The thing that should not be: predictive coding and the uncanny valley in perceiving human and humanoid robot actions
Source record
saygin-2012-predictive-coding· 2012 · Social Cognitive and Affective Neuroscience
Evidence tier: B (peer-reviewed-journal) · Verification: verified-fulltext
Bottom line: Repetition suppression in bilateral anterior intraparietal sulcus was stronger for the android (appearance–motion mismatch) than for the human or robot, interpreted as prediction error; authors state this cannot establish a causal link to the uncanny valley.
Citation
Ayse Pinar Saygin, Thierry Chaminade, Hiroshi Ishiguro, Jon Driver, Chris Frith (2012). The thing that should not be: predictive coding and the uncanny valley in perceiving human and humanoid robot actions. Social Cognitive and Affective Neuroscience, 7(4), 413-422. https://doi.org/10.1093/scan/nsr025
Verification entry (identifiers, resolvers, claims with source locations): /data/citations.json#saygin-2012-predictive-coding
Study design
| Field | Value |
|---|---|
| Design | Event-related fMRI repetition suppression (3T); 2-s grayscale clips of 8 actions by a human, the Repliee Q2 android (human appearance, mechanical motion) and the same robot with skin removed (mechanical appearance and motion) |
| Modality | video, robot |
| Stimuli | Action videos: Human, Android, Robot |
| Participants (N) | 19 |
| N detail | 20 recruited; 1 excluded for head motion |
| Topics | uncanny-valley, neural, predictive-coding |
Results
| Metric | Value | Note |
|---|---|---|
| aIPS repetition suppression | Android > Human and Robot (bilateral) | key action-perception node |
| Left extrastriate body area | Human and Android > Robot | sensitive to humanlike form |
| Selectivity for appearance or motion alone | little (outside EBA) | network responded to the match between them |
Mechanism relevance
perceptual-mismatch, predictive-coding
Limitations
- N=19
- Single android
- Authors: no conclusive or causal link to the uncanny valley
Cited in
- MR-0002: The Uncanny Valley and Synthetic Media: Does It Exist, What Causes It, and Have Deepfakes Crossed It?
Verification
| Field | Value |
|---|---|
| Status | verified-fulltext |
| Compiler confidence | high |
| Checked on | 2026-10-07 |
| Via | academic.oup.com article page |