Summary: Researchers at the University of Hong Kong have created an AI-driven brain model that tracks fear as it unfolds in realistic, naturalistic situations. This approach departs from conventional laboratory methods that rely on static images, better reflecting how fear operates in dynamic, everyday contexts. Using the new model, the team showed that the hormone oxytocin specifically reduces fear during social experiences, revealing a promising targeted mechanism for treating social anxiety and related conditions.
These results suggest a path toward clinical interventions that reflect real-life emotional processing and may improve treatment for people with excessive social fear, including social anxiety disorder, social phobia, and autism-related social challenges.
Key Facts
- Real-World Fear Mapping: An AI-inspired brain model tracked fear more accurately during naturalistic experiences than models based on static laboratory images.
- Social Fear Reduction: Oxytocin selectively lowered both reported fear and a brain-wide neural signature of fear in social—but not non-social—contexts.
- New Treatment Pathway: The findings support a targeted strategy for clinical conditions characterized by excessive social fear and provide a tool to design interventions grounded in real-life emotional dynamics.
Source: University of Hong Kong
Researchers at The University of Hong Kong (HKU) have introduced a practical, ecologically valid approach to measuring and modulating fear in everyday settings.
Traditional fear research often uses simplified stimuli—static images or isolated cues—that fail to capture the complexity of fear in dynamic, real-world situations. To bridge that gap, Professor Benjamin Becker and colleagues in HKU’s Department of Psychology developed an AI-driven brain model capable of tracking the conscious experience of fear as it unfolds during naturalistic tasks such as watching immersive video scenes.
The researchers first demonstrated that existing fear models, trained on static laboratory images, do not reliably predict fear responses during lifelike experiences. They then trained and validated a new model tailored to capture fear representations in dynamic, social and non-social contexts. Using this model, they ran a preregistered, double-blind placebo-controlled trial testing intranasal oxytocin (24 IU) in healthy male participants.
Results showed that oxytocin selectively reduced subjective fear ratings and diminished the brain-wide neural signature of fear—but only when participants experienced fear in social settings. At the neural circuit level, oxytocin increased activation in the left middle cingulate cortex (lMCC) and strengthened functional connectivity between the lMCC and the opposite-side amygdala. Both of these neural changes were inversely related to the participants’ self-reported fear after oxytocin administration.
On a broader scale, oxytocin also modulated large-scale brain network communication, enhancing interactions between the dorsal attention network (DAN), the fronto-parietal network (FPN), and the default-mode network (DMN). Applying an independent neuromarker that captures activity and connectivity patterns associated with fear in naturalistic contexts (the CAFE marker) confirmed that oxytocin reduced brain-wide expressions of fear.
Taken together, these findings indicate that oxytocin exerts an ecologically valid, socially specific anxiolytic effect by enhancing top-down regulation from the middle cingulate cortex over amygdala-related fear processes and by reshaping brain-wide fear representations. This points to a precision-based strategy for treating disorders where social fear is prominent.
Key Questions Answered:
A: Traditional paradigms typically use static images or simplified cues that lack the temporal and social complexity of everyday fear. The AI-based model shows that fear processing differs substantially in dynamic settings, revealing limitations in decades of laboratory work.
A: Oxytocin reduced both subjective fear ratings and the neural patterns associated with fear, but only during social contexts. Neural data indicate oxytocin boosts middle cingulate activity and its regulation of the amygdala, which together correlate with lower reported fear.
A: By identifying fear signatures in realistic social scenarios and showing a selective oxytocin effect, the model supports developing targeted, mechanism-based interventions that specifically address the neural circuitry underlying social fear.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was read and reviewed in full by the editorial team.
- Relevant context and clarifications were added by staff writers to improve readability and accuracy.
About this AI and fear research news
Author: Jaymee Ng
Source: University of Hong Kong
Contact: Jaymee Ng – University of Hong Kong
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Oxytocin Reduces Subjective Fear in Naturalistic Social Contexts via Enhancing Top-Down Middle Cingulate Amygdala Regulation and Brain-Wide Fear Representations” by Benjamin Becker et al. Advanced Science
Abstract
Oxytocin Reduces Subjective Fear in Naturalistic Social Contexts via Enhancing Top-Down Middle Cingulate Amygdala Regulation and Brain-Wide Fear Representations
A growing body of research suggests that the neuropeptide oxytocin (OT) can regulate fear, but translating those findings into clinical interventions requires evidence gathered under conditions that mimic real life. This study combined a naturalistic fMRI design that produced immersive fear experiences in both social and non-social contexts with a preregistered, randomized, double-blind, placebo-controlled intranasal oxytocin trial (24 IU; n = 67 healthy men).
Oxytocin selectively reduced subjective fear in social contexts while leaving non-social fear unchanged. Neurofunctionally, oxytocin increased activation in the left middle cingulate cortex and strengthened its connectivity with the contralateral amygdala; these neural changes were inversely associated with subjective fear. At the network level, oxytocin enhanced communication between the dorsal attention network, the fronto-parietal network, and the default-mode network, and it shifted brain-wide communication patterns.
Using an independent activity-connectivity neuromarker designed for naturalistic fear (CAFE) confirmed that oxytocin attenuated brain-wide fear expressions. These results indicate a socially specific, ecologically valid fear-reducing effect of oxytocin and highlight its potential as a treatment for disorders characterized by excessive social fear.