Threat extinction is the decline in responding to a stimulus that previously signaled danger, following repeated non-consequential exposures. The amygdala, prefrontal cortex, and hippocampus are known contributors to the neural circuits driving threat extinction, however, research has not yet resolved the precise neuronal interactions necessary for extinction success. At times, research even appears to present conflicting conclusions concerning the neurocircuitry that drives extinction learning. Here, we propose a model of threat extinction that accounts for the range of findings within the literature. Our model consists of two parallel pathways that differ in their temporal dynamics, such that a fast extinction circuit requires only one training session and exerts its effects rapidly, while a slow circuit impacts behavior only after multiple training sessions. This dual-mechanism of threat extinction circuits is rooted in a hybrid associative learning model whereby parallel fast and slow pathways differentially rely on new learning and unlearning, respectively, to optimize survival as an animal's probabilistic beliefs concerning threat and danger evolve over repeated extinction trials.

Original languageEnglish
Pages (from-to)96-103
Number of pages8
JournalCurrent Opinion in Behavioral Sciences
StatePublished - Dec 2018


Dive into the research topics of 'Fast and slow extinction pathways in defensive survival circuits'. Together they form a unique fingerprint.

Cite this