Description
Models information seeking as foraging that maximises information gain per unit cost, quantifying information scent, patch residence and diet choice.
Borrowing from optimal foraging theory in ecology, Pirolli and Card modelled users as foragers who move between information patches (pages, search results, documents) and decide when to leave a patch using Charnov's marginal value theorem: leave when the rate of gain in the current patch drops below the average rate in the environment. Information scent, the perceived value of a link from its proximal cues, determines navigation, and was computationally modelled with spreading activation (SNIF-ACT). Rigorous application measures scent quantitatively, predicts navigation choices and patch-leaving, and compares predicted and observed paths. UX retains 'information scent' as a slogan while dropping the models and measurements.
Attributes
- Discipline
- Cognitive Psychology & Perception
- Loss type
- Vestigial
- Stage
- Evaluation
- Data
- Mixed
- Effort
- Weeks
Steps
- Define the user's information goal precisely and the information environment (patches, links, cues).
- Measure information scent: have participants or a semantic model rate how likely each link leads to the goal.
- Predict navigation by choosing the highest-scent link at each step, and predict patch-leaving via the marginal value theorem.
- Collect navigation and dwell data with logs or a tree-test-like task.
- Compare predicted and observed paths and patch residence times; identify scent misleads and dead patches.
- Rewrite labels and summaries to strengthen accurate scent and re-measure.
Sources
- Pirolli & Card, Information foraging (1999), Psychological Review
- Pirolli, Information Foraging Theory: Adaptive Interaction with Information (2007)
Also known as
- Information foraging theory
- Information scent analysis
- IFT
Categories
- User Research
- Cognitive Psychology & Perception