For as long as humans have walked the earth, survival has depended on a delicate balance between two opposing strategies: staying with the familiar or venturing into the unknown. Early humans knew the safety of reliable hunting grounds or dependable foraging sites. These familiar places provided a sense of stability and security. Yet clinging too tightly to them carried risks: resources could dry up, the environment could shift, or competitors might overrun those same grounds.
On the other hand, exploration promised opportunity but came with its own dangers. A new valley might offer plentiful game, or it might be barren. Fresh berries could provide nourishment, or they could prove poisonous. Even in modern life, the same tension lingers. We return to our favorite restaurant for comfort, but sometimes we take the leap and try a new place, hoping for a discovery that delights—or bracing for disappointment.
Psychologists and neuroscientists call this constant tug-of-war the exploration-exploitation dilemma. It reflects a fundamental challenge of decision-making: should we exploit what we already know, or explore in search of something better?
Shifting the Lens: From Gains to Losses
Until recently, much research on this dilemma focused on how we make decisions in the context of potential gains. When there’s a chance to win, people tend to explore more when they feel uncertain about which choice will bring the best outcome. Brain imaging studies have shown that both surface areas like the cortex and deeper regions such as the amygdala light up during this process.
But avoiding losses is not the same as pursuing rewards. Strategies driven by loss aversion are often linked to psychological conditions like anxiety, post-traumatic stress disorder, and mood disorders. The way the brain processes potential losses has unique implications, and researchers have long wondered whether the mechanisms behind exploration might differ depending on whether the stakes involve winning or avoiding losing.
A Window Into the Living Brain
A groundbreaking study at the Weizmann Institute of Science set out to answer this question. Published in Nature under the title “Rate and noise in human amygdala drive increased exploration in aversive learning”, the research offered a rare look into the activity of single neurons in the human brain.
Seventeen patients with epilepsy, already undergoing treatment that required electrodes implanted deep in their brains, volunteered to take part. This gave researchers an extraordinary opportunity to monitor the firing of 382 neurons across 22 sessions, primarily in the amygdala and the nearby temporal cortex.
The patients played a simple game. Each round began with a tone that indicated whether it was a gain trial or a loss trial. Two geometric shapes then appeared, each with a different probability of outcome—one offering a 70% chance of success, the other just 30%. In gain trials, participants could earn 10 points or nothing. In loss trials, they risked losing 5 points or nothing.
Behind this simple design lay a profound question: how does the brain’s activity shift when we try to avoid losses instead of chasing gains?
Firing, Noise, and the Nature of Exploration
The scientists measured not only how often neurons fired just before each decision but also how variable that firing was from trial to trial. This variability, which they called “neural noise,” turned out to be crucial.
The results revealed that participants explored more in loss trials than in gain trials. Even after they had learned which option was statistically better, people remained more willing to deviate from the safe choice when losses were at stake. Their accuracy dropped more sharply in loss contexts than in gain contexts, suggesting a persistent pull toward exploration when the possibility of losing loomed.
Two key neural signals explained this behavior. First, there was a valence-independent rise in firing rates in both the amygdala and the temporal cortex before exploratory choices. This signal appeared in both gain and loss contexts, suggesting it acts as a general marker of exploration.
Second—and most striking—was a surge of noise in the amygdala during loss trials. This extra variability correlated with higher uncertainty and a stronger tendency to explore. As participants learned more and uncertainty declined, the noise decreased. When researchers built computational models of decision-making and plugged in these noise patterns, the models reproduced the human tendency to explore more aggressively under threat of loss. Importantly, this behavior could not be explained by loss aversion alone.
Why the Brain Pushes Us to Explore Under Threat
From an evolutionary perspective, the findings make sense. If a familiar food source is no longer reliable, survival may depend on quickly testing alternatives, even if they come with risks. Under threat of loss, sticking too rigidly to the old and familiar could mean starvation. Neural variability—the “noise” in the amygdala—may be the brain’s way of tilting the balance toward exploration when the cost of staying put could be catastrophic.
This mechanism also resonates with everyday life. Imagine that your favorite restaurant closes unexpectedly. In that moment, exploration becomes more urgent. You might try several new places, some disappointing, some surprising. The closure of the safe option shifts your mindset, making you more willing to tolerate uncertainty for the chance of discovering something new.
Implications for Mental Health and Decision-Making
The study also sheds light on why negative experiences can push people toward maladaptive exploration. In conditions like anxiety or PTSD, the brain’s threat systems are often hyperactive. If amygdala noise drives exploration under loss, an overactive or unstable system could lead to decision-making patterns that feel erratic, restless, or self-defeating.
By identifying the neural signals that distinguish gain-driven and loss-driven exploration, researchers open new doors for understanding—and eventually treating—these conditions. If scientists can learn to modulate the balance of firing and noise in the amygdala, they may one day help people make choices that feel less chaotic and more adaptive.
A Glimpse Into the Future of Neuroscience
This research is part of a broader shift in neuroscience toward studying the brain not as a static organ but as a dynamic decision-making system, finely tuned by evolution to balance risk and reward. The use of single-neuron recordings in humans represents a leap forward, allowing scientists to go beyond broad brain imaging to watch the very cells that shape behavior.
Yet the work also raises new questions. Can neural noise be directly manipulated, and if so, does that change exploration behavior? How do other brain regions, like the prefrontal cortex, interact with the amygdala in this process? Could therapies someday harness or stabilize this neural variability to improve mental health outcomes?
The Endless Dance Between Safety and Discovery
At its core, this study reminds us that the human mind is not simply a calculator of probabilities. It is a survival machine shaped by millions of years of evolution. When faced with potential loss, our brains nudge us toward exploration—even if it means stepping into the unknown.
This mechanism, born in the savannas and forests of our ancestors, still guides us today in ways large and small. It influences the risks we take in business, relationships, and personal growth. It is why the closing of one door often compels us to search for another.
Exploration and exploitation remain the twin poles of survival. Physics may describe the forces that move the stars, but neuroscience reveals the forces that move us—the sparks, the noise, the subtle rhythms of neurons firing as we navigate a world of both promise and peril.
More information: Tamar Reitich-Stolero et al, Rate and noise in human amygdala drive increased exploration in aversive learning, Nature (2025). DOI: 10.1038/s41586-025-09466-1
