🧬 biology

Implied Motion as a Mechanism for Spontaneous Belief Computation in Humans

This study demonstrates that the human brain spontaneously computes others' beliefs by encoding them as implied agent-to-object motion, a mechanism supported by intracranial EEG evidence showing shared neural processing in the temporoparietal junction for both belief tracking and visual flow.

Original authors: Patricia Christian, James Cahill, Adeline Fecker, Martha Paskin, Ronja Löfström, Zane Zook, Ahmed Raslan, Kelly Collins, Arvid Guterstam

Published 2026-06-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Patricia Christian, James Cahill, Adeline Fecker, Martha Paskin, Ronja Löfström, Zane Zook, Ahmed Raslan, Kelly Collins, Arvid Guterstam

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 ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed.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 Big Idea: The Brain's "Invisible String"

Imagine you are watching a movie. A character looks at a box and thinks, "There is a ball inside that box!" Even though you can't see the ball, and even though the character is just standing still, your brain automatically knows what they are thinking.

This paper asks a simple but deep question: How does your brain do this so quickly and without you even trying?

The researchers propose a surprising answer: Your brain treats someone else's belief like an invisible string of motion. When a person looks at an object, your brain doesn't just see "eyes" and "object." It automatically draws an invisible line of energy connecting them, as if something is physically moving from the person's head to the object.

The Experiment: The "Ghost Ball" Game

To test this, the researchers created a video game for the brain. Here is how it worked:

The Setup: You watch a short cartoon. A character (the "Agent") sees a ball get stuck behind a screen. Then, the character leaves the room. While they are gone, the ball disappears.
The Twist: The character comes back.
- Scenario A (The Belief): The character thinks the ball is still behind the screen (because they didn't see it leave).
- Scenario B (No Belief): The character knows the ball is gone (because they saw it leave).
The Test: The screen goes blank, and a field of moving dots appears between where the character's head was and where the screen was. You have to quickly guess which way the dots are moving (left or right).

The Discovery:
When the character believed the ball was behind the screen, your brain reacted as if there was a real invisible force moving from the character's head toward the screen.

If the dots moved in the same direction as this "belief string," your brain got confused and you were slightly slower to answer.
If the dots moved in the opposite direction, you were faster.

This is called motion adaptation. It's like if you stare at a waterfall for a minute, then look at a stationary rock; the rock looks like it's moving up. In this study, the "waterfall" was the invisible belief, and the "rock" was the dots. The fact that your brain got "tired" of the belief direction proves it was treating the belief like real motion.

The "Magic Beam" Trick

In a second experiment, the researchers tried to hack this system. They added a very faint, almost invisible beam of moving dots connecting the character to the screen. Most people didn't even notice it was there.

When the beam matched the character's belief: Your brain got a "boost." You were faster at detecting the ball because the invisible beam helped your brain's "belief string" work better.
When the beam went the wrong way: It acted like a traffic jam, slowing your brain down.

Crucial Check: When they replaced the human character with a pile of boxes, this magic beam trick stopped working. This proves that your brain only does this "motion tracking" for people, not for inanimate objects. It's a social superpower.

The "Wiring" Inside the Brain

Finally, the researchers looked directly inside the brains of 15 people (who were already there for epilepsy monitoring) to see which part of the brain was doing the work.

They found that a specific area called the TPJ (Temporoparietal Junction) is the control center for this.

This area is famous for understanding what others are thinking.
It is also famous for processing physical motion (like seeing a car drive by).

The study found that the TPJ lights up in the exact same way whether you are watching a person move physically or watching a person "move" their belief toward an object. It's as if the brain uses the same hardware to track a running dog and to track a person's thoughts.

The Takeaway

Your brain is a master of efficiency. To keep track of what other people are thinking without having to do heavy mental math, it uses a shortcut. It turns "what they believe" into "where they are looking" and draws an invisible line of motion between them.

The Metaphor: Think of your brain as a puppeteer. When you watch someone else, your brain doesn't just watch them; it pulls an invisible string from their eyes to the object they care about. This string helps you instantly know what they are thinking, making social interaction fast and effortless.

What the paper does NOT claim:

It does not say this explains autism (though it suggests future research could look there).
It does not say this can be used to cure social anxiety.
It strictly shows that this "motion" mechanism exists and is used by the brain to compute beliefs spontaneously.

Technical Summary: Implied Motion as a Mechanism for Spontaneous Belief Computation in Humans

Problem Statement

The ability to reconstruct other people's beliefs (Theory of Mind, or ToM) is fundamental to human social interaction. While "explicit" ToM is understood as a slow, effortful process often assessed via false-belief tasks, "spontaneous" ToM refers to swift, unconscious belief processing crucial for rapid social interaction. Despite evidence that humans implicitly track others' beliefs even when task-irrelevant, the underlying neural and cognitive mechanisms allowing the brain to automatically and effortlessly track "who is believing what" remain poorly understood. Specifically, it is unknown how the brain encodes others' attention and belief states during spontaneous ToM without explicit awareness.

Methodology

The study employed a multi-experiment approach combining behavioral paradigms, psychophysics, and intracranial electroencephalography (iEEG) to test the hypothesis that the brain encodes others' attention as an "implied motion" linking an agent to their belief target.

