How Headgame scores your games

The score in one paragraph

Each game gives a raw score, a time, a level or a number of points. Where a public norm exists, that score becomes a percentile, the share of games you beat. Each stat takes the average of your two best sessions of the week, and turns the percentile into a number from 35 to 99, with the formula 35 plus 0.64 times the percentile. Your weekly card mixes three brain stats, reflexes, memory and focus, for 40%, and three discipline stats from your phone down sessions, hours, consistency and streak, for 60%.

Where the norms come from

Reaction time, sequence memory, the chimp test and number memory are compared with Human Benchmark distributions read on 13 October 2026, from 24,084 to 70,940 games each (Human Benchmark, 2026). Lab studies give the reference points, 231 ms for simple reaction time in 1,469 adults (Woods et al., 2015a), 6.52 digits of forward span in 763 adults (Woods et al., 2011).

Public online norms come from volunteers, on every kind of device, without control for cheating. They are a demanding opponent rather than a picture of the population, and every page says so. All norms are on the norms page, with their sample sizes and a CSV file.

Why the card averages the week

One session of less than a minute is noisy. In a study of 219 adults who took brief tests on their smartphones many times, the reliability of the average score reached 0.97 or more, comparable to the lab (Sliwinski et al., 2018). A precise reaction time test, retaken once a week for three weeks, reached 0.83 (Woods et al., 2015b). So the card never trusts one morning, it keeps your two best sessions of the week.

The touchscreen correction

Touchscreens add between 50 and 200 ms of latency depending on the model (Van Patten et al., 2021), and people on phones, Android especially, score slower on reaction tests than people on computers (Passell et al., 2021). The app records the phone model and the refresh rate of the screen, and adjusts reaction scores by type of screen. On this website, the percentile uses the public norm, which mixes devices, and the page tells you so.

Which tests count, and why Stroop and flanker do not

A test only enters the card if its reliability, or that of a close parent task, is known to be good. Simple reaction time 0.83 (Woods et al., 2015b), go/no-go commission errors 0.76 (Hedge et al., 2018), Trail Making A, parent of the Schulte grid, 0.87 (Woods et al., 2015c), Corsi block span, parent of sequence memory, 0.85 (Morawietz et al., 2024).

The Stroop and flanker costs are robust effects with weak individual reliability, 0.60 to 0.66 for Stroop and 0.40 to 0.57 for flanker (Hedge et al., 2018). They stay free games. Season games such as the F1 start, number memory and the flanker join the card only after their reliability has been measured in beta.

The practice effect

Scores rise when a test is repeated, partly because people learn the test (Scharfen et al., 2018a). For working memory tests, the gain was about a quarter of a standard deviation from the first to the second session, and kept growing up to the fourth (Scharfen et al., 2018b). This is why the first sessions in the app are a warm-up, so the card does not reward simple familiarity.

What practice does

The largest review of brain training found extensive evidence that people get better at the tasks they practise, and to a lesser degree at closely related tasks (Simons et al., 2016). Headgame does not claim more. Each game asks for one skill, reaction speed, working memory or focus, and your card shows how your scores in that skill move from week to week, next to the time you take back from scrolling.

Who reviewed this method

The method was written by the Headgame team from the sources listed below. No independent scientific reviewer has been appointed yet. When one is, their name, title and institution will appear here and on every page they have read. Until then, every figure links to its source so anyone can check it, and errors can be reported to contact@brainovr.app.

The Headgame games →

Sources

1. Sliwinski M.J., Mogle J.A., Hyun J. et al. (2018). Reliability and validity of ambulatory cognitive assessments. Assessment 25(1), 14 to 30. PMC5690878. doi.org/10.1177/1073191116643164
2. Human Benchmark, score distributions embedded in humanbenchmark.com, read on 13 October 2026. Reaction time n 70 940, sequence memory n 35 868, chimp test n 27 233, number memory n 24 084. humanbenchmark.com
3. Woods D.L., Wyma J.M., Yund E.W., Herron T.J., Reed B. (2015). Factors influencing the latency of simple reaction time. Frontiers in Human Neuroscience 9, 131. doi.org/10.3389/fnhum.2015.00131
4. Woods D.L., Kishiyama M.M., Yund E.W. et al. (2011). Improving digit span assessment of short-term verbal memory. Journal of Clinical and Experimental Neuropsychology 33(1), 101 to 111. PMC2978794. doi.org/10.1080/13803395.2010.493149
5. Woods D.L., Wyma J.M., Yund E.W., Herron T.J. (2015). The effects of repeated testing, simulated malingering, and traumatic brain injury on high-precision measures of simple visual reaction time. Frontiers in Human Neuroscience 9, 540. PMC4637414. doi.org/10.3389/fnhum.2015.00540
6. Van Patten R., Iverson G.L., Muzeau M.A., VanRavenhorst-Bell H.A. (2021). Test-retest reliability and reliable change estimates for four mobile cognitive tests administered virtually in community-dwelling adults. Frontiers in Psychology 12, 734947. PMC8566921. doi.org/10.3389/fpsyg.2021.734947
7. Passell E., Strong R.W., Rutter L.A. et al. (2021). Cognitive test scores vary with choice of personal digital device. Behavior Research Methods 53(6), 2544 to 2557. PMC8568735. doi.org/10.3758/s13428-021-01597-3
8. Hedge C., Powell G., Sumner P. (2018). The reliability paradox. Behavior Research Methods 50(3), 1166 to 1186. PMC5990556. doi.org/10.3758/s13428-017-0935-1
9. Woods D.L., Wyma J.M., Herron T.J., Yund E.W. (2015). The effects of aging, malingering, and traumatic brain injury on computerized Trail-Making Test performance. PLoS One 10(6), e0124345. PMC4465490. doi.org/10.1371/journal.pone.0124345
10. Morawietz C. et al. (2024). Consistency of spatial ability performance in children, adolescents, and young adults. Frontiers in Psychology 15, 1365941. PMC10938598. doi.org/10.3389/fpsyg.2024.1365941
11. Scharfen J., Peters J.M., Holling H. (2018). Retest effects in cognitive ability tests, a meta-analysis. Intelligence 67, 44 to 66. doi.org/10.1016/j.intell.2018.01.003
12. Scharfen J., Jansen K., Holling H. (2018). Retest effects in working memory capacity tests, a meta-analysis. Psychonomic Bulletin and Review 25, 2175 to 2199. doi.org/10.3758/s13423-018-1461-6
13. Simons D.J., Boot W.R., Charness N. et al. (2016). Do brain training programs work? Psychological Science in the Public Interest 17(3), 103 to 186. doi.org/10.1177/1529100616661983