Fitts's Law Pointing Experiment

Fitts's Law task

Condition

–

Trial

–

Distance D

–

Width W

–

Index of difficulty

–

Last MT

–

Your fitted model

a (intercept)

–

b (slope)

–

Fit R²

–

Throughput

–

Error rate: –

How this experiment works

Fitts's Law (1954) models the time it takes to move a pointer to a target of width W across a distance D. It says the movement is not governed by distance alone, but by the ratio between distance and target size — the index of difficulty:

ID = log2(D / W + 1)   [bits]
MT = a + b · ID        [seconds]

The circle of targets below is the ISO 9241-9 "multi-directional tapping" task: clicking alternates roughly across the ring, so every trial is a fresh movement in a new direction — no muscle memory from repeating the same stroke. Every click is timed with performance.now() and its exact pixel position is recorded relative to the target center.

Once you finish every distance/width condition, the simulator runs an ordinary-least-squares regression of your own MT against ID to solve for your personal a and b. It then reports throughput using the Shannon formulation of Fitts's Law:

We = 4.133 · SD(endpoint error)
IDe = log2(D / We + 1)
TP = IDe / MT           [bits/second]

We (effective width) is derived from the actual spread of where you clicked, not the nominal target size — it accounts for the speed/accuracy trade-off you personally chose, which is why throughput is considered a more device-comparable metric than raw movement time.

Fitts's Law pointing experiment

You'll click a moving circular target through several rounds of distance/size combinations. Click "Start experiment" to begin — the first click in each round just starts the timer.