Imagine touching an object in total darkness. Then, a few minutes later, you recognize it effortlessly... just by looking at it. Without even thinking, you've just done something remarkable: you've connected two different senses—touch and sight—to understand the same reality.
But how does the brain achieve this feat? Is it a trick of the mind, or a more precise and localized mechanism? That's exactly the question a team from the University of Geneva (UNIGE) sought to answer. And what they discovered could change our understanding of intelligence—both human and artificial.
A hidden sensory GPS in the brain
The brain is capable of remarkable feats. For instance, it can learn with one sense (like sight) and reuse that knowledge with another (like touch). This ability of
sensory abstraction is a cornerstone of intelligence. Yet, we didn't know exactly where this happens.
The UNIGE team conducted experiments on mice to find out. The principle: teach them to differentiate between stimuli coming from above or below, first through touch using their whiskers. If they responded correctly, they received a reward. Within a week, they had mastered the task.
But the most fascinating part was yet to come.
When whiskers give way to light
Once the mice were trained, the researchers replaced tactile stimuli with visual signals—a shadow passing through the upper or lower part of their field of vision. And here's the surprise: without specific training, the rodents continued to respond correctly, as if the brain had automatically translated what it learned through touch into sight.
This mouse stands between red and blue lights representing the lower and upper parts of the surrounding space, where touch and vision converge.
© Sami El-Boustani
This was proof of sensory generalization: a kind of "instant translation" between senses.
A key player: the RL area
By closely observing the brain activity of these mice, the researchers discovered a very specific brain region: the RL (rostro-lateral) area, located in the upper cortex. This is where tactile and visual information converge.
When this area was deactivated, the mice lost their ability to generalize between senses, even though they could still learn well with one sense at a time. Even better: by stimulating it, the researchers could artificially trigger this ability.
"The RL area acts as a sensory translator. It helps us understand that what we feel with our whiskers in the dark is the same as what we later see in full light," explains Giulio Matteucci, co-author of the study.
Real-world implications
This discovery opens many doors. In medicine, it could help us better understand certain sensory disorders where the brain struggles to integrate information correctly. And in artificial intelligence, it offers an inspiring model: systems capable of
abstracting data to move from image to sound, or from text to video, just as our brain does naturally.
A beautiful illustration of how a small brain region can reveal immense potential for intelligence... and innovation.