Back to docs

Tool Cognitive Coverage

Category: Cognitive Transport · Since: v18.27.0 · Tier: All

Select the N most cognitively different personas from a list. Uses greedy farthest-point sampling in Wasserstein space. Each selected persona is as far as possible from all previously selected ones.

Why This Matters

Testing with 3 random personas might pick 3 similar ones. For example: first-timer, cautious-user, elderly-user all have low risk tolerance. This tool guarantees maximum cognitive diversity. Selected personas cover the widest possible range of cognitive profiles.

Parameters

Parameter Type Required Description
personas string[] Yes List of persona names to select from
count number Yes Number of personas to select

Example

Select the 3 most different personas from: first-timer, power-user, elderly-user, mobile-user, impatient-user, screen-reader-user, cognitive-adhd

Response

{
  "selectedPersonas": ["power-user", "cognitive-adhd", "screen-reader-user"],
  "coverage": "Maximum cognitive diversity",
  "distanceMatrix": [
    { "persona": "power-user", "distances": { "power-user": 0, "cognitive-adhd": 0.2847, "screen-reader-user": 0.2103 } },
    { "persona": "cognitive-adhd", "distances": { "power-user": 0.2847, "cognitive-adhd": 0, "screen-reader-user": 0.1956 } },
    { "persona": "screen-reader-user", "distances": { "power-user": 0.2103, "cognitive-adhd": 0.1956, "screen-reader-user": 0 } }
  ]
}

Algorithm

Greedy farthest-point sampling:

  1. Start with the first persona
  2. Find the remaining persona with the maximum minimum distance to all selected
  3. Add it to the selection
  4. Repeat until N personas are selected

This maximizes the minimum pairwise distance. It gives optimal coverage of the cognitive space.

Related Tools

From the Blog