This study looked at whether a laptop can quietly track signs of attention while students do schoolwork. It followed 21 university students with ADHD symptoms, aged 19 to 27. To take part, each student needed four or more ADHD symptoms on a rating scale plus a clinical interview. The authors say this is not an official diagnosis. About 35% had a prior ADHD diagnosis, and 50% had none but suspected ADHD.
Each student did up to ten one-hour sessions in a virtual reality study space, working on their own homework on a laptop. Software logged their keyboard, mouse and window activity. After each session, students rated their own attention, effort and motivation.
The researchers built four measures from the logged data. Two of them tended to move with the students' own ratings. One was how often the keyboard and mouse sat idle. The other was how much time went to homework windows. These links were close to moderate and unlikely to be down to chance. How often students switched between windows was not linked to any rating.
The authors call this a proof-of-concept. They also name its limits. The headset may have made the setting less like real life. No students without ADHD were studied. Only schoolwork was tracked. They say larger studies with a comparison group are needed.
The numbers Stats and what the terms mean
The statistics behind the plain-language claims above.
- da Costa, 2026: link between idle keyboard and mouse time (Activity Gap) and self-rated effort, −0.40 (p ≤ 0.0001).
- da Costa, 2026: link between idle time and self-rated attention, where a lower attention score means stronger focus, 0.40 (p ≤ 0.0001).
- da Costa, 2026: link between idle time and self-rated motivation, −0.31 (p ≤ 0.0001).
- da Costa, 2026: link between time in homework windows (On/Off Task) and self-rated effort, 0.34 (p ≤ 0.0001).
- da Costa, 2026: link between time in homework windows and self-rated attention, −0.30 (p ≤ 0.0001).
- da Costa, 2026: link between time in homework windows and self-rated motivation, 0.25.
- da Costa, 2026: links between window switching and effort, attention and motivation, 0.02, −0.01 and −0.09, none significant.
- da Costa, 2026: share of the spread in self-rated attention explained by idle time plus time in productive windows, adjusted R2 = 0.218.
- da Costa, 2026: share of the spread in self-rated attention explained once an ADHD inattention rating score was added, 29.3%. What the terms mean:
- Repeated measures correlation A number showing how closely two things move together across each person's repeated sessions; here it compares logged activity with the student's own ratings. A minus sign shows the direction of the link, not its size.
- Correlation Two things tend to show up together, which does not show that one brings about the other.
- P-value A number from a statistical test that shows how unusual a result would be if there were really no effect; here results counted only below a stricter cutoff of 1.25% because four measures were tested.
- Adjusted R2 The share of the differences in a rating that a set of measures can account for, adjusted for how many measures were used.
- Self-report People answer questions about themselves, here rating their own attention, effort and motivation after each session.
Source: da Costa M, Chan E, Langberg J, et al.. Towards ecological validity when assessing ADHD symptoms: Patterns in automatically collected, real-world PC activity data. International Journal of Human-Computer Studies (2026). doi:10.1016/j.ijhcs.2025.103724