Dynamics of sequence learning in rats: The influence of reminders and training by blocks.
1. Univ. Arizona, Tucson, AZ (2) Psychology (3) Psychology, Cognitive Sci., (4) Psychology, Applied Mathematics
5. Psychology, McGill Univ., Montreal, PQ, Canada
Memory reconsolidation involves previously fixed memories returning to labile state in which they are subject to change. Experiments investigating reconsolidation in rodents have used protein synthesis inhibitors to block the reconsolidation process, resulting in memory impairments, while reconsolidation studies with humans have also demonstrated constructive memory effects. These findings suggest that reconsolidation may serve to update and integrate episodic memories encoded in similar environmental contexts. In an attempt to design an animal paradigm that will more closely relate to the human reconsolidation studies, we investigated the dynamics of learning in rats trained on spatial sequences involving feeders spaced around an open-field circular arena. We first examined the effects of reminders of a previously learned sequence on the acquisition of a novel sequence. In one condition a rat received no reminder of the prior sequence before training for the new one, whereas a second rat briefly rehearsed the first sequence. The rat in the reminder condition had more intrusions of the first sequence into the second one, and also continually visited feeders from both sequences during free recall. The rat in the no reminder condition steadily decreased and stopped its visits to the old sequence feeders. These results indicate that reminders of a previously learned sequence slow down acquisition of a new sequence and preserve memory for the old sequence.
We then explored the effect of the type of sequence training on errors and on subsequent recall. One rat was trained on two sequences in two separate sessions (blocked condition), while the other rat learned both sequences in a random order in each session (mixed condition). Results show that the rat in the blocked condition made significantly fewer errors than the rat in the mixed condition and that the majority of these errors were intrusions from the sequence the rat was not currently running. During free recall, the rats appeared to rely on different strategies. The rat in the blocked condition more often tried the feeders from whichever sequence it had run most recently, whereas the rat in the mixed condition consistently visited feeders from one specific sequence over the other. These findings suggest that training on sequences in a mixed, but not blocked, fashion primes the rat to make errors from the wrong sequence and that the structure of training affects recall strategy.
These experiments are the first steps in the design of a sequence learning paradigm that will enable the study of constructive memory effects in memory reconsolidation in rats.
Keyword (Complete): reconsolidation; spatial learning; memory