Foundation Track·Core Concept

Anthropomorphism and the Uncanny Valley

35 min
1 / 10

The Question That Defines Robot Design

One single design decision dictates user trust, emotional response, and long-term safety. How much should this robot look and act like a human being?

This question is deceptively simple. Most designers answer it with pure aesthetic instinct. The marketer wants a "friendly" face. The engineer wants a functional box. The safety team wants something that is glaringly mechanical. All of these impulses are guesses.

HRI design replaces guesswork with testable decisions informed by psychology, interaction research, and the deployment context. Anthropomorphism is not simply about making things cute; it shapes the expectations people may form about a robot's agency, social ability, and competence.

The Anthropomorphism Spectrum
FIGURE 4.1: Anthropomorphic cues can be treated as design variables whose effects should be evaluated in context.
References3 selected references · updated Aug 9, 2026

These sources support cautious use of anthropomorphic cues and the uncanny-valley hypothesis. The effect varies with stimuli, measurement, and context; it is not a universal law or a single fixed realism threshold.

  1. 01
    The Uncanny Valley

    Masahiro Mori; translated by Karl F. MacDorman and Norri Kageki · 2012 · IEEE Robotics & Automation Magazine

    Supports: The original uncanny-valley hypothesis and its relationship to human likeness and movement.

    Peer-reviewed research
  2. 02
    A Meta-analysis of the Uncanny Valley’s Independent and Dependent Variables

    Alexander Diel, Sarah Weigelt, and Karl F. MacDorman · 2022 · ACM Transactions on Human-Robot Interaction

    Supports: Evidence, measurement differences, and limitations across uncanny-valley studies.

    Peer-reviewed research
  3. 03
    Anthropomorphism and the Social Robot

    Brian R. Duffy · 2003 · Robotics and Autonomous Systems

    Supports: Judicious use of anthropomorphism in robot appearance and behavior.

    Peer-reviewed research
← PreviousMultimodal Communication in HRI
Next →Trust, Safety, and Graceful Failure