How a handshake could help diagnose autism

To you, simple actions such as picking up a coffee mug might be effortless and almost insignificant; your hand opens, your fingers adjust, and you lift it without a thought. But beneath the surface, your brain is running a highly choreographed show: coordinating muscles, timing movements, and predicting the mug’s weight and shape. Now imagine if those tiny, almost invisible details in how you grasp an object could reveal something as complex as autism.

That’s exactly what a study published in Autism Research suggests. The work comes from a team led by Dr. Martin Freud and colleagues. The researchers addressed a question with a strong societal impact. Could we detect autism not by observing social interactions or communication patterns, as is standard today, but by analyzing how someone moves their hand when they pick up an object?

It sounds almost unbelievable, but their findings point to a powerful new way to classify autism using nothing more than motion data.

Autism spectrum disorder (ASD) affects how people perceive the world and interact with others. Current diagnostic methods rely heavily on behavioral observations and structured interviews. A specialist might watch how a child communicates, plays, or responds to questions.

While these approaches are effective, they have big challenges. One of them is subjectivity, as diagnosis depends on human judgment. Also, it often takes hours of testing and months of waiting. And finally, many children don’t receive a diagnosis until years after the first signs appear.

Researchers around the world are searching for faster, more objective ways to identify autism, ideally using measurable biological or physical markers. That’s where this study tries to leap.

At first glance, hand movements may seem far removed from autism. But motor differences have long been noted in autistic individuals. Children on the spectrum sometimes show unique movement patterns, slightly different posture, coordination, or timing in their actions.

Dr. Freud and his colleagues honed in on grasping kinematics, the detailed timing and shape of how fingers and hands move when reaching for and picking up objects. Think of it as the “fingerprint” of your hand in motion. These patterns are so subtle that the human eye usually can’t detect them, but with motion sensors and machine learning, they can be measured precisely.

The team suspected that autistic individuals might show characteristic signatures in their grasping movements, and that these signatures could be strong enough to classify autism accurately.

Inside the experiment

Participants, both autistic individuals and neurotypical controls, were asked to perform simple grasping tasks. Imagine being in a lab, reaching out to grab different objects while small motion sensors or cameras track every millisecond of your hand’s trajectory.

The researchers didn’t want to analyze whether the grasp “looked normal.” Instead, they studied the kinematics, including how quickly the hand opened, the exact angle of the fingers, the timing between movement phases, and how the hand adjusted right before contact.

Once this treasure trove of data was collected, they turned to artificial intelligence. Machine learning models were trained to spot patterns invisible to humans, tiny differences that consistently separated autistic from non-autistic participants.

The results? Classification with a never-before-seen accuracy.

Simple, but breakthrough results

What makes this work so exciting is its elegance. Instead of relying on long interviews or invasive tests, the method just requires a person to reach for objects while their hand is tracked.

The study found that autism classification based on grasping movements was not only possible but also surprisingly effective. These motor signatures appear to carry reliable information about neurodevelopmental differences.

That means, in the future, a simple test using motion capture or even consumer devices like smartphones could provide an early, objective screening tool for autism.

The results of this study could really change lives. In fact, the earlier autism is identified, the sooner children can access therapies and support that help them thrive. Moreover, access to expert clinicians is uneven across regions. Therefore, a simple, movement-based test could democratize autism screening worldwide. Also, instead of focusing solely on social “deficits,” this approach emphasizes measurable, biological differences, helping to reframe autism in more neutral, scientific terms. And finally, the study offers new windows into the brain. Ultimately, understanding how movement relates to autism may shed light on the broader neurological foundations of the condition.

The idea that something as ordinary as picking up a glass of water could reveal deep truths about how the brain processes the world could be a game-changer for healthcare.

If you want to learn more, read the original article titled "Effective autism classification through grasping kinematics" on Autism Research at http://dx.doi.org/10.1002/AUR.70049.