Fitts' Law Modelling

Description

Predicts and measures the time to acquire a target from its distance and size, calibrating target sizes and placements for pointing and touch interfaces.

Fitts' law states that movement time to a target grows linearly with the index of difficulty, ID = log2(D/W + 1), where D is distance and W is target width: MT = a + b·ID. The constants a and b depend on device, limb and context, so the rigorous use of the law is calibration, not quotation. Researchers run a multidirectional tapping task (ISO 9241-411) across a range of distances and widths, fit the regression and compute throughput in bits per second, then use the fitted model to predict times for proposed layouts. For touch, effective width and error rates matter as much as speed, and context such as gloves, vibration or one-handed use changes the parameters. A well-fitted model (R² above about 0.9) lets teams compare layouts quantitatively before building them.

Attributes

Loss type
Active
Stage
Design
Effort
Days

Steps

  1. Define the device, input method and context of use (e.g. gloved touch, one-handed mobile, mouse).
  2. Set up an ISO 9241-411 style multidirectional tapping task covering at least 3 distances and 3 widths.
  3. Collect data from 12-20 participants with enough trials per condition (around 15-25) and record movement time and errors.
  4. Compute effective width from endpoint spread and fit MT = a + b·log2(D/We + 1); check R² is above about 0.9.
  5. Calculate throughput and use the fitted model to predict acquisition times for proposed layouts.
  6. Set minimum target sizes and spacing that keep both predicted time and error rate within thresholds (e.g. errors below 4%).

Sources

  • Human-Computer Interaction
  • Statistics for HCI
  • Designing for Touch (Josh Clark)
  • Shneiderman & Plaisant
  • Touch IT
  • Methods Atlas (Gopinath)
  • The Inquiry Series, Ch 3 — The Ballistics of the Human Body — Fitts' Law & Motor Control
  • Garcia-Ruiz, M. (ed.), Games User Research: A Case Study Approach (2016)

Also known as

  • Fitts' Law
  • Fitts' Law calibration
  • Pointing performance modelling
  • Target acquisition modelling

Categories

  • User Research
  • Task Analysis & Cognitive Modelling