Fitts

Fitts’s law predicts movement time from the distance to a target and the target’s effective width. A farther target generally takes longer to reach, while a wider target is easier to acquire because it permits more endpoint error. The original formulation expressed this relationship as MT = a + b log₂(2D/W), where MT is movement time, D is the distance from the starting position to the target, W is target width along the movement axis, and a and b are empirically estimated constants.¹

The logarithmic term is called the index of difficulty. It treats movement as an information-processing task: increasing distance or reducing width raises the number of distinctions the performer must resolve. The law describes a statistical tendency rather than an exact limit, and performance varies with practice, posture, device, visual conditions, and the definition of the target.

Paul Morris Fitts developed the law while studying human motor control at Ohio State University. His 1954 paper, “The Information Capacity of the Human Motor System,” compared rapid reciprocal movements between targets of different widths and separations.¹ Fitts found a near-linear relation between movement time and the logarithm of the movement’s spatial difficulty, linking motor behavior to the language of information theory.

The work was part of a broader postwar research program concerned with communication, control, and skilled performance. Fitts later contributed to aviation psychology and experimental methods, and the Fitts–Welford model became one of several formulations developed to account for movement conditions that the original equation treated imperfectly. The law’s lasting value lies less in a universal constant than in its compact, testable prediction.

Fitts’s law helps designers evaluate how quickly people can point to buttons, menu items, links, controls, and other selectable targets. Interfaces can reduce acquisition time by enlarging important targets, placing related controls near the starting region, and using the edges and corners of a screen, where the pointer can stop against a physical boundary. The latter observation is especially useful for menus and edge-mounted controls because the effective target can extend beyond the visible region.²

The model informs graphical user interfaces, touchscreens, game controls, digital instruments, and accessibility design. It does not by itself determine whether a control is understandable, visually discoverable, safe, or comfortable. A large target may still be poor if it is ambiguous, crowded, difficult to reach, or likely to trigger an unwanted action. Designers therefore combine Fitts-based measurements with usability testing and accessibility guidance, including recommendations for adequate pointer and touch-target size.³

Fitts’s law is strongest for rapid, visually guided aiming movements, but its parameters change when the task changes. Touch input introduces finger occlusion, contact-size uncertainty, and differences between the visual target and the physical region that accepts a touch. Two-dimensional pointing also requires choices about how target width is measured; the effective width approach estimates the spread of observed endpoints rather than assuming a perfectly controlled rectangle.

Researchers have extended or modified the law for serial pointing, steering through constrained paths, discrete key selection, and movements involving cognition or corrective submovements. The related steering law, for example, predicts the time needed to guide a pointer through a path whose width varies along its length.⁴ Fitts’s law also has boundary cases: very large targets can produce ceiling effects, extremely small targets can invite corrective movements, and novice users may spend more time interpreting a control than physically acquiring it.

“Fitts” is a surname, while the possessive form “Fitts’s law” identifies the principle associated with Paul Fitts. The index of difficulty is commonly measured in bits, reflecting the logarithm’s information-theoretic interpretation, but the value should not be mistaken for a literal count of neural decisions. The slope and intercept are properties of a particular participant, device, task, and experimental setup.

Modern human–computer interaction treats the law as a design heuristic and empirical model. It is most useful when comparing alternatives under controlled conditions: a larger target, a shorter movement, or an edge location can often be expected to improve acquisition speed, but the magnitude must be measured in the intended context. This cautious interpretation preserves the law’s practical strength without treating it as a complete theory of skilled action.

Index of difficulty

A logarithmic measure of the spatial challenge of a pointing movement, commonly written as log₂(2D/W) in the classic formulation.

Effective width

A width estimated from the distribution of movement endpoints, often used when actual performance does not match an ideal geometric target.

Steering law

A model predicting the time required to guide a pointer through a constrained path, with narrower sections increasing difficulty.

Aiming movement

A movement directed toward a spatial target, usually ending when the target is acquired or selected.