Choice Reaction Time Task

Choice reaction time - responding to one of several stimuli

The choice reaction time (CRT) task extends the simple reaction time paradigm by requiring the participant to discriminate among several stimuli and map each to a different response. It measures not just detection speed but the time to identify the stimulus and select the correct action — the core of everyday decision latency.

The Choice Reaction Time Procedure

On each trial, one of several possible stimuli appears (e.g., a left or right arrow, a red or green light, or one of N letters). The participant must press the response key that corresponds to that specific stimulus. The number of alternatives, N, is varied across blocks or experiments.

Performance is summarized by:

  • Mean RT per condition.
  • Error rate (confusions between stimuli/responses).
  • The RT–N relationship — how reaction time increases as the number of choices grows.

Why CRT is more than a speed test

Simple RT asks whether a signal occurred; CRT asks which signal occurred and which response fits it. That extra step — stimulus identification plus response selection — is the component that interests decision researchers. Two individuals with identical simple RTs can differ dramatically on CRT, and those differences track everyday tasks such as driving, sorting, and time-pressured decisions far better than simple RT does.

What Choice Reaction Time Measures

CRT is famous for demonstrating Hick's Law: reaction time increases roughly linearly with the logarithm of the number of equally likely alternatives (RT ≈ a + b·log₂N). This implies that people process choices by eliminating alternatives at a roughly constant rate, as if performing a binary search. In compatible mappings, the slope is shallow; in arbitrary mappings, it is steeper.

The task thus indexes:

  • Decision and selection capacity (not just motor speed).
  • Stimulus–response compatibility — spatially compatible mappings are faster than arbitrary ones (related to the Simon Task).
  • Practice and automatization — the slope shrinks with training.

Typical choice reaction times

For two alternatives (2CRT), healthy adults typically respond in roughly 300–400 ms, compared with about 250–300 ms for simple RT. Adding alternatives increases average RT, but the increments follow the logarithmic pattern of Hick's Law rather than growing linearly. Error rates usually stay low (under 5%) in easy discriminations; when errors climb, researchers often interpret the result as a speed–accuracy trade-off rather than pure slowing. As with simple RT, individual scores vary with attention, age, fatigue, and equipment, so single-session results should be read as a rough personal benchmark, not a diagnosis.

Factors That Shape Choice Reaction Time

  • Number of alternatives (N) — more choices, slower responses, in a roughly log-linear fashion (Hick's Law).
  • Stimulus–response compatibility — pressing a key on the same side as the stimulus is faster than an arbitrary mapping; the Simon Task exploits this for conflict research.
  • Discriminability — similar-looking stimuli are harder to tell apart and produce both slower RTs and more errors.
  • Practice — CRT improves quickly with repetition as mappings become automatized.
  • Age — choice RT slows with age more steeply than simple RT, because it involves more cognitive processing.
  • Fatigue and arousal — as in simple RT, sleepiness slows responses and increases errors, especially on higher-N blocks.

The Error–Speed Trade-off

CRT produces two dependent variables — time and accuracy — and they trade off against each other. If participants are told to be fast, RT drops and errors rise; if told to be accurate, the reverse happens. Researchers therefore rarely report RT alone. A common approach is to interpret CRT performance through both measures together, or to model them jointly. In self-administered online tests, this is why "fast but sloppy" and "slow but careful" results are not directly comparable — a point worth remembering when comparing your own sessions.

Relationship to Other Paradigms

CRT sits between Simple Reaction Time (N = 1) and conflict tasks. When two responses compete, it shades into the Stroop Effect and the Eriksen Flanker Task. The Psychological Refractory Period paradigm pushes CRT further by overlapping two choice tasks. Related inhibition paradigms add a "withhold" requirement on top of a choice baseline:

  • The Go/No-go Task asks participants to respond on go signals and withhold on no-go signals.
  • The Stop Signal Task adds a cancel signal after a response has begun.
  • The Deary–Liewald Reaction Time Task is a modern computerized battery that pairs simple and choice RT with multiple trials for a reliable estimate of processing speed.

Why It Matters

CRT is a cornerstone of human-factors engineering (control-panel design, interface layout) and a classic tool for studying how the brain turns perception into action. Its clean quantitative law makes it a model system for testing theories of information processing. Practical applications include:

  • Interface and product design — predicting how long a menu or control panel takes to search when users face several options.
  • Driving and safety research — brake-response and hazard-response paradigms are choice RT tasks at heart.
  • Cognitive and clinical studies — CRT slowing is used alongside other tasks to study aging, attention, and executive function; results in individual contexts should be interpreted by qualified professionals.

Measuring CRT: What to Watch

If you run a CRT experiment (or an online adaptation), keep these points in mind:

  • Use enough trials per condition — 20–40 per mapping reduces noise.
  • Handle anticipation and lapses explicitly — very fast responses (< 100–150 ms) usually mean anticipation, while very slow ones are lapses; the median is more robust than the mean.
  • Record errors alongside RT — a condition that produces more errors is not being answered more slowly for the same reason.
  • Compare within-person — because baselines differ widely, within-session comparisons across N levels are the most interpretable.

Methodology

Methodology: This topic is part of the experimental workflow covered in Experimentology. For the relevant methods, see Design (number of alternatives as a manipulated independent variable; compatibility as a factor), Measurement (RT and error rate as dependent variables), and Models (Hick's Law as a quantitative prediction to fit).

Related Topics

Implementation

A jsPsych implementation of the choice reaction time task is part of the ExpFactory experiments collection, which provides browser-ready timing and decision paradigms.

Try it Now

Measure your reaction time online - Start with simple reaction time, then compare it with choice reaction time to see decision-making costs

Try It Yourself

References

  • Hick, W. E. (1952). On the rate of gain of information. Quarterly Journal of Experimental Psychology, 4(1), 11-26.

Last updated: 2026-10-20

Menu
Loading