Attention
Road Hazard Task
Also called: hazard perception task, hazard perception test, driving hazard detection
Measure how quickly people notice developing hazards in drawn road scenes, with response time, misses and extra presses.
5-7 minTimed: desktop or laptop with a mouse or trackpad
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 84
- Evidence grade
- C
- Registered clinical studies
- 5
What it measures
Measures how quickly people notice developing road hazards. Participants watch drawn scenes from the driver's seat and press the space bar when a road user moves into their path. Scores response time, missed hazards and extra presses.
Constructs
- Hazard perception
- Selective attention
- Response timing
Research uses
- Visual attention research
- Road-scene perception studies
- Within-study group comparison
How the paradigm works
The Road Hazard Task measures how quickly people notice a developing road hazard. Participants watch original drawn scenes from the driver's seat and press the space bar when a pedestrian, car or cyclist starts to enter their path. Each scene continues after a press and ends without a crash. The hazard-perception method follows published studies using timed responses to road scenes, including computer animations (Crundall and colleagues, 2021).
More about the paradigm
The standard session has 4 practice scenes with feedback, followed by 24 scenes in two sets. Each set has 3 pedestrian hazards, 3 car hazards, 3 cyclist hazards and 3 safe scenes, in seeded order. Hazards begin 3 to 6 seconds into a scene, with a 2,500 ms response window and one further second before the scene ends. Studies can choose 1 to 3 sets. The scene pauses and blanks if the window loses focus. Desktop or laptop with a keyboard.
Scores
- Hazard Response Timems
- Median first response from the displayed hazard onset, at or after 100 ms and within 2,500 ms. Missed hazards are excluded.
- Lower is faster; read with hit rate
- Hazards Detected%
- Hazards with a first response between 100 and 2,500 ms after onset, divided by all hazard scenes.
- Higher means more hazards detected
- Missed Hazardscount
- Hazards without a qualifying first response.
- Lower means fewer misses
- Pedestrian Hazards Detected%
- Share of pedestrian hazards detected within the response window.
- Higher means more detected
- Car Hazards Detected%
- Share of car hazards detected within the response window.
- Higher means more detected
- Cyclist Hazards Detected%
- Share of cyclist hazards detected within the response window.
- Higher means more detected
- Out-of-Window Pressescount
- Presses in safe scenes or outside the hazard window.
- Lower means fewer extra responses
- Safe Scenes with a Press%
- Share of safe scenes with any space-bar response.
- Lower means fewer false alarms
- Repeated Hazard Pressescount
- Presses after the first response within a hazard window.
- Informational
- Anticipationscount
- First responses under 100 ms after hazard onset.
- Lower supports interpretation
- Pausescount
- Scene interruptions after focus loss, hiding the tab or resizing.
- Lower supports interpretation
- Long Frames%
- Share of animation frames longer than 50 ms.
- Above 0.05 is flagged
- Completed Scenescount
- Completed scenes in the scored run.
- Informational
- Planned Scenescount
- Scenes planned for the run.
- Informational
Use and evidence
84 papers from 2021 to 2025 name this task in their title or abstract, up 22% compared with 2016 to 2020. That ranks it 267 of 582 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 25 |
| 2024 | 14 |
| 2023 | 9 |
| 2022 | 13 |
| 2021 | 23 |
| 2020 | 13 |
| 2019 | 12 |
| 2018 | 10 |
| 2017 | 14 |
| 2016 | 20 |
| 2015 | 13 |
| 2014 | 16 |
| 2013 | 8 |
| 2012 | 7 |
| 2011 | 12 |
| 2010 | 9 |
| 2009 | 12 |
| 2008 | 5 |
| 2007 | 1 |
| 2006 | 1 |
| 2005 | 2 |
| 2004 | 2 |
| 2003 | 1 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 0 |
Use grade D
30 to 99 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 16
- Registered clinical studies using it as an outcome
- 5
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
Note on the counts: includes hazard perception tests for train drivers and work settings.
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.39.0, counted 10 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Scenes
- Original first-person road drawings, with a pedestrian, car and cyclist. No filmed clips or crashes.
- Trial mix
- 12 scenes per set: 9 hazards, 3 of each road user, and 3 safe scenes. Standard 2 sets (24 scenes), with 1 to 3 sets available.
- Timing
- Hazard onset at 3,000 to 6,000 ms, a 2,500 ms response window, then 1,000 ms until the scene ends. Times and order are set from the session seed.
- Practice
- 4 practice scenes with feedback: one of each hazard and one safe scene.
- Response
- Press the space bar once when the hazard develops. Presses do not change the scene. The first response in the window determines detection and timing; presses under 100 ms are anticipations.
- Scores
- Median hazard response time, hit rate, misses, detection by road user, out-of-window presses and frame timing. Misses remain misses and are not replaced with response times.
Our simplified drawings, scene timings and item set differ from the published animations. Scores describe performance in these scenes and do not establish fitness to drive. Compare response time with misses and false alarms. No percentile output.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: complete scenes from the scored run. Misses and presses outside hazard windows are reported separately. Incomplete runs, frequent extra presses, pauses and long frames are flagged; responses under 100 ms follow the timed-test rule.
Why these rules
The scenes measure responses to developing hazards. There is no fixed chance level for pressing during an animation.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Sets of scenes | 2 | 1 to 3 |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Timed test: participants need a desktop or laptop with a mouse or trackpad. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- No percentile. Scores and quality flags only. How norms work
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Road Hazard Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Crundall, van Loon, Baguley and Kroll, 2021), with the standard settings (sets of scenes: 2). The main score was hazard response time (median first response from the displayed hazard onset, at or after 100 ms and within 2,500 ms).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Crundall, D., van Loon, E., Baguley, T., & Kroll, V. (2021). A novel driving assessment combining hazard perception, hazard prediction and theory questions. Accident Analysis & Prevention, 149, 105847. https://doi.org/10.1016/j.aap.2020.105847Computer-animated first-person scenes and hazard-window detection; our drawings and scoring are original.
Source papers
- Crundall, D., van Loon, E., Baguley, T., & Kroll, V. (2021). A novel driving assessment combining hazard perception, hazard prediction and theory questions. Accident Analysis & Prevention, 149, 105847.Paradigm source, Adults
Questions
How long does the Road Hazard Task take?
About 5 to 7 minutes, as listed in the test library.
What device does a participant need?
Timed test: participants need a desktop or laptop with a mouse or trackpad. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Hazard Response Time, Hazards Detected, Missed Hazards, Pedestrian Hazards Detected, Car Hazards Detected, Cyclist Hazards Detected, Out-of-Window Presses, Safe Scenes with a Press, Repeated Hazard Presses, Anticipations, Pauses, Long Frames, Completed Scenes and Planned Scenes. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Road Hazard Task used in research?
84 papers from 2021 to 2025 name it in their title or abstract, up 22% compared with 2016 to 2020. That ranks it 267 of 582 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.39.0, counted 10 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Road Hazard Task have?
In the published research, 0 papers report its test-retest reliability and 5 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Sets of scenes (1 to 3). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.