Saved in:
Bibliographic Details
Main Authors: Pegler, Dominik, Steyrl, David, Zhang, Mengfan, Karner, Alexander, Arato, Jozsef, Scharnowski, Frank, Melinscak, Filip
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.04889
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918315113316352
author Pegler, Dominik
Steyrl, David
Zhang, Mengfan
Karner, Alexander
Arato, Jozsef
Scharnowski, Frank
Melinscak, Filip
author_facet Pegler, Dominik
Steyrl, David
Zhang, Mengfan
Karner, Alexander
Arato, Jozsef
Scharnowski, Frank
Melinscak, Filip
contents Phobias are common and impairing, and exposure therapy, which involves confronting patients with fear-provoking visual stimuli, is the most effective treatment. Scalable computerized exposure therapy requires automated prediction of fear directly from image content to adapt stimulus selection and treatment intensity. Whether such predictions can be made reliably and generalize across individuals and stimuli, however, remains unknown. Here we show that pretrained convolutional and transformer vision models, adapted via transfer learning, accurately predict group-level perceived fear for spider-related images, even when evaluated on new people and new images, achieving a mean absolute error (MAE) below 10 units on the 0-100 fear scale. Visual explanation analyses indicate that predictions are driven by spider-specific regions in the images. Learning-curve analyses show that transformer models are data efficient and approach performance saturation with the available data (~300 images). Prediction errors increase for very low and very high fear levels and within specific categories of images. These results establish transparent, data-driven fear estimation from images, laying the groundwork for adaptive digital mental health tools.
format Preprint
id arxiv_https___arxiv_org_abs_2509_04889
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SpiderNets: Vision Models Predict Human Fear From Aversive Images
Pegler, Dominik
Steyrl, David
Zhang, Mengfan
Karner, Alexander
Arato, Jozsef
Scharnowski, Frank
Melinscak, Filip
Computer Vision and Pattern Recognition
Artificial Intelligence
Human-Computer Interaction
Machine Learning
Phobias are common and impairing, and exposure therapy, which involves confronting patients with fear-provoking visual stimuli, is the most effective treatment. Scalable computerized exposure therapy requires automated prediction of fear directly from image content to adapt stimulus selection and treatment intensity. Whether such predictions can be made reliably and generalize across individuals and stimuli, however, remains unknown. Here we show that pretrained convolutional and transformer vision models, adapted via transfer learning, accurately predict group-level perceived fear for spider-related images, even when evaluated on new people and new images, achieving a mean absolute error (MAE) below 10 units on the 0-100 fear scale. Visual explanation analyses indicate that predictions are driven by spider-specific regions in the images. Learning-curve analyses show that transformer models are data efficient and approach performance saturation with the available data (~300 images). Prediction errors increase for very low and very high fear levels and within specific categories of images. These results establish transparent, data-driven fear estimation from images, laying the groundwork for adaptive digital mental health tools.
title SpiderNets: Vision Models Predict Human Fear From Aversive Images
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Human-Computer Interaction
Machine Learning
url https://arxiv.org/abs/2509.04889