Sustained Attention to Response Task (SART)
Follow a sequence of numbers and explore sustained attention and response inhibition in this session. This is an original browser implementation without population norms or diagnostic validation.
- About 5 minutes
- Physical keyboard
- Practice included
- No sign-up
Press Space once for every number except 3. When 3 appears, do not press anything. Give equal importance to speed and accuracy. You can respond after the number disappears, until the next number appears.
18 practice rounds, then 225 scored rounds. Practice is not scored.
A physical keyboard is required for this version. The questionnaire works on touchscreens; these two tasks are currently designed for computer keyboards.
Practice round
Press Space for each number except 3.
Practice complete
This round was stopped.
The page lost focus, became hidden, or the round was stopped. We have discarded this round so an interruption is not counted as an attention lapse. Start again when you are ready.
Your results
Sustained Attention to Response Task (SART) · Performance and data quality are shown separately.
Understand the task
How this online SART works
The Sustained Attention to Response Task (SART) is a go/no-go task introduced in 1997. Digits 1–9 appear one at a time; you press Space for the frequent non-target digits and withhold your response when 3 appears.
What this task samples
It samples sustained attention and response inhibition while one simple rule repeats. This short session cannot represent every part of attention in everyday life.
What appears in your result
Your report shows accuracy, omissions, commission errors, median response time, and response-time variation for usable trials. Read speed and accuracy together.
How to read a SART result
- Omissions are missed responses to digits other than 3.
- Commission errors are responses made when 3 appears.
- Median response time describes the middle timing value among correct, usable responses.
- Response-time variation describes how much those correct responses differed within this session.
There is no universal “good” online SART score. Procedures and reference groups differ, and this implementation has no population norms. Use the measures as a description of this session.
Method and quality checks
18 practice rounds, then 225 scored rounds. Practice is not scored. Responses under 150 ms are marked early. Early and timing-affected trials are excluded from accuracy/rate denominators and response-time summaries. Counts refer only to usable trials unless marked otherwise.
FixMind’s original implementation of a published experimental paradigm. Our code, instructions and geometric stimuli are original. This is not the NIH Toolbox, TOVA, or another commercial test, and it does not use their norms.
Before you begin
Use a computer and a physical keyboard in a quiet setting. Close distractions and keep this tab in view. Sleep, fatigue, stress and device timing can affect your experience.
Privacy and data
Your answers and results stay in this page and are not uploaded or saved automatically. Leaving or refreshing clears your progress. This site uses Cloudflare Web Analytics for basic traffic and performance measurements; assessment answers and scores are not included.
Questions about the SART test
Can a SART test diagnose ADHD?
No. A meta-analysis of go/no-go tasks, including the SART, found that performance differences were not sensitive or specific enough to serve as a stand-alone diagnostic measure. FixMind also has no population norms or diagnostic validation.
What is the difference between an omission and a commission error?
An omission is a missed Space press for a digit other than 3. A commission error is a Space press made for 3, when the rule asks you to withhold the response.
Does a faster response time mean better attention?
Not by itself. Responding faster can come with more errors, so timing should be read alongside accuracy, omissions, and commission errors.
What can affect an online SART result?
Sleep, fatigue, stress, distractions, understanding the rule, keyboard or browser timing, and interruptions can all affect a single session.
How long does this SART take?
Allow about five minutes after the short practice. The scored task contains 225 trials, including 25 appearances of the digit 3.
Research and implementation
- Robertson et al. (1997), the original SART study ↗
- Wright et al. (2014), meta-analysis of response-inhibition tasks and diagnostic limits ↗
FixMind built this browser task from the published paradigm. The code, instructions, quality rules, and report are our own and have not been clinically validated.