A multisession evaluation of an adaptive competitive arm rehabilitation game.

Only four sessions were tested, so the paper cannot show that engagement would persist through a full rehabilitation program. The study measured enjoyment, perceived competence, movement intensity, and session duration, not arm function or clinical recovery. The alone session used a computer-controlled opponent, which may have behaved differently from a human opponent. All participants experienced three competitive sessions before the alone session, which may have biased ratings and duration. Many outcomes came from questionnaires, and conversation was rated by the experimenter without consulting participants, so it was likely biased. The home group was self-selected and had close relationships with opponents, more computer game experience, and preferred more difficult games than the clinical group. The clinical group had shorter time since injury, was in inpatient rehabilitation, and participants in pairs were not truly independent for analysis. Cognitive state was measured only in the clinical group, so differences in cognition between groups cannot be confirmed.

The paper names the National Institute of General Medical Sciences as funder.

Do not conclude that competitive games improve arm function. This paper measured enjoyment and movement intensity over four sessions, not clinical recovery, and the clinic group showed no advantage over playing alone.