Important
This application was developed over 2 days and then deployed for 5 days for Unit 4 Specialist Mathematics SAC (School Assessed Coursework).
A simple Flask and sqlite3 application to collect human reaction time data with built-in live comparison against your normal distribution and collected sample distribution.
Important
Thank you to the 52 Participants who took part in the survey over 5 days, generating 391 datapoints. Your contributions are greatly appreciated.
The raw dataset is available here (last updated 2025/09/07)
Note
- Personal reaction time distribution with mean, standard deviation, and trial count.
- Comparison to the collected sample distribution within the set.
- Percentile ranking against the global distribution.
According to (Human Benchmark, 2007) the median reaction time is 273 ms across their dataset. According to their about the test section, computer differences and latency can add as much as 150 ms.
The standard deviation was calculated to be 140.92 ms while the median was calculated to be 373 ms across the 391 datapoints with 52 unique users. The range was 961 ms with a minimum of 4 ms and maximum of 965 ms with the test boundaries being 0 ms to 1000 ms.
Interestingly, the difference between the Human Benchmark dataset and this dataset is 100 ms, which may indicate a systematic bias introduced by the CSS transition applied during the visual stimulus
transition: background-color 0.3s, transform 0.2s;Note
This line (49) has been commented out. Create a new sqlite3 database for future studies, as data collected before 2025/10/23 may contain a systematic measurement bias.
pip install -r requirements.txt
python app.py
Then open http://127.0.0.1:8000 in your browser
- Human Benchmark. (2007). Reaction Time Test. Humanbenchmark.com; Human Benchmark. https://humanbenchmark.com/tests/reactiontime