Neural structure of perception and memory

The ability to specifically manipulate functionally defined circuits is crucial for understanding the cellular basis of perception because percepts do not arise from a single brain region or anatomically defined population of neurons but from a distributed, sparse population of neurons whose demographic characteristics are unknown. Furthermore, manipulation of circuits generated by natural sensory experience is necessary to understand how percepts allow comprehension of a coherent world from independent basic sensory input signals. Additionally, controlled regulation of internally generated activity will provide an understanding of the causal role of translational computations on perception generation. Chapter one provides a history of evidence for neural correlates of perception and possible computational roles of internally generated activity on percept generation. Chapter two discusses a novel approach for studying perception by manipulating sparse, distributed circuits functionally defined by their activity during natural sensory experience. Specifically, the effect of activating a competing, artificially generated neural representation on encoding of contextual fear memory is described. The work was published in Science in March 2012. In summary, a cfos based transgenic approach was used to introduce the hM₃Dq receptor into neurons based on their natural activity patterns. Neural activity could then be specifically and inducibly increased in the hM₃Dq expressing neurons by an exogenous ligand. When an ensemble of neurons for one context was artificially activated during conditioning in a distinct context, animals formed a hybrid memory representation. Reactivation of the artificially stimulated network within the conditioning context was required for retrieval of the memory. The memory was specific for the spatial pattern of neurons artificially activated during learning while similar stimulation impaired recall when not part of the initial conditioning. Chapter three considers several research questions that arose as a result of the findings presented in chapter two as well as other previous studies. The chapter describes a new transgenic system using light activation, and proposes experiments to address whether a circuit that represents a new associative memory is constructed entirely de novo or if an existing neural network is modified to code for a new experienc