparisons of short- and long-term memory, probably because other procedures (see below) offer better possibilities of interpretation. 

There have now been parallel studies of 6-OHDA-induced lesions of the mesostriatal and mesocortical DA systems (Robbins et al., 1998 Baunez and Robbins, 1999). Both studies produced results that were different from those of mesolimbic DA loss, in that there were impairments in choice accuracy when the visual stimuli were presented in a temporally unpredictable manner. The impaired choice accuracy resulting from mesostriatal DA depletion was found in the context of many other behavioral deficits, including slowed responding and large increases in response latency (similar to those seen following mesolimbic DA loss). However, despite these effects, no deficits in accuracy were observed under baseline conditions. The selective disruption produced by the variable intertrial intervals may be related to the basic impairments in the readiness to respond described in earlier studies on simple and choice reaction time (Brown and Robbins, 1991). 

Figure 8 Effects of 7 mg kg trimethyltin on delayed alternation performance. Lower accuracy values were evident in TMT-treated rats at all delay values, but no impairment was seen in the zero-second delay condition, consistent with a specific effect on memory function. (Reproduced from Bushnell PJ (1988) Effects of delay, intertrial interval, delay behavior and trimethyltin on spatial delayed response in rats. Neurotoxicology and Teratology 10 237-244, with permission from Elsevier.)...

In the shuttle box the acquisition of a two-way CAR was analyzed during 5 consecutive days. The rat was put in a box divided inside into two parts by a barrier with a small gate in the middle, and the animal was trained to cross the barrier under the influence of a conditioned stimulus (CS, fight flash). If it failed to respond within 5 s, it was punished with an unconditioned stimulus (US), a footshock (1 mA). If the rat failed to respond within 5 s to the US, it was classified as an escape failure (EF). One trial consisted of a 15 s intertrial interval (IR), followed by 15 s CS. The last 5 s of CS overlapped the 5 s of US. At each learning session, the number of CARs, EFs and IRs were automatically counted and evaluated by multi-way ANOVA. 

The effects of selected compounds on the performance of C57/B10J mice on a spatial working memory task in a square water maze was assessed 30 min following the administration of a solution of drug by oral gavage. Four trials were conducted, separated by 30-s intertrial intervals. Latency to find a platform hidden below the surface of the opaque water served as the dependent measure (Symons et al., 1988). 

Three experiments were performed. In each experiment 4 groups of 10 trained animals were used. According to a 4 x 4 Latin-square scheme each group was successively subjected to each of 4 possible procedures. As shown in Fig. 2, column 2, each procedure consisted of 3 trials (indicated by I, II and III), separated by a first intertrial interval of 2 h and a second one of only 1 min. Three injections were given the first 20 min before trial I, the second immediately after trial I and the third 20 min before trial II. One of the injections of procedures B, C and D contained 5 mg/kg PCMG, while all other injections contained the vehicle (phosphate-citric acid buffer, 10 times diluted, pH 6). Dose and time were chosen on the basis of preliminary experiments. The same animals were subsequently used in a second experiment. 

Eighty trained animals, previously used in the first series of experiments ( one PCMG treatment ), were administered 2.5 mg/kg PCMG twice 20 min before trial I and 20 min before trial II. There were 2 trials with an intertrial interval of 2 h on 1 day only. Three animals refused to run within 5 min and were therefore discarded. This experiment was meant to maintain the drug-state at the same level during each of the two trials. 

Squire (1969) examined the effect of 1 mg/kg scopolamine hydrobromide administered 25 + 10 min before trial I or trial II, with varying intertrial intervals. The animals were not trained to alternate. Hence, with long intertrial intervals low alternation percentages were observed and therefore no influence of administration prior to trial II could be detected at intertrial intervals of more than 1 h. Still, it was concluded that such administrations had no effect and that the observed effect at shorter intervals was due to an influence on consolidation. 

The decrease in the alternation percentage observed after treatment with PCMG either before trial I or trial II cannot easily be explained by state-dependent learning , since with both trials under the influence of PCMG or scopolamine alternation is strongly suppressed as well. Only with a short intertrial interval of 1 min no decrease to chance levels of alternation was noted (Fig. 4). 

The spontaneous alternation behaviour shown by rats during successive trials in a T-maze is dependent on the intertrial interval. Lengthening the intertrial interval beyond 1 h is attended by a decrease in the percentage of animals alternating. 

A procedure is described to train rats to alternate at a rate of about 70 % at an intertrial interval of 2 h. 

With an intertrial interval of 2 h, administration of PCMG 20 min prior to trial I or 20 min prior to trial II reduced the alternation percentage to about 50%, whereas administration immediately following trial I had much less effect. 

Van der Poel May I try to complete the picture for the alternation behaviour of rats Squire (1969), who did not train his animals and still used intertrial intervals of more than 1 h (with these long intervals low alternation percentages are observed), was able to demonstrate that a dose of 0.4 mg/kg physostigmine caused an increase in the percentage of animals alternating. 

Fig. 3. Effects of dietary supplementation with arachidonic acid (AA 0.5%) and docosahexaenoic acid (DHA 1%) on performance in Trial 2 of spontaneously hypertensive (SHR) and Wistar-Kyoto (WKY) groups at 60-s and 60-min intertrial intervals (ITI). Data are presented as means + SEM. Both strains performed significantly worse on the 60-min ITI and with the novel stimulus than on the 60-s ITI. There were no significant effects of diet. Source adapted from Reference 37.

Rapid habituation reduces response rates in subsequent trials. Therefore, randomization or counterbalancing of the sequence in which odor treatments are presented is mandatory for experiments with repeated measures designs. To minimize effects of habituation, an intertrial interval of at least one day was established in all but one experiment. 

RL Doty, TP Gregor, RG Settle. Influence of intertrial interval and sniff-bottle volume on phenyl ethyl alcohol odor detection thresholds. Chem Senses 11 259-264, 1986. 

---