Environmental volatility amplifies Pavlovian control in decision-making

Every day, our brains act as constant calculators, weighing the likely results of our actions to decide what to do next. If you press a button and get a treat, you learn to press it again; if you press it and get nothing, you stop. This is how we navigate a world where cause and effect usually hold steady. But the real world is rarely that stable. Sometimes, the rules change without warning. A path that led to safety yesterday might lead to danger today. When the environment becomes unpredictable, the brain faces a difficult problem: the information it relies on to make smart choices becomes noisy and unreliable. In these moments of high uncertainty, our minds have to decide whether to keep trying to figure out the new rules or to fall back on a simpler, older system that reacts automatically to signals in the environment.

This question of how we switch between careful calculation and automatic reaction is at the heart of a new study by researchers at the University of Bologna and the National Research Council of Italy. They wanted to understand what happens when the world becomes chaotic. Specifically, they investigated whether a shaky, unpredictable environment makes us more likely to ignore our own calculations and instead follow the pull of familiar cues. Imagine a person walking through a forest where the trail markers keep moving; eventually, they might stop trying to read the map and simply follow the scent of a flower that usually leads to water, even if the map says the water is elsewhere. The researchers set out to see if this kind of shift happens in the human brain when the environment is volatile, and whether this shift changes the very nature of how we make choices.

To find the answer, the team designed a computer-based game that mimicked a slot machine. In the first part of the experiment, fifty healthy volunteers learned to play this game under two different conditions. In the "low volatility" condition, the machine was predictable: one lever gave a reward most of the time, and the other gave a reward rarely. The rules stayed the same for a long time. In the "high volatility" condition, the rules changed frequently and without warning. The lever that was usually good suddenly became bad, and the bad one became good, sometimes five times in a single session. The participants had to learn these shifting patterns just by watching the results of their choices. As expected, people found the changing rules much harder to master. Their choices were less accurate, and they seemed more unsure about which lever to pick.

After learning the rules of the slot machine, the participants moved to a second phase where they were shown pictures of food rewards. They learned that certain pictures predicted which food would appear. This was a passive learning stage where they simply watched the pictures and the outcomes, without pressing any levers. This created a strong link in their minds between a specific picture and a specific food. Finally, the researchers brought the two parts together. The participants returned to the slot machine, but this time, a picture appeared above the levers before they had to choose. The picture might match the lever that was currently the best choice, or it might point to the lever that was currently the worst choice. The researchers then watched to see if the participants would stick with the lever that had been paying off recently, or if they would be pulled toward the lever suggested by the picture.

The results revealed a clear and surprising pattern. When the participants had learned the game in the stable, low-volatility environment, the pictures had very little effect. They mostly ignored the pictures and stuck with the lever that was currently giving them the best rewards. However, when they had learned the game in the chaotic, high-volatility environment, the pictures took control. If a picture pointed to the lever that was currently the worst choice, the participants in the high-volatility group followed the picture far more often than those in the stable group. They abandoned the strategy that had worked for them in the past and let the cue drive their decision. This happened even though the picture was irrelevant to the current rules of the machine. The brain, overwhelmed by the constant changes, had shifted its trust from its own calculations to the simpler signal.

The researchers also looked at how consistent the participants' choices were. They measured the "entropy" of the decisions, which is a way of describing how much a person's behavior varies from moment to moment. In the high-volatility group, when the picture and the best choice disagreed, the participants' choices became much more erratic. They seemed to be torn between the two systems, flipping back and forth between following the picture and following the rules. This internal conflict created a noisy, unpredictable pattern of behavior. But the story changed in a second experiment designed to test what happens when the feedback stops. In this version, the researchers removed the reward feedback during certain parts of the test. Without the feedback to update their calculations, the instrumental system had nothing to work with. In this state, the participants didn't just become erratic; they became rigidly consistent. They followed the picture every single time, even when it led to the wrong choice. The conflict vanished because the brain had stopped trying to calculate and had fully surrendered to the cue.

These findings suggest that environmental volatility does not just add confusion or noise to our behavior. Instead, it actively changes how our brain weighs different types of information. When the world is stable, we rely on our ability to learn from our actions. When the world becomes unpredictable, our brain seems to decide that tracking every change is too costly and risky. Instead, it shifts control to the simpler, cue-based system that reacts to signals in the environment. This shift is not a failure of learning, but a strategic adaptation. The brain realizes that in a chaotic world, the old rules are gone, and the only stable thing left is the signal. This mechanism helps explain why people in unstable or stressful situations might become more impulsive or driven by immediate cues, a pattern seen in conditions like addiction and compulsive disorders. The study shows that when the map keeps changing, we stop reading it and start following the scent.