Effects of Facial Photo Filters on Judgments of Attractiveness and Trustworthiness: An Online Experiment with AI-Generated Faces

Facial photo filters are widely used when sharing selfies, but how others perceive filtered faces has rarely been examined experimentally. In this online experiment, I examined whether a typical facial photo filter from Instagram affects judgments of the attractiveness and trustworthiness of female faces. To rule out familiarity effects, 40 AI-generated (StyleGAN) faces of women (39 analyzed) were presented either unfiltered or modified with a filter that combines soft-focus, contrast, and exposure adjustments. Participants (N = 69, 39 women and 30 men, aged 18–63 years) rated each face on 10-point scales. The four questionnaire versions were randomly allocated to recruitment channels rather than to individual participants, so condition was confounded with viewer gender (filter: 28 men and 6 women, no filter: 33 women and 2 men). Filtered faces received slightly lower mean ratings of attractiveness (Hedges’ g = −0.32, 95% CI [−0.79, 0.15]) and trustworthiness (g = −0.32, 95% CI [−0.79, 0.15]), but neither difference was statistically significant. Mixed-effects models with crossed random effects for participants and faces led to the same conclusion. Adjusting for viewer gender reduced the attractiveness difference to nearly zero but changed the trustworthiness difference only slightly. Gender differences could not be tested. About half of the women and a quarter of the men reported editing photos of themselves. The extent of editing was not significantly correlated with age or education. The findings do not support the assumption that such filters lower these judgments. Adequately powered studies with randomization at the participant level are needed.

Authors

Institutions

Publication Details

Journal
Journal of Medical - Clinical Research & Reviews
Published
2026-11-16
DOI
https://doi.org/10.33425/2639-944x.1509
Primary Topic
Evolutionary Psychology and Human Behavior
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Effects of Facial Photo Filters on Judgments of Attractiveness and Trustworthiness: An Online Experiment with AI-Generated Faces

Darius Krutze
2 links
Journal of Medical - Clinical Research & Reviews
Evolutionary Psychology and Human Behavior
article

Effects of Facial Photo Filters on Judgments of Attractiveness and Trustworthiness: An Online Experiment with AI-Generated Faces

Darius Krutze
article en

Abstract

Facial photo filters are widely used when sharing selfies, but how others perceive filtered faces has rarely been examined experimentally. In this online experiment, I examined whether a typical facial photo filter from Instagram affects judgments of the attractiveness and trustworthiness of female faces. To rule out familiarity effects, 40 AI-generated (StyleGAN) faces of women (39 analyzed) were presented either unfiltered or modified with a filter that combines soft-focus, contrast, and exposure adjustments. Participants (N = 69, 39 women and 30 men, aged 18–63 years) rated each face on 10-point scales. The four questionnaire versions were randomly allocated to recruitment channels rather than to individual participants, so condition was confounded with viewer gender (filter: 28 men and 6 women, no filter: 33 women and 2 men). Filtered faces received slightly lower mean ratings of attractiveness (Hedges’ g = −0.32, 95% CI [−0.79, 0.15]) and trustworthiness (g = −0.32, 95% CI [−0.79, 0.15]), but neither difference was statistically significant. Mixed-effects models with crossed random effects for participants and faces led to the same conclusion. Adjusting for viewer gender reduced the attractiveness difference to nearly zero but changed the trustworthiness difference only slightly. Gender differences could not be tested. About half of the women and a quarter of the men reported editing photos of themselves. The extent of editing was not significantly correlated with age or education. The findings do not support the assumption that such filters lower these judgments. Adequately powered studies with randomization at the participant level are needed.

Journal of Medical - Clinical Research & ReviewsVol. 10(11)
Hochschule Fresenius (DE)
Evolutionary Psychology and Human Behavior
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon