Associative Representational Plasticity in Primary Auditory Cortex: Importance of Learning Strategies in Acquisition and Maintenance

Classical views on the functional organization of the cerebral cortex portray primary sensory regions as passive recipients of sensory impressions. Over the past half century or so, a body of work has accumulated in opposition of this view, which endows primary sensory regions with cognitive functions, such as learning and memory, classically localized elsewhere. Most extensively studied in the primary auditory cortex (A1), learning has been repeatedly shown to coincide with the induction of neural changes that not only enhance coding of behaviorally relevant sounds, but alter the way auditory stimuli are processed in general. The term representational plasticity (RP) has been adopted to collectively refer to these learning related changes in the processing or representation of behaviorally relevant stimuli. Recently a number of studies have demonstrated that RP, induced during learning, dissipates with extended training without impacting a subject's performance. Such findings stand at odds with a body of evidence supporting a mnemonic function for RP but might be explained by when considered in light of learning strategy use. Learning strategy, i.e. the way animals use environmental cues to solve a problem, has been identified as a critical factor regulating the induction of RP. The first experiment sought to better understand the relationship between learning strategy and RP by examining the effects of overtraining on strategy use and RP. During overtraining, map renormalization occurred and the degree of renormalization was correlated with the degree to which animals stopped utilizing a tone-onset-to-error (TOTE) strategy. We developed a novel behavioral analysis method to track and quantify strategy use during training. This revealed that strategy use was much more dynamic of a process than was previously realized. This prompted a critical test of the factor underlying the relationship between TOTE and RP. This test showed that use of the TOTE strategy alone was not sufficient to produce RP, indicating that RP was in fact related to learning. These results compelled a reconceptualization of learning strategy in a more general light. Collectively, the experiments in this dissertation provide the framework for a new understanding of the relationship between learning strategy and RP