![[Adobe Stock]](https://sp-ao.shortpixel.ai/client/to_webp,q_glossy,ret_img,w_750,h_375/https://www.psypost.org/wp-content/uploads/2026/07/scary-smile-750x375.jpg)
A new study published in the Quarterly Journal of Experimental Psychology suggests that scary music can subtly alter how people interpret human facial expressions, making even smiling faces seem somewhat threatening. By measuring both subjective ratings and involuntary eye responses, scientists observed that ominous soundtracks biased participants toward perceiving greater danger in human faces. These findings demonstrate how background auditory cues might unconsciously shape everyday social evaluations and media interpretation.
Human interaction relies heavily on reading facial expressions to evaluate whether someone poses a threat or offers safety. Background sounds often establish an emotional context that influences how subsequent information is processed, a phenomenon known as affective priming. When an initial stimulus carries a strong emotional tone, it tends to frame how a person interprets visual cues that follow shortly after.
Horror film soundtracks are constructed to evoke anxiety through rapid volume changes, jarring pitches, and acoustic patterns that mimic human cries. These fear-inducing sounds tend to trigger subcortical brain regions that monitor the environment for danger. The authors conducted this research to explore whether auditory fear cues alter subjective perceptions of social threat, and whether these changes are accompanied by physiological markers of arousal.
Shumetha Sidhu, a lecturer in psychology and director of postgraduate research studies at the University of Reading Malaysia, explained what led her team to pursue this topic. “It has always been fascinating to me how preconceived notions or other cognitive factors can influence perception, and this study came about because my students were interested in emotional face perception research too,” Sidhu told PsyPost. “A bit of literature reviewing showed this was an underexplored area, and we decided to run with it using the angle of threat perception as it seems highly relevant to a lot of people.”
To test this possibility, the researchers recruited 45 university students in Malaysia, aged 18 to 25, with an average age of approximately 21 years. The participant group was predominantly female, including 42 females and three males, with a multiethnic background comprising Chinese, Malay, Indian, and other ethnicities. Participants sat 60 centimeters away from an eye-tracking screen while wearing over-ear headphones fixed at a set volume.
The study used a within-subjects experimental design, meaning every participant experienced all combinations of sounds and facial expressions. The researchers controlled for environmental influences by maintaining constant overhead lighting and using a fixed monitor display. In each trial, participants listened to a 15-second audio clip followed immediately by a photograph of a face displayed for one second.
The auditory clips featured either a fear-inducing track from the film *Insidious* or neutral white noise as a control baseline. The visual stimuli consisted of 90 photographs depicting young adult Chinese models showing angry, neutral, or happy expressions. Participants completed 90 randomized trials in total, split across three separate testing blocks to minimize fatigue.
After viewing each face, participants rated how threatening the image appeared on a continuous scale from 0, indicating not at all threatening, to 100, indicating extremely threatening. To hide the true intent of the experiment, participants also rated how exciting and relaxing the faces appeared. Physiological response was measured using pupillometry, a technique that tracks changes in pupil diameter to index autonomic nervous system activity and heightened attention.
For the threat rating analysis, data from three participants were excluded as statistical outliers, leaving a sample of 42 individuals. Exposure to scary music led to a notable overall shift in perceived threat across all facial expressions. The average threat rating across all face types rose from 18.90 points out of 100 during white noise to 41.65 points during scary music, representing an absolute increase of 22.75 points and more than doubling the baseline score.
Angry faces received the highest threat scores in both conditions, moving from 34.92 points under white noise to 52.93 points under scary music. Neutral faces saw an increase from a baseline of 14.18 points with white noise up to 39.62 points with scary music. Happy faces, which averaged a low threat rating of 7.58 points following white noise, rose to 32.41 points when paired with scary music.
The reaction to smiling faces proved particularly unexpected for the research team. “We didn’t expect that fear-inducing music would increase threat perception towards happy faces!” Sidhu noted.
During the neutral white noise condition, a distinct hierarchy emerged where angry faces were judged as most threatening, followed by neutral faces, and finally happy faces. When scary music was played, angry faces remained the highest in threat, but the difference between neutral faces and happy faces was no longer statistically significant. This outcome suggests that fear-inducing audio may blur the boundary between emotionally neutral signals and friendly expressions.
Pupillometry data were analyzed for all 45 participants, as no outliers were present in the eye-tracking measurements. The main difference in average pupil diameter between the scary music condition (2.94 millimeters) and the white noise condition (2.93 millimeters) was not statistically significant. This indicates that scary music did not produce a widespread, uniform enlargement of pupil size across all facial expressions.
The scientists observed a specific interaction between music type and facial expression regarding pupil size. When participants viewed neutral faces after listening to scary music, their average pupil diameter expanded to 2.94 millimeters compared to 2.91 millimeters after white noise. In contrast, pupil size changes for angry faces (2.95 millimeters under both sounds) and happy faces (2.93 millimeters with white noise versus 2.92 millimeters with scary music) were not statistically significant across sound conditions.
Regarding the broader message of these outcomes, Sidhu emphasized how external environments shape social processing. The primary takeaway, she explained, is “that fear-inducing music can influence perception of not just negative and neutral emotions, but also positive ones (happy faces). It highlights the importance of contextual cues when processing information so as not to be easily biased.”
The findings do not suggest that scary music induces full-blown panic or causes people to view ordinary faces as severe physical threats. The threat ratings for smiling faces paired with scary music averaged approximately 32 on a 100-point scale, indicating mild unease or ambiguity rather than extreme terror. The auditory priming effect appears to introduce a subtle interpretive bias rather than a total distortion of reality.
The authors also pointed out that readers should consider the specific demographics of the study when evaluating the results. “We have a predominantly female sample, so I will not overly generalize these findings,” Sidhu stated. “Likewise, we used a South East Asian sample, so no guarantee that it generalizes to other demographics.”
Moving forward, the research team aims to explore why positive expressions become eerie under scary audio conditions. “The unexpected finding we had was that people perceived happy faces as more threatening when exposed to the fear-inducing music,” Sidhu said. “One interpretation for this could be that a smile coupled with that type of soundtrack was conveying sinister intent. We plan on investigating this a bit further down the line.”
Future work could also examine whether music-induced changes in facial threat perception spill over into real-world choices and social behaviors. Scientists could explore this question by pairing different musical backgrounds with faces during economic trust games to observe whether ominous sounds alter financial decisions or social cooperation.
The study, “The Sound of Fear: How Horror Soundtracks Shape Threat Perception of Emotional Faces,” was authored by Li Fun Lua, Xuan Nee Kong, Jasmine K. W. Lee, and Shumetha Kaur Sidhu.