Embodied Simulation and Interface Metaphor Grounding

Description

Tests whether interface labels, gestures and metaphors evoke mental simulations compatible with the required action, using response-time compatibility effects and comprehension measures.

Embodied simulation theory holds that understanding language involves partially re-enacting the perceptual and motor experiences it describes. Glenberg and Kaschak's action-sentence compatibility effect showed that people respond faster when the response direction matches the motion implied by a sentence. Bergen summarises evidence that people simulate orientation, direction, shape and effort. For interfaces, labels and metaphors such as 'swipe up to archive' or 'pull to refresh' are understood partly by simulation, so mismatches between the implied and the required action slow responses and cause errors. The method identifies the action or spatial content each label implies, measures response times and errors when required actions match or conflict with it, and favours wording and gesture mappings that evoke compatible simulations.

Attributes

Loss type
Unmigrated
Effort
Days

Steps

  1. Test simulation ease of interface labels: present labels in isolation, ask users to describe the action the label refers to and mime the motion (if applicable). Measure:
  2. response latency (faster = easier simulation),
  3. action consistency across participants (% agreement on the described action),
  4. simulation accuracy (does the simulated action match the actual afforded action?). Labels with latency > 1.5s or accuracy < 70% are redesign candidates.

Sources

  • Louder Than Words — Benjamin K. Bergen
  • Glenberg & Kaschak, Grounding language in action (Psychonomic Bulletin & Review, 2002)
  • Barsalou, Perceptual symbol systems (Behavioral and Brain Sciences, 1999)

Also known as

  • Embodied simulation testing
  • Action-compatibility testing of labels

Categories

  • User Research
  • Cognitive Psychology & Perception