Cognitive Task Design Audit

Description

Auditing qualitative findings against three task design dimensions (complexity match, feedback adequacy, error detectability) to turn observations into design-relevant diagnoses.

Cognitive task design, as framed by Hollnagel and colleagues, treats a task as joint work between people and systems, designed to fit the actual cognitive demands of the work. This audit applies three lenses to qualitative evidence. Complexity match asks whether the designed task reflects the real cognitive complexity, signalled by elaborate workarounds or decisions users cannot articulate. Feedback adequacy asks whether users can tell their actions had the intended effect, signalled by re-checking and repeated actions. Error detectability asks whether users can notice errors before consequences become severe, signalled by errors found only at submission or by others. Mapping observations onto these dimensions converts scattered findings into structured design problems.

Attributes

Loss type
Siloed
Effort
Days

Steps

  1. Assemble qualitative evidence from interviews, observation or diary data for the task under study.
  2. Code each relevant observation against the three dimensions: complexity match, feedback adequacy and error detectability.
  3. For each dimension, rate the task (for example adequate, partial, failing) and cite the supporting incidents.
  4. Identify the task steps where failures cluster and estimate the severity of consequences at each.
  5. Translate each failing dimension into a design requirement, such as an explicit state indicator or pre-commit review step.
  6. Review the audit with designers and engineers and track requirements into the backlog.

Sources

  • Handbook of Cognitive Task Design (Hollnagel)
  • Human-Computer Interaction (Alan Dix)
  • Shneiderman & Plaisant
  • Movie Booking Playbook (Gopinath)
  • The Inquiry Series, Ch 16 — Cognitive Task Modeling — GOMS, KLM & CPM-GOMS

Also known as

  • CTD audit
  • Complexity-feedback-detectability audit

Categories

  • Accessibility
  • Human Factors & Error