Perception
Do people prone to hallucinations expect to see faces in noise?
Open access · cc by · source: Europe PMC
Healthy people who reported more hallucination-like experiences were more inclined to 'see' faces in visual noise and became aware of faces looking straight at them faster, fitting the idea that psychosis involves overly strong expectations for socially meaningful signals.
Study at a glance
- Design
- Cross-sectional — Correlational study in a non-clinical sample: questionnaire measures of psychosis proneness related to a Bayesian-modelled face-detection-in-noise task and a breaking continuous flash suppression (b-CFS) gaze task, using rank and Bayesian correlations
- N
- N=39 · Thirty-nine participants from the general population completed the face task; three were excluded from the gaze task (one technical problem, two with excessive suppression), leaving 36 for gaze analyses.
- Population
- Healthy adults recruited from the general population in Berlin, with relatively high psychosis-proneness scores
- Outcome
- Estimated prior and sensitivity for detecting faces in noise; direct-gaze bias (difference in breakthrough time for averted versus direct gaze); correlations with Peters Delusion Inventory and Cardiff Anomalous Perceptions Scale scores
Structured fields used in claim comparison tables when every cited study has a complete layer.
Key findings
A stronger prior for detecting faces correlated with both hallucination proneness (r = 0.496) and delusion proneness (r = 0.461), whereas sensitivity to real faces was unrelated to either, so the effect was a bias rather than better or worse vision. Across the sample, faces with direct gaze broke through suppression faster than averted-gaze faces (on average 3.385 versus 4.055 seconds). The size of this direct-gaze advantage correlated with hallucination proneness (r = 0.429) but not significantly with delusion proneness.
Methodology
Thirty-nine adults from the general population completed questionnaires measuring delusion proneness and anomalous (hallucination-like) perceptual experiences. In a face task they judged whether each noisy image contained a face (40 images hid a real face and 60 were pure noise); a Bayesian model separated each person's prior expectation of faces from their sensitivity to real faces. In a gaze task, faces with direct or averted gaze were shown to one eye while flashing masks were shown to the other (continuous flash suppression), and the researchers timed how long each face took to break into awareness.
Limitations
The sample was small and consisted only of healthy volunteers, so the authors note a study in people with psychosis is needed to confirm that the findings apply clinically. The design is correlational, so it cannot show that strong priors cause hallucination-like experiences. Because only face stimuli were used, the study cannot tell whether the bias is specific to social signals or reflects a general tendency to see meaning in noise; the authors call for a non-social control condition. The 'prior' is inferred from behaviour, not experimentally manipulated, and the evidence for the gaze–hallucination link was only substantial rather than strong.
How this study connects
Role on claims
Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.
Not yet placed on a claim. This paper has study layers, but no concept page yet cites it as support, challenge, or qualifier.
Related papers in this topic
Same topic cluster — not a recommendation engine.