New neuroscience study reveals sex-specific brain responses to threat

A new study published in Nature Neuroscience has found that male and female mice rely on different brain circuits to process threatening situations, even though they behave in much the same way. While both sexes learned to associate specific cues with danger and showed comparable defensive reactions, the brain pathways they used to make those associations were strikingly different. This finding challenges long-standing assumptions in neuroscience that similar behavior implies similar brain function, and it highlights the importance of including both sexes in brain research to ensure that results are broadly applicable.

The study, led by Rosemary Bagot at McGill University, was designed to address a persistent gap in neuroscience: the underrepresentation of females in experimental research. Historically, male animals have been used as the default subjects in many studies under the assumption that findings would apply universally. But mounting evidence suggests that males and females often exhibit different patterns of brain activity, especially when it comes to processing emotions and stress. Understanding these differences is critical for developing treatments for mental health conditions like anxiety and depression, which frequently affect women and men in different ways.

To investigate how the brains of male and female mice process threatening versus non-threatening signals, the researchers focused on two key brain pathways. Both pathways originate in brain regions known to regulate emotion and decision-making: the medial prefrontal cortex and the ventral hippocampus. These areas send information to a brain region called the nucleus accumbens, which plays a central role in integrating signals related to both reward and threat. Previous research has shown that these connections help balance fear and motivation, but how they function in females has remained largely unexplored.

The study used a variety of advanced techniques to examine brain activity in both sexes. In one part of the experiment, 17 mice (eight males and nine females) underwent a form of Pavlovian conditioning, where one cue (such as a tone or light) signaled an upcoming mild footshock, while another cue predicted no shock. Over time, mice learned to freeze—a common fear response—when the threat-predicting cue appeared. Researchers used fiber photometry to record real-time changes in calcium levels in specific brain cells, a proxy for neural activity. The recordings targeted pathways from the prefrontal cortex and hippocampus to the nucleus accumbens.

In parallel experiments, the researchers used a method called chemogenetics to selectively inhibit activity in these pathways. Mice received injections of designer receptors that could be activated by a drug, allowing researchers to temporarily shut down either the prefrontal-to-accumbens or hippocampus-to-accumbens connections. This allowed them to observe how silencing each pathway affected behavior in both male and female mice. A total of 90 mice participated in these chemogenetic experiments, with roughly equal numbers of males and females.

Across several days of conditioning, both sexes learned to distinguish between the threatening and neutral cues. They froze more during the threat cue and resumed normal behavior during the safe one. But when the researchers analyzed the brain recordings, they discovered that the pathways responsible for learning this discrimination differed by sex. In male mice, activity in the pathway connecting the hippocampus to the nucleus accumbens was most involved in distinguishing between threat and safety. In female mice, however, the prefrontal cortex–to–accumbens pathway carried this information instead.

To confirm this, the team used a machine learning classifier trained to predict whether a mouse was hearing the threat or safe cue based on its brain activity. In males, only activity from the hippocampal pathway predicted the correct cue, while in females, only the prefrontal pathway did. This showed that the two sexes relied on distinct brain circuits to solve the same learning problem.

Interestingly, these neural differences did not correspond to major differences in basic freezing behavior, which remained similar between the sexes. But when the researchers added a reward-based task, new sex differences emerged. Mice were trained to press a lever to receive a sweet treat. Then, during testing, the threat and safety cues were reintroduced. Mice were expected to reduce their lever pressing during the threat cue but continue pressing during the safe cue.

Here, the researchers found that disabling the hippocampal pathway disrupted the males’ ability to suppress reward-seeking during threat cues, while disabling the prefrontal pathway did the same in females. Not only did this impair their discrimination between threat and safety, but it also revealed that the circuits governing how threat cues affect motivated behavior are not the same in males and females. In female mice, shutting down the prefrontal pathway led to more generalized fear, with some mice failing to press the lever at all. But in males, shutting down this pathway had no such effect, suggesting they did not rely on it for discriminating threat from safety in the same context.

To determine whether these behavioral effects were caused by differences in brain wiring, the researchers examined how the prefrontal and hippocampal pathways connect to neurons in the nucleus accumbens in both sexes. They used optogenetics and brain slice electrophysiology to measure how strongly these inputs excited or inhibited their target neurons. The results showed no significant differences in the strength or balance of these connections between males and females. This suggests that the observed sex differences arise not from structural differences in the brain, but from how similarly connected circuits are recruited during learning.

Another surprising finding involved the synchrony between the two pathways. By analyzing the timing of neural activity in both circuits, the researchers found that threat cues tended to reduce synchrony between the hippocampal and prefrontal inputs to the nucleus accumbens. This effect was more pronounced and lasted longer in females. Conversely, safety cues increased synchrony in females, but not in males. These patterns suggest that female brains may be more attuned to detecting and signaling safety, perhaps as a behavioral strategy to avoid excessive threat generalization.

This research highlights that even when behavior appears the same, the brain may be working in very different ways depending on sex. These findings could help explain why some psychiatric disorders, such as anxiety and depression, show sex-specific patterns in how they develop and how they respond to treatment. Because the pathways involved in this study are also sensitive to chronic stress, the work may also inform future research into how long-term stress differentially affects men and women at the neural level.

The study does have some limitations. It focused only on mice, so it’s unclear how directly the results apply to humans. While rodents are a useful model for basic neuroscience, human brains are far more complex, and social and hormonal influences differ significantly. The sample sizes were also relatively small, which is typical for studies involving invasive neural recording, but it limits generalizability. Finally, the study examined only two brain pathways; other circuits may also contribute to threat processing in sex-specific ways.

Future research could explore how these pathways interact with hormonal changes across the lifespan, such as during puberty or after pregnancy, and whether similar sex-specific circuit recruitment occurs in humans. Longitudinal studies might also reveal whether stress or trauma alters how these circuits function over time.

The study, “Sex-biased neural encoding of threat discrimination in nucleus accumbens afferents drives suppression of reward behavior,” was authored by Jessie Muir, Eshaan S. Iyer, Yiu-Chung Tse, Julian Sorensen, Serena Wu, Rand S. Eid, Vedrana Cvetkovska, Karen Wassef, Sarah Gostlin, Peter Vitaro, Nick J. Spencer, and Rosemary C. Bagot.