ConFER: A Neurally Constrained Computational Model of Context-Dependent Fear Extinction Recall and Relapse

Shreya K. Rajagopal,Thad A. Polk
2024-11-13
Abstract:Exposure therapy, a standard treatment for anxiety disorders, relies on fear extinction. However, extinction recall is often limited to the spatial and temporal context in which it is learned leading to fear relapse in novel contexts or after delays. While animal studies provide valuable insights into the mechanisms underlying extinction recall, their invasive nature limits direct applicability to human research. Computational models that incorporate findings from animal research and generate testable hypotheses for human studies can bridge this gap. Current models, however, often focus either on detailed neuron-level activity, limiting their scope, or simulate behavior by abstracting away from neural mechanisms. Moreover, they tend not to disentangle the distinct roles of cue and context information in fear extinction and recall. To address these limitations, we present the Context-Dependent Fear Extinction Recall Model (ConFER), a systems-level, neurally constrained model of fear extinction and recall, and fear relapse. ConFER integrates findings from the neural fear circuit, including distinct pathways for processing cue and context information and valence-based stimulus-responsive regions in the basolateral amygdala (BLA) that encode extinction and fear memories. These pathways independently activate positive and/or negative BLA representations based on learned associations, which compete to produce the final fear response. ConFER simulates fear renewal across various contexts and spontaneous recovery after delays, while also generating novel, testable predictions. Notably, it predicts that counterconditioning may confer greater resilience than extinction in preventing relapse in new contexts or after delays. By providing a mechanistic view of the relapse of fear, ConFER offers potential insights to enhance clinical interventions by improving exposure therapy outcomes.
Neurons and Cognition
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the limitations of fear - extinction recall in different environments and how to enhance the understanding of this process through a computationally - modeled neural constraint, thereby improving clinical interventions. Specifically, the paper focuses on the following points: 1. **Environmental Dependence of Fear Extinction**: Exposure therapy (a standard method for treating anxiety disorders) relies on fear extinction, but this extinction is usually limited to specific temporal and spatial environments. When an individual is in a new environment or after a long - time delay, fear may recur. This limits the effectiveness of exposure therapy. 2. **Limitations of Existing Models**: Current computational models either simulate the activities of individual neurons in too much detail, resulting in a limited scope of the model; or they abstractly simulate behavior, ignoring neural mechanisms. In addition, these models often fail to distinguish the different roles of cue information and context information in fear extinction and recall. 3. **Integrating Findings from Animal Studies**: Animal studies have shown that the mechanisms of fear extinction involve multiple neural pathways, especially different pathways for processing cue information and context information. However, due to the invasiveness of these experiments, it is difficult to directly apply them to human studies. To solve these problems, the paper proposes a computational model named ConFER (Context - Dependent Fear Extinction Recall Model). ConFER is a system - level, neurally - constrained model, aiming to integrate the findings from animal studies and generate testable hypotheses to improve the accuracy of human studies. The main features of ConFER include: - **Distinguishing Cue and Context Information**: The model contains different neural pathways that process cue information and context information respectively. - **Competition Mechanism**: The model simulates the positive and negative emotional representations in the basolateral amygdala (BLA) through a competition mechanism and ultimately determines the intensity of the fear response. - **Simulating Fear - Recovery Phenomena**: ConFER can simulate fear - recovery phenomena in different environments, such as fear renewal and spontaneous recovery. Through these features, ConFER can not only explain the environmental dependence of fear extinction but also generate new, testable predictions, providing potential insights for improving exposure therapy. For example, the model predicts that counter - conditioning may be more effective than simple fear extinction in preventing recurrence in new environments or after a long - time delay.