Neural correlates of fear learning in the amygdala and the infralimbic cortex

Abstract

Fear is one of the most intensely studied fields in emotion, due to its simple and well-known animal model, the Pavlovian fear conditioning. Numerous studies have reported that the amygdala and its surrounding brain network are critically involved in fear conditioning and extinction. However, the long-term effects of fear learning have remained largely unknown since most of the previous studies used behavioral paradigms in which memory retrieval was tested only in the short-term. Therefore, I employed a fear learning paradigm that consists of fear conditioning and extensive extinction that spans several days. In the first chapter, I examined how neurons in the lateral amygdala (LA), a key brain structure of fear associative learning, represents fear memory during fear conditioning and subsequent extensive extinction, reconditioning. I found that the ensemble activity of LA neurons correlated tightly with conditioned fear responses of rats in the reconditioning paradigm. Further analysis revealed that among the LA neurons that displayed increased responses to the CS after fear conditioning, some exhibited weakened responses after extinction (extinction-sensitive), whereas others remained potentiated (extinction-resistant) after extinction. These results suggest the existence of distinct neuronal populations that encode various facets of fear memory and provide insights into the neuronal mechanisms underlying fear memory modulation. In the second chapter, I questioned whether the inhibitory network, which consists of the infralimbic cortex (IL) and the intercalated amygdala cells (ITC), is crucial for fear extinction, represents long-term correlates of fear learning that consisted of fear conditioning and extensive extinction. Single unit recordings and biochemical lesion techniques were employed to investigate the long-term effects of fear learning. I found that the CS-responses of IL neurons which emerged after single extinction dissipated with additional extinction. In keeping with this, I also found that ITC lesions that impaired the retrieval of extinction caused no deficit if lesions were made after multiple extinction sessions. These results suggest that single and extensive extinction involves different neural mechanisms. In summary, I investigated the long-term neural correlates of fear learning involving extensive extinction and reconditioning. First, LA neuronal population represented dynamic changes in CS-US association, while distinct sub-populations encoding various aspects of fear learning existed. Second, IL neurons and ITC activities were critical for single extinction, but not for extensive extinction. Together, these findings provide insights into the neural mechanisms underlying fear memory modulation and the treatment of fear-related mental disorders.