Feasibility of multimodal behavioral, gaze, and EEG measurement of natural scene segmentation in autism
This paper introduces and validates a novel multimodal framework combining EEG, eye-tracking, and behavioral measures to investigate natural scene segmentation in individuals with Autism Spectrum Disorder, demonstrating its feasibility for capturing dynamic, real-world perceptual processing differences between neurotypical and neurodivergent populations.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
The human brain is constantly at work sorting the visual world into manageable pieces. When you look at a busy street, you do not see a chaotic blur of pixels; you see a car, a tree, a person, and the sky as distinct objects. This mental process of cutting a complex scene into meaningful parts is called segmentation. It is the foundation of how we understand our surroundings. For most people, this happens automatically and instantly. However, for many individuals with Autism Spectrum Disorder, a condition defined by differences in social communication and repetitive behaviors, the way the brain handles sensory information can be different. Research has long suggested that people with autism might process visual details in unique ways, sometimes noticing small patterns that others miss or struggling to see the "big picture" of a scene. Understanding exactly how this happens is crucial, because these sensory differences can shape daily life, from how someone navigates a room to how they interact with others. Yet, most previous studies have relied on simple, artificial pictures that do not capture the messy, rich complexity of the real world.
A new study by researchers at Albert Einstein College of Medicine seeks to bridge this gap by observing how people with and without autism segment real-world scenes. The team designed a novel experiment that combined three different ways of measuring the brain and behavior: asking people to make decisions about what they see, tracking where their eyes look, and recording the brain's electrical activity. They recruited twenty-four neurotypical adults and thirteen adults with autism, all aged sixteen or older. In the lab, these participants sat in front of a screen and viewed natural photographs, such as landscapes with rocks and water, as well as synthetic images made of textured patterns. Before each image appeared, the participants were told to mentally divide the picture into a specific number of parts, like two or three distinct segments.
Once the image appeared, the participants were briefly shown two red circles at random spots on the picture. They had to decide instantly whether the two circled areas belonged to the same mental segment or to different ones. By repeating this hundreds of times with different pairs of circles, the researchers could reconstruct a detailed map of how each person mentally sliced the image. This method allowed the team to see not just what a person saw, but how certain they were about their choices. The study also recorded eye movements to see where participants looked while they formed these mental maps, and used electroencephalography, a technique that measures electrical signals on the scalp, to see how the brain reacted in real-time as the images were processed.
The results showed that people with autism were fully capable of performing the task. They successfully created mental maps of the scenes that were just as clear and organized as those made by neurotypical participants. When looking at the final maps, the researchers found that both groups generally agreed on where the boundaries of objects were, whether the image was a natural scene or a simple texture. However, subtle differences began to emerge when the researchers looked closer at the timing and the uncertainty of the decisions. People with autism took slightly longer to answer the questions, with a median reaction time of 582 milliseconds compared to 491 milliseconds for the neurotypical group. This delay was consistent across both natural images and textures, suggesting a difference in the speed of decision-making rather than a difficulty with the task itself.
The study also revealed differences in how the two groups explored the images with their eyes. On average, participants with autism looked at a wider area of the image, covering about 38.5 percent of the picture, whereas neurotypical participants focused on about 31.5 percent. This broader gaze happened almost immediately, within the first second of viewing, indicating that the difference in attention was not due to a lack of focus later in the trial but was a fundamental part of how they approached the scene. When the researchers analyzed the brain activity, they found that the neurotypical group showed stronger electrical signals in the back of the brain, the area responsible for processing vision. These signals were larger in magnitude, but the timing of the brain's peak response was the same for both groups. This suggests that while the brain's visual centers were working just as fast, the intensity of the activity differed.
Importantly, the researchers ruled out the idea that the slower reaction times in the autism group were caused by losing focus or getting tired during the long experiment. Both groups actually got faster as they went through the trials, showing that they remained engaged throughout. The study also found that while the overall level of uncertainty in making decisions was similar between the groups, the specific patterns of where people felt unsure were more varied among those with autism. This hints that while the final result of seeing the scene might look similar, the internal process of getting there involves different strategies and levels of confidence.
This work does not claim to have solved the mystery of autism or to have found a single cause for these differences. Instead, it establishes a new, reliable way to study how the brain organizes the visual world in real-time. By using natural images alongside controlled textures, the researchers created a method that captures the complexity of everyday life while still allowing for precise measurement. The findings suggest that the differences in autism are not necessarily about an inability to see the world correctly, but rather about how the brain prioritizes information, allocates attention, and processes the speed of decisions. The study provides a foundation for future research to explore how these sensory differences might connect to the social and communication challenges often associated with autism, offering a clearer, more nuanced view of the autistic experience.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.