Repeated exposures to stress, and especially severe stress, are known to impair the brain’s ability to process thoughts and emotions, and can have a harmful impact on health. Among these impacts are the emergence of psychiatric conditions including depression, anxiety disorders, and (in cases of traumatic stress) PTSD.

For decades, BBRF grantees have been conducting research devoted to understanding how stress affects the structure and function of the brain. Recently, a team led by 2020 BBRF Young Investigator Luis E. Rosas-Vidal, M.D., Ph.D., of Northwestern University School of Medicine, reported important new findings on this subject based on experiments in rodents—whose brains are structurally and functionally similar to the human brain in how they respond to aversive experiences and are often used as a model in research. A co-author of the team’s paper, which appeared in Neurobiology of Stress, was Sachin Patel M.D., Ph.D., a 2015 BBRF Independent Investigator also at Northwestern.

The team provided a potentially clarifying explanation for a mysterious process called “representational drift” that they observed in real time while rodents in a lab setting were being subjected to a series of mild but unpleasant shocks to their feet. Called foot shocks, these are not unlike the static electricity shocks that humans experience.

The researchers were especially interested in observing a part of the rodent brain called the prelimbic prefrontal cortex (PL). The PL receives inputs that carry information about the internal and external environment from such regions of the brain as the amygdala and hippocampus, and serves to coordinate behavioral, physiological, and emotional responses to stress exposure.

A sophisticated real-time imaging system was used to observe at single-neuron resolution the activity of large populations of neurons in the PL while rodents were being subjected to trials in which multiple foot shocks were delivered in an unpredictable pattern, raising stress levels. The question was: how was this stress represented in the PL?

Analysis revealed that large numbers of neurons in the PL changed their activity levels during each session of multiple foot shocks. Particularly interesting was the observation that the “ensemble” of cells that were active during any specific foot shock—capturing the event, registering the discomfort—tended to change over time as additional foot shocks were experienced. There was relatively little overlap in these ensembles of activated neurons when one compared those activated in the beginning, middle, and near the end of a session.

The neural representation of the stressful experience, in other words, was not stable, but rather exhibited “representational drift” over short periods of time, within a session. The concept that different groups of neurons capturing a similar experience may spatially shift in the brain has been suggested previously. But what the team saw in the rodent brain “occurred on a much faster time scale—minutes rather than days—and remained persistently unstable across days,” they wrote.

This raised the question of why a “drift” in the representation of stress might be occurring. The team proposed that it may be part of the mechanism that enables the brain to construct a robust but flexible representation of the aversive experience which allows the individual to incorporate new information as needed. By incorporating new information into the representation, the brain can also “habituate” to repeated stress exposures—in other words, become accustomed to the experience as one way of coping with it.

This could be relevant to understanding stress-related psychiatric conditions. “Impairments in adaptation to stress,” the team noted, “are thought to lead to such disorders as depression, anxiety disorders, and PTSD.” These disorders have been associated with impaired cognitive flexibility. Stress adaptation is a form of such flexibility, “a dynamic process that requires broad, circuit-wide adaptations.”

The researchers postulate that high levels of rapid and persistent representational drift in the PL may help the brain make computations involved in associative learning. Shifting ensembles of neurons registered the stressful experience over time, meaning that a large number of cells in the PL ultimately took part in the process of learning to anticipate the shocks—they were part of the “sensory representation of the stressor.” As this drift eventually stabilized over time, the animals, the team proposes, may have been habituating to the repeated stress.

One implication of the research is that when habituation of this kind does not occur, or is not robust, the individual may then be vulnerable to developing depression, anxiety, or PTSD.