Treat humanlikeness as a tunable design parameter and anthropomorphism as an evaluation metric¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The review recommends that once the constructs are distinguished, designers can "intentionally manipulating humanlike features of robots" to systematically amplify or attenuate attribution of specific human qualities aligned with deployment contexts, and treat anthropomorphism as an outcome measure rather than a confound. For example, in hospital contexts designers may suppress facial cues inviting moral-agency attributions while leveraging other humanlike features for task-focused interaction. If people anthropomorphize a socially evocative robot consistent with its design goals, the design achieved its intended interaction purpose.

Design Implications¶

Context¶

Requirements¶

• Requires treating humanlikeness as a multidimensional set of design inputs and anthropomorphism as a non-monolithic set of attributional outcomes

Constraints¶

• The review notes attempts to eliminate human anthropomorphism entirely may be too difficult an endeavor

Target Learners¶

• Robot design teams
• HRI practitioners

Target Learning Goals¶

• Aligning robot design features with desired attribution outcomes in specific deployment contexts

Affordances¶

• Humanlikeness Design Anthropomorphism Inference Framework

Related Strategies¶

Examples¶

-

Key Sources¶

• Phillips, E. K., & de Visser, E. J. (2026). Humanlikeness as design, anthropomorphism as inference: a conceptual framework for human–robot interaction. Frontiers in Cognition. https://doi.org/10.3389/fcogn.2026.1786256