Why humans and animals keep experimenting, even when there's no need

In a fast-paced world where change is constant, habitually experimenting and veering away from the known can be viewed as both clever and risky.

A fascinating new study reveals that this apparent curiosity is not exclusive to humans but is also shared with other animals.

The study from MIT suggests we have a built-in tendency to continuously tweak our methods to a task, despite having already mastered a successful approach – and, surprisingly, even when these alterations can lead to mistakes.

Why do we keep experimenting?

When defined roles and rules seem to be in place, why do we still feel the need to adapt, explore, and deviate from the norm?

The answer, according to Mriganka Sur, the lead author of the study at MIT, lies in two compelling reasons.

First, he says, simply because a task’s rules seem set one moment doesn’t mean they’ll stay that way in this uncertain world, so altering behavior from the optimal condition every so often could help reveal necessary adjustments.

Second, and of equal importance, continuous exploration could also offer a chance to discover something superior to our current best.

“If the goal is to maximize reward, you should never deviate once you have found the perfect solution, yet you keep exploring. Why? It’s like food. We all like certain foods, but we still keep trying different foods because you never know, there might be something you could discover,” noted the researchers.

The experiment: How we react to change

Former research technician Tudor Dragoi, now a graduate student at Boston University, led the study alongside fellow researchers in the Sur Lab. They investigated how humans and marmosets – a small primate – predict event timing.

Three humans and two marmosets participated in a straightforward task. They observed an image on a screen for a variable duration, within a set range, and had to respond by pressing a button as soon as the image disappeared.

The marmosets tapped a tablet, while the humans clicked a mouse. Success was determined by reacting as quickly as possible but without responding prematurely.

Marmosets were rewarded with juice for correct responses. Although marmosets required more training than humans, all participants eventually adopted a similar behavioral pattern.

The longer the image remained on the screen, the faster their reaction time was when it disappeared.

This aligns with the hazard model of prediction – if an image has a limited duration, the longer it stays visible, the more likely it is to disappear soon. With experience, reaction times improved overall.

Expectation versus experimenting

However, as the experiment progressed, the team observed an unexpected trend.

Mathematical modeling of the reaction time data revealed that both humans and marmosets adjusted their responses based on the previous trial’s outcome, despite already having learned the task.

If the image disappeared quickly in one round, participants slightly reduced their reaction time in the next, seemingly anticipating another short duration.

Conversely, if the image was visible for longer, they extended their reaction time, as if expecting another longer wait.

These findings build on a 2023 study from Sur’s lab, which revealed a similar pattern in mice. Even after mastering the rules of a different cognitive task, the mice occasionally deviated from the successful strategy.

As in this study, learning the optimal approach did not stop them from exploring alternatives – even at the expense of missing a reward.

“The persistence of behavioral changes even after task learning may reflect exploration as a strategy for seeking and setting on an optimal internal model of the environment,” concluded the researchers.

The link between learning and autism

The striking similarity between human and marmoset behaviors in the study carries profound implications for studies on autism spectrum disorders.

Autism is often associated with differences in making environmental predictions.

As marmosets show higher cognitive abilities than mice and possess inherent social traits, they are increasingly used in autism studies. Hence, establishing that marmosets can model neurotypical human behavior in prediction is a significant stride.

This revelation strengthens the argument that marmosets can indeed be valuable models for autism-related studies.

The study reiterates the astounding intricacies of learning and adaptation in both humans and animals. It not only sheds light on the flexibility of our learning processes but also paves the way for more comprehensive research into autism.

The full study was published in the journal Current Biology.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–