Skip to main content
Category: Laws
Type: Cognitive / psychophysical regularity (choice reaction time)
Origin: William E. Hick, Quarterly Journal of Experimental Psychology, 1952
Also known as: Hick–Hyman law (related formulation)
Quick Answer — Hick’s Law says that, in simple choice tasks, reaction time grows with the logarithm of the number of equally likely alternatives—not linearly with raw option count. It helps explain menu design, emergency controls, and any interface where scanning and deciding compete with acting. It is a tendency across conditions, not a single universal constant for every real-world decision.

​
What is Hick’s Law?

Hick’s Law describes a robust pattern in choice reaction time: when stimuli are equally probable and the task is cleanly defined, mean reaction time increases roughly in proportion to the logarithm of the number of alternatives—often written with a base-2 log form because information is measured in bits. The relationship connects cognitive psychology to Weber–Fechner scaling intuitions (perception compresses stimulus ratios) and complements Yerkes–Dodson concerns about arousal and performance. It differs from choice overload in emphasis: Hick focuses on timed discrimination among labeled options, while overload often includes preference formation and regret.
More forks on the road do not add equal seconds—complexity compounds gently at first, then demands structure.

​
Hick’s Law in 3 Depths

  • Beginner: Expect visibly slower taps when a screen adds a few equally plausible buttons; shrinking counts helps more than “bigger font.”
  • Practitioner: Chunk options into progressive steps or defaults so each step keeps n small; measure task time, not opinion polls about “simplicity.”
  • Advanced: Recognize domain limits—semantic search, expertise, and unequal priors break the tidy equal-probability setup.

​
Origin

William Edmund Hick published “On the Rate of Gain of Information” in 1952 in the Quarterly Journal of Experimental Psychology, reporting choice-reaction experiments (including setups with 10 response alternatives arranged around a subject). The paper helped crystallize the idea that humans process choice information at a roughly steady bits-per-second rate in those paradigms. Ray Hyman reported related linear relations between reaction time and transmitted information in 1953, which is why the pairing is often called Hick–Hyman in textbooks.

​
Key Points

Use Hick to budget attention in timed tasks, not to excuse bad information architecture.
1

Log growth, not linear growth

Doubling alternatives does not double time in these models—yet zero shortcuts still vanish if n explodes.
2

Equal likelihood matters

The cleanest fits assume alternatives are equiprobable and clearly labeled; skewed priors change optimal strategies.
3

Motor + cognitive stack

Observed latency includes perception, memory retrieval, and movement—Hick isolates a slice, not the whole product experience.
4

Defaults and classification shrink effective *n*

Good IA moves decisions from one giant menu to several small, meaningful choices.

​
Applications

Translate the law into measurable design moves.

Interfaces & products

Reduce parallel top-level actions; use progressive disclosure and strong defaults so each step’s option count stays small.

Safety & operations

In high-stress consoles, match control layout to muscle memory and limit simultaneous decisions—Hick adds latency where seconds matter.

Education & testing

Multiple-choice length affects speed and error; fair comparisons hold item difficulty constant when changing option counts.

Organizations

Approval chains multiply alternatives—clarify decision rights so individuals face fewer simultaneous competing moves.

​
Case Study

Hick’s empirical anchor is his 1952 paper’s controlled laboratory setting: participants responded to one of 10 alternative stimuli in a choice reaction paradigm, enabling quantitative fits between number of alternatives and reaction time distributions. That work is why modern HCI references cite Hick when arguing that shaving a menu from many peer options to fewer meaningful branches can reduce mean selection latency—an effect you can approximate in usability tests with median task-time and error-rate metrics, even though office software rarely matches equiprobable lab conditions perfectly.

​
Boundaries and Failure Modes

Boundary 1: Experts compress menus
Practice and chunking change effective n; experts are not bound by novices’ curves. Boundary 2: Preference is not discrimination
Choosing what you want involves taste and tradeoffs beyond simple reaction tasks. Common misuse: Shrinking options by hiding critical functions—latency drops while failures rise.

​
Common Misconceptions

Keep the law’s laboratory roots visible.
Reality: Search, filters, and recommendations change the task; Hick targets simultaneous equally likely choices.
Reality: Input modality, target size (see Fitts-style constraints), and fatigue shift measured times.
Reality: Rushed selections raise errors—pair time with accuracy and safety margins.
Connect these when tuning decisions under time pressure.

Yerkes–Dodson Law

Arousal and performance follow an inverted-U—stress interacts with Hick-style latency.

Weber–Fechner Law

Perception often scales logarithmically—cousin intuition for compressed stimulus scales.

Choice Overload

Beyond reaction time—preference, satisfaction, and regret when assortments grow.

​
One-Line Takeaway

Cut effective alternatives per step—defaults, grouping, and search—before polishing pixels.