Retroactive Interference of an Operant Discrimination induced by exposure to a Pavlovian-Conditioning Task: Task specific and neuronal factors

Abstract

Retroactive interference (RI) is a phenomenon where a new learning experience interferes with the recollection of a previous one. This phenomenon has been mainly studied in rodents using behaviourally conflicting contextual memories and these forms of memories critically involve the normal functioning of the hippocampus. Studies have shown that silencing the activity of the hippocampus during the second (interfering) learning experience significantly impacts the strength of RI in a contextual memory paradigm. The current study used the protocol established for the contextual memory paradigm and applied it to an appetitive learning setting that does not involve the hippocampus. Rats were first subjected to the learning of an action-outcome (A-O) association where pressing one of two levers yields a food reward. One hour later, rats were subjected to learning stimulus-outcome (S-O) association where rats are provided with cue that reliably predicts the delivery of food. The rats were tested on the A-O condition without reinforcement 24 hours later. Rats performed poorly in the A-O test when compared to a group that did not undergo the S-O condition. This suggests that the non-behaviourally conflicting S-O learning experience had interfered with the A-O task. These rats were also poorer in performance when compared to those who were tested immediately after experiencing the S-O association, suggesting that the S-O learning experience did not affect short term memory of the A-O task. Furthermore, when the delay between the initial A-O task and the rats were interfering S-O task was increased, the strength of RI tended to be diminished. Based on these initial results, it was hypothesised that the common, critical neuronal substrate, the NAc core was also critical in the manifestation of retroactive interference in our paradigm. Therefore rats were subjected to neuronal silencing of the NAc core using the designer receptor exclusively activated by designer drug (DREADD) when undergoing the classical conditioning paradigm. NAc inhibition did not impact the strength of RI. The results indicate that rats may have overlearned the initial task, and therefore future optimization of the protocol may increase the chances of obtaining a significant effect of NAc inhibition on RI. This paradigm would also be useful in testing the validity of two conflicting theories of forgetting, the consolidation and the temporal distinction theories. The former suggests that the effect of RI is due to the prevention of memory storage and the latter suggests that it is due to confusion of the two stored memories during retrieval. The behavioural findings have in this study so far have agreed with both theories and a further manipulation is proposed to tease these methods of RI apart.