Primate Gaze Behavior Reflects Change of Mind in a Naturalistic Navigation Task

Every day, our brains make countless decisions, often while we are already in motion. We decide which path to take down a hallway, which turn to make while driving, or which item to grab from a shelf. These choices are rarely final; as new information arrives, we frequently change our minds, adjusting our course on the fly. Scientists have long studied how the brain makes decisions, but most experiments force people to sit still and stare at a fixed point, limiting their natural ability to look around and gather information. This creates a gap between what we know about decision-making in a lab and how we actually behave in the real world, where our eyes are constantly moving to explore our surroundings and update our understanding of where we are going.

A team of researchers at New York University and the Korea Brain Research Institute wanted to bridge this gap. They set out to see if the way our eyes move could reveal the hidden process of changing our minds. Specifically, they asked whether the precision of our gaze while navigating toward a goal could predict if we would stick with our original plan or switch to a new one when a better option appeared. By studying both humans and monkeys in a virtual world, they discovered that our eyes do more than just look at things; they act as a window into our internal beliefs, showing us exactly when and why we decide to change our course.

To investigate this, the researchers created a virtual environment where both humans and monkeys could navigate freely. The participants used a joystick to steer through a three-dimensional space that looked like a flat ground covered in blinking, triangular shapes. These shapes moved past them, creating a sense of speed and direction, but there were no fixed landmarks like trees or buildings to help them orient themselves. In this world, a brief "firefly"—a small, dark circle—would flash on the ground for a fraction of a second and then vanish. The task was to navigate to the spot where that firefly had been and stop there.

The experiment had two main variations. In some trials, two fireflies appeared at the same time, and the subject had to pick one and stick with it. In other trials, the first firefly appeared, and after a short delay, a second one flashed into view. This second setup was crucial because it allowed the researchers to see if the subject would change their mind. If the new firefly appeared in a more convenient spot, would the subject abandon their original path and steer toward the new one? The results showed that both humans and monkeys were very good at this task, often choosing the goal that required the least amount of effort, such as the closer target or the one that required less turning. However, when a new target appeared, they frequently changed their minds, showing a flexible ability to weigh the cost of their current path against the promise of a new option.

The most revealing part of the study came from watching the participants' eyes. The researchers found that the eyes did not just react to what was visible; they tracked the invisible goal the subject intended to reach. When a subject decided to stick with the first target, their gaze remained steadily fixed on that remembered location, even after the target had disappeared. But when a subject decided to change their mind and head toward the second target, their eyes shifted smoothly toward the new goal, often before their body or the joystick movement fully reflected the change. This suggested that the brain had already updated its plan, and the eyes were simply following the new internal direction.

Crucially, the researchers discovered that the quality of this eye tracking could predict a change of mind before it happened. If a subject's gaze wavered or drifted away from the first target while they were still navigating toward it, it signaled that their belief in that choice was shaky. This uncertainty made them much more likely to switch to the second target once it appeared. In contrast, when their gaze held steady and precise on the first target, they were far more likely to ignore the new option and finish their original journey. The eyes were essentially broadcasting the strength of their confidence: a steady gaze meant a firm decision, while a drifting gaze meant the mind was still open to change.

To confirm this, the team built a computer model that analyzed the eye movements and other data, such as how fast the subject was moving and which direction they were facing. They found that by looking only at the horizontal position of the eyes, they could predict which target the subject would eventually choose. This prediction worked even before the second target appeared, proving that the subtle movements of the eyes contained hidden information about the subject's future actions. The model showed that the eyes were not just a passive result of a decision but an active part of the planning process, carrying the latent variables of the brain's internal map.

The study also highlighted a striking similarity between humans and monkeys. Both species showed the same patterns of behavior, preferring closer and more central targets, and both used their eyes in the same way to signal when they were changing their minds. This suggests that the mechanism linking eye movements to decision-making is a fundamental part of how primates navigate an uncertain world. The findings challenge the old idea that eye movements are just a side effect of looking at something. Instead, they show that our eyes are deeply involved in the complex work of updating our beliefs and planning our next move.

In the end, this research offers a new way to understand how we make choices. It suggests that the act of looking is not separate from the act of deciding; the two are intertwined. When we look at a path, our eyes are testing our confidence in that path. If our gaze is steady, we are committed. If our gaze drifts, we are reconsidering. By watching where our eyes go and how they move, we can see the invisible process of changing our minds in real time, revealing that the path we take is often decided long before we take the first step.