Editors' ChoiceNeuroscience

Switching tracks in fear memories

See allHide authors and affiliations

Science Translational Medicine  15 Feb 2017:
Vol. 9, Issue 377, eaam6062
DOI: 10.1126/scitranslmed.aam6062


The brain circuits underlying fear memories can be reorganized by updating old memories with new information.

Abnormalities surrounding the encoding, retrieval, updating, and extinction of fear-related memories are thought to be core components of clinical anxiety disorders. The best current treatments for anxiety disorders are extinction-related psychotherapies, but unfortunately many patients do not respond adequately to treatment. One aspect of improving outcome in patients is better understanding how new fear-relevant information, such as might occur in psychotherapy, influences how the brain handles fear memories. Stable stored memories can become labile when retrieved in the context of new information, but can that new information reorganize how the original memory itself is stored in the brain? In principle, this could lead to very different brain signatures in patients and differences in treatment outcome.

Kwapis and colleagues addressed this question, working with a well-characterized experimental system for studying fear memories in rodents, wherein the animal learns that a previously neutral tone predicts an aversive footshock, after which the animal freezes when it hears the tone. Their experiment took advantage of the distinction between two types of fear conditioning, called trace and delay conditioning. These forms of learning differ with respect to whether the tone and shock are coincident or separated in time, which leads to reliance on different brain regions. After training in one type of conditioning combined with pharmacologic disruption of new memory formation in specific brain regions, the animals updated their memories with a single trial of the other type of conditioning. The amygdala, which is usually only required for extinction of delay conditioning, became necessary when trace conditioning was updated with a single trial of delay conditioning. Likewise, the retrosplenial cortex, only involved in retrieval of trace conditioning, became superfluous when trace conditioning was updated with delay conditioning. Extending these findings to clinical populations, the authors suggest that getting memories to “switch tracks” to neural pathways more amenable to intervention, through built-in memory updating processes, may facilitate treatment.

Highlighted Article

Stay Connected to Science Translational Medicine

Navigate This Article