The term uncanny valley describes the dip in our comfort or affinity that often occurs when a non-human figure becomes highly realistic yet still not perfectly human. As realism increases, people’s positive reactions rise up to a point, but a small gap where the appearance is almost human but not quite triggers a sense of unease or eeriness. This effect is widely discussed in robotics, computer animation, and virtual avatars, and it helps explain why extremely lifelike characters can feel off-putting to some viewers rather than warmly familiar. The concept was first introduced by Masahiro Mori in 1970 and has since been observed across robots, CGI characters, lifelike dolls, and virtual beings in VR/AR contexts.

Key points you might find helpful

• Origins and core idea: The “valley” represents a rapid drop in comfort as resemblance to humans crosses from “clearly non-human” toward “almost human.”

• Where it appears: Not limited to robots—also relevant for CGI characters, realistic avatars, and other near-human digital or mechanical figures.

• Why it happens: A mix of perceptual mismatch, cognitive categorization challenges, and potential threat or moral animacy concerns can contribute to unease when details conflict with expectations of human appearance and behavior.

• Classic examples: Some motion-captured or highly realistic animated characters can feel stiff or uncanny to viewers, illustrating the valley in action. Contemporary discussions also cover advances in AI-generated faces and virtual agents.

• Designer takeaway: To avoid the uncanny valley, many creators opt for stylized or intentionally non-photorealistic designs, or strive for near-perfect human likeness with extremely careful attention to movement and expressiveness.

If you’d like, I can summarize these sources in more detail, compare different explanations for why the uncanny valley occurs, or provide a quick glossary of related terms in multiple languages.