Think-Aloud Protocol

Description

Participants verbalise their thoughts while (or after) performing a task, giving direct evidence of expectations, interpretations and confusions behind their behaviour.

The think-aloud protocol asks participants to say what they are thinking as they work through a task. In concurrent think-aloud they verbalise during the task, which gives access to working-memory contents with minimal reconstruction. In retrospective think-aloud they verbalise afterwards, usually while watching a recording, which avoids interfering with task performance but relies on memory. Ericsson and Simon's theory distinguishes pure verbalisation of attended information from explanations, which can alter cognition. Rigour comes from a short practice task, neutral prompts such as 'keep talking' rather than 'why did you do that?', avoiding interruptions during critical steps, and analysing verbalisations alongside behaviour rather than treating comments as opinions.

Attributes

Loss type
Active
Effort
Days

Steps

  1. Explain the procedure and run a short practice task, such as counting windows in one's home, to train verbalisation.
  2. Choose concurrent or retrospective think-aloud according to whether task timing matters.
  3. During tasks, use only neutral reminders such as 'please keep talking' after about 15 seconds of silence.
  4. Avoid asking for explanations mid-task; save probes for after the task.
  5. Record audio, screen and, where useful, face video.
  6. Transcribe or annotate verbalisations against task steps and code expectations, confusions and decisions.

Sources

  • Think Like a UX Researcher
  • Observing the User Experience
  • Protocol Analysis (Ericsson & Simon)
  • Drachen, A., Mirza-Babaei, P. & Nacke, L. (eds), Games User Research (2018)
  • Garcia-Ruiz, M. (ed.), Games User Research: A Case Study Approach (2016)

Also known as

  • Concurrent think-aloud
  • Retrospective think-aloud
  • Verbal reporting

Categories

  • User Research
  • Cognitive Psychology & Perception