INVESTIGATING NEUROMODULATOR RELEASE DURING IMPLICIT LEARNING IN A STOCHASTIC ALTERNATING SERIAL REACTION TIME TASK IN MICE

Introduction
To survive in dynamic environments, the brain employs various learning strategies, including implicit sequential learning. This fundamental mechanism enables the subconscious acquisition of action sequences, such as those used in speaking or riding a bicycle.
Objective
The primary goal of our research was to develop a mouse model analogous to human studies, allowing the comparison of characteristics measured in humans and animals. Additionally, we aimed to understand how the dopaminergic and cholinergic systems encode this process.
Methods
We trained mice to follow a four-light sequence via nose pokes for water rewards. Rewards were delivered after every fourth correct choice. The sequence structure alternated between fixed and random elements, creating frequent and rare triplets. We included test trials with all four lights on, requiring the animal to choose. Using neuromodulator biosensors, we measured dopamine (DA) release from the ventral striatum (VS) and prefrontal cortex (PFC), and acetylcholine (ACh) release from the basolateral amygdala (BLA).
Results
Animals responded faster and with fewer omissions to frequent triplets compared to rare ones. In test trials, accuracy for frequent triplets was superior and exceeded chance levels. Based on performance, mice were classified into "implicit" or "explicit" learner groups. The "implicit" learners showed higher DA release in the VS and increased ACh in the BLA, while "explicit" learners displayed higher DA release in the PFC.
Conclusions
Behavioral data indicate animals distinguished between frequent and rare triplets, while the difference in brain activity patterns suggest network dominance depends on the learning type.