Behavioral Experiments (Experiments 1–4)

Paradigm: Participants viewed animated movies involving an agent, a ball, and an occluder. The critical manipulation involved the agent's belief state regarding the ball's location behind the occluder.
- Agent-Belief Condition: The agent leaves before the ball disappears, returning with a false belief that the ball is still behind the occluder.
- No-Belief Conditions: Variations where the agent either knows the ball is gone or where neither the agent nor the participant believes the ball is present.
Experiment 1 (Motion Adaptation): Following the belief formation phase, a random dot motion stimulus appeared between the agent and the occluder. Participants performed a direction discrimination task. The study measured reaction time (RT) differences between congruent (motion direction matching the implied agent-to-ball direction) and incongruent trials to detect motion adaptation effects.
Experiment 2 (Correlation with ToM): Participants performed both the dot motion task and a ball detection task (measuring spontaneous ToM engagement via detection latency). Pupil dilation was also recorded as a measure of surprise when the agent's expectation was violated.
Experiment 3 (Causal Manipulation): A subthreshold motion "beam" (25% coherence, masked by 100% random noise) was introduced between the agent and the occluder. The beam flowed either congruent or incongruent with the agent's belief direction. Participants performed the ball detection task to see if this covert motion signal manipulated spontaneous ToM responses.
Experiment 4 (Control): Replicated Experiment 3 but replaced the social agent with a pile of boxes to rule out low-level visual motion effects or non-social attentional shifts.

Neurophysiological Experiment (Experiment 5)

Participants: 15 preoperative epilepsy patients with intracranial electrodes implanted in the Temporoparietal Junction (TPJ).
Design: An event-related design presenting either actual visual motion (100% coherent dots) or the belief animations (Agent-Belief vs. Agent-NoBeliefB).
Analysis: High-gamma broadband (HGB) activity (70–200 Hz) was extracted from TPJ contacts. A Support Vector Machine (SVM) classifier was trained to distinguish left vs. right visual motion. This classifier was then tested on the belief trials to see if it could decode the agent's belief direction (left vs. right) based on the neural patterns, using the No-Belief condition as a control.

Key Results

Behavioral Findings

Motion Adaptation to Beliefs (Exp 1): Participants showed a significant motion adaptation effect (slower RTs for congruent motion) only when the agent held a false belief that the ball was behind the occluder (Agent-Belief: $\Delta$ RT = 28 ms, $p < 0.001$ ). No such effect occurred when the agent did not hold this belief, despite identical visual input during the critical 2-second belief representation phase.
Correlation with Spontaneous ToM (Exp 2): There was a significant positive correlation ( $r = 0.37, p = 0.035$ ) between the magnitude of the motion adaptation effect and the degree of spontaneous ToM engagement (measured by ball detection latency). Additionally, pupil dilation was significantly greater in Agent-Belief trials when the agent's expectation was violated, indicating surprise.
Covert Manipulation (Exp 3): Introducing a subthreshold motion beam flowing from the agent to the occluder (congruent with the belief) significantly facilitated ball detection ( $\Delta$ RT = 20 ms, $p = 0.021$ ). Conversely, incongruent motion inhibited detection.
Social Specificity (Exp 4): When the agent was replaced by inanimate boxes, the effect of the subthreshold motion on ball detection disappeared ( $p = 0.401$ ), confirming the effect is specific to social agents.

Neurophysiological Findings (Exp 5)

TPJ Encoding: In 7 out of 79 TPJ electrode contacts (across 7 of 15 patients), a classifier trained on low-level visual motion successfully decoded the direction of the agent's belief during the belief representation phase.
Specificity: This decoding was significantly better than chance and significantly better than performance in the No-Belief control conditions. This indicates that the TPJ processes the belief states of others and low-level visual motion in a similar manner.

Key Contributions

Mechanistic Insight: The study provides evidence that spontaneous ToM relies on an implicit model where attention is encoded as "implied motion" streaming from an agent to a belief target.
Neural Substrate: It identifies the Temporoparietal Junction (TPJ) as a key region where belief states are spontaneously encoded using mechanisms similar to those used for processing physical visual motion.
Causal Evidence: By using subthreshold motion signals, the study demonstrates that visual motion mechanisms can covertly manipulate spontaneous belief computation, suggesting a functional role for motion-based models in social cognition.
Differentiation: The results distinguish spontaneous ToM from explicit reasoning, showing that these processes occur automatically, are task-irrelevant, and rely on specific neural pathways distinct from low-level visual features alone.

Significance and Claims

The authors claim that these findings suggest the brain uses internal motion signals to facilitate belief computations by enhancing the directional connection between agents and belief targets. This "motion-based attention model" allows the brain to efficiently keep track of "who is believing what" without conscious effort.

The paper posits that this mechanism may have implications for understanding autism spectrum disorder, given that autistic individuals often exhibit differences in both spontaneous ToM and visual motion processing. The authors suggest a potential shared neurobiological underpinning in the TPJ, where visual motion systems might be disengaged during the encoding of others' attention in autism, affecting the ability to automatically compute beliefs. However, the paper maintains modesty regarding these implications, framing them as predictions for future testing rather than established conclusions.

Ultimately, the work demonstrates how low-level sensory mechanisms (visual motion systems) are recruited to support high-level social cognition (belief computation), offering a novel link between sensory processing and Theory of Mind.