🧬 biology

Autistic Traits and Social-Context-Dependent Prefrontal Hemodynamic Responses: An fNIRS Gameplay Study

This study demonstrates that higher subclinical autistic traits are associated with attenuated prefrontal hemodynamic responses and reduced neural differentiation between social and non-social contexts during live competitive gameplay, suggesting that dimensional autistic variation specifically impacts brain activity during reciprocal human interaction.

Original authors: Mohammad Ghalavand, Javad Hatami, Fatimah Nosrati

Published 2026-08-26

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mohammad Ghalavand, Javad Hatami, Fatimah Nosrati

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 ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed.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

Human beings are not isolated thinkers; our minds are shaped by the people around us. When we interact with another person, our brains do not simply watch a neutral object move; they try to guess what that person is thinking, what they want, and what they might do next. This constant, invisible work of predicting another person's intentions is a core part of how we navigate the world. Scientists have long studied this by showing people pictures of faces or videos of social scenes, but real life is rarely that static. In the real world, we are active participants, constantly adjusting our own actions based on how others respond to us. A newer approach in science, sometimes called second-person neuroscience, argues that to truly understand social thinking, researchers must study people while they are actually interacting with one another, not just watching from the sidelines. This shift from observing to participating is crucial for understanding how our brains handle the complexity of live human connection.

A team of researchers at the University of Tehran wanted to see how this live interaction affects the brain, specifically looking at people who have varying levels of autistic traits. Autism is often discussed as a diagnosis, but many scientists believe these traits exist on a spectrum, meaning everyone has some degree of them, from very low to very high. The researchers used a standard questionnaire called the Autism-Spectrum Quotient to measure these traits in a group of forty-three young men. They were not looking for a diagnosis of autism, but rather for natural differences in how these individuals process social information. The central question was whether these differences show up in the brain's activity when a person is playing a game against a computer versus playing against another human being.

To find the answer, the researchers designed a virtual tennis game. The participants played two different versions of the game. In the first version, they played against a computer program. In the second version, they played against another real person sitting in the same room. While they played, the researchers measured blood flow in the front part of the brain, an area known as the prefrontal cortex, which is heavily involved in planning, decision-making, and controlling our actions. They used a special light-based scanner that sits on the head like a cap, allowing them to see brain activity without the need for a large, noisy machine. This method is particularly useful for social studies because it lets people move and interact more naturally than traditional brain scanners.

The results revealed a clear pattern that depended entirely on who the player was facing. When the participants played against the computer, the level of their autistic traits did not seem to change how their brains responded to the game. Their brain activity looked much the same regardless of their scores on the questionnaire. However, the situation changed dramatically when they played against a human opponent. In this live, social condition, the researchers found a strong link between a person's autistic traits and their brain activity. Specifically, individuals with higher scores on the autistic traits questionnaire showed a noticeably weaker response in a specific area on the left side of their prefrontal cortex when playing against a human. This area is known to help us control our impulses, make quick decisions, and adjust our behavior based on what others are doing.

The study suggests that for people with higher levels of autistic traits, the brain does not ramp up its activity in the same way when facing a real human opponent compared to when facing a computer. It is as if the brain treats the human opponent differently, perhaps not engaging the same level of predictive effort or social monitoring that it does for other people. The researchers also found that the difference in brain activity between playing against a human and playing against a computer was smaller for those with higher autistic traits. In other words, their brains did not distinguish as sharply between the two types of gameplay as the brains of those with lower autistic traits did. This finding supports the idea that the way these individuals process social interaction is fundamentally different, not just in how they behave, but in how their brains physically respond to the presence of another person.

It is important to note that this study looked at natural variations in a group of healthy university students, not at people with a clinical diagnosis of autism. The researchers are careful to say that their findings do not mean that a specific part of the brain is "broken" or that these traits are a disorder. Instead, the results show that autistic traits are linked to how the brain handles the demands of live social competition. The study also highlights the value of using games and real-time interaction to understand the brain, moving beyond static pictures to see how we think while we do. While the researchers acknowledge that their study has limitations, such as the small number of participants and the focus on young men, the results offer a compelling glimpse into how the brain adapts to the presence of another mind. They suggest that the difference between playing against a machine and playing against a person is not just a matter of rules or strategy, but a deep, biological shift in how our brains prepare for the unpredictable nature of human connection.

Technical Summary: Autistic Traits and Social-Context-Dependent Prefrontal Hemodynamic Responses

Problem Statement
Real-world social cognition involves continuously predicting an agent's behavior and adapting one's own actions during unfolding interactions. While traditional social neuroscience has relied on passive observation paradigms (e.g., viewing faces or animations), "second-person neuroscience" argues that social cognition must be studied through direct engagement with another agent. In dyadic interactions, action selection becomes recursively structured; individuals must anticipate not only an opponent's actions but also how the opponent interprets their own actions.

Autism research has traditionally focused on categorical differences between diagnosed and non-diagnosed groups. However, a dimensional approach suggests that autism-related traits vary continuously across the general population, measurable via the Autism-Spectrum Quotient (AQ). A critical gap exists in understanding how these dimensional autistic traits influence neural responses during competitive, action-based social interactions. Specifically, it is unclear whether AQ scores predict task-evoked prefrontal hemodynamic responses when the cognitive field shifts from individual gameplay to reciprocal human interaction.

Methodology
The study employed a within-subjects design using functional near-infrared spectroscopy (fNIRS) to measure prefrontal hemodynamic responses in 43 adult male university students.

Participants: The final analytic sample consisted of 43 right-handed males (mean age 23.42 years) who completed the Autism-Spectrum Quotient (AQ) questionnaire. AQ scores ranged from 13 to 42 (mean 22.72), allowing for dimensional analysis rather than diagnostic grouping.
Experimental Task: Participants performed a virtual tennis task under two conditions:
1. Single-player: Against a computer-controlled opponent (less socially interactive).
2. Dual-player: Against a human opponent in real-time (socially interactive).
  The task utilized a block design (10 cycles of 60s task / 30s rest). To ensure comparable effort, rewards were determined by a coin flip after the session, masking which condition would count toward the final score.
Data Acquisition: Prefrontal activity was recorded using a six-channel continuous-wave fNIRS system (OxyMon) operating at 740 nm and 860 nm wavelengths. Optodes were positioned according to the 10–20 system, covering fronto-medial and lateral prefrontal areas (channels near FPz, F7, and F8).
Preprocessing: Raw signals were band-pass filtered (0.01–0.09 Hz) to isolate task-related hemodynamic fluctuations. Motion artifacts were corrected via spline interpolation, and baseline correction was applied.
Statistical Analysis: Task-evoked responses were estimated using a General Linear Model (GLM) with a canonical hemodynamic response function (HRF). Beta coefficients were extracted for oxygenated (HbO) and deoxygenated (HbR) hemoglobin across all channels.
- Correlation analyses (Pearson) were performed between AQ scores and beta estimates for single-player, dual-player, and dual–single contrast conditions.
- Multiple comparisons were controlled using Bonferroni correction across 12 signals (6 channels × 2 hemoglobin types) per condition.
- Robustness was assessed via bootstrap resampling (95% CI) and permutation testing.

Key Results
The study found that autistic traits were associated with prefrontal hemodynamic responses selectively during reciprocal human interaction, not during isolated gameplay.

Dual-Player Condition (Social Interaction):
- Channel 6-HbR: A robust negative correlation was observed between AQ scores and task-evoked deoxygenated hemoglobin (HbR) responses in Channel 6 (located approximately over the left lateral prefrontal region near F7). Higher AQ scores predicted attenuated HbR responses ( $r = -0.549$ , $p_{Bonf} = 0.0017$ ). This finding survived Bonferroni correction, bootstrap resampling (95% CI: $[-0.696, -0.392]$ ), and permutation testing ( $p_{perm} = 0.0003$ ).
- Channel 1-HbO: An exploratory negative correlation was found in Channel 1 (anterior prefrontal region) for HbO ( $r = -0.423$ ), which was significant in resampling analyses but narrowly missed Bonferroni correction ( $p_{Bonf} = 0.056$ ).
Single-Player Condition (Non-Social):
- No significant correlations survived Bonferroni correction between AQ scores and hemodynamic responses in any channel. The strongest uncorrected association (Channel 1-HbR) was deemed weak and exploratory. Notably, Channel 6-HbR showed no association in this condition ( $r = 0.038$ ).
Dual–Single Contrast:
- The difference between dual-player and single-player responses (dual–single contrast) in Channel 6-HbR showed a significant negative correlation with AQ scores ( $r = -0.440$ , $p_{Bonf} = 0.038$ ). This indicates that higher AQ scores predicted a reduced hemodynamic differentiation between social and less-social gameplay contexts.
Robustness:
- A post-hoc sensitivity analysis excluding participants with extreme AQ scores (below 5th or above 95th percentile) confirmed that the primary Channel 6-HbR findings remained robust.

Significance and Claims
The paper claims that the study provides preliminary evidence that dimensional variation in autistic traits is reflected in prefrontal hemodynamic responses specifically during reciprocal human interaction, rather than during general task performance.

Context-Dependence: The findings support the "second-person neuroscience" perspective, suggesting that the neural mechanisms engaged during live interaction differ from those in passive or isolated contexts. The association between AQ and brain activity emerged only when participants had to predict and respond to an intentional human agent.
Prefrontal Mechanisms: The attenuation of HbR responses in the left lateral prefrontal region (Channel 6) among individuals with higher AQ scores suggests differences in cognitive control, response selection, or the flexible updating of actions required for competitive social interaction.
Methodological Contribution: The study demonstrates that interactive fNIRS paradigms, particularly within a hyperscanning framework, can detect dimensional variability in social prediction and adaptive action updating in the general population.
Modesty of Claims: The authors explicitly state that these findings should not be interpreted as identifying a specific neural marker for autism or a diagnostic tool. They emphasize that fNIRS signals are influenced by physiological factors and that social cognition relies on distributed networks beyond the measured prefrontal channels. The work is presented as an initial step toward understanding the relationship between AQ scores and brain responses in social interaction, requiring further validation with larger, more diverse samples and higher-density montages.

Technical Summary: Autistic Traits and Social-Context-Dependent Prefrontal Hemodynamic Responses

More like this