EEG in novices learning archery: Rampp 2022 study

Eleven adults with no archery experience at all, all right-handed and right-eye dominant, shot 100 arrows a day on three days in the same week with a 14 lb recurve bow at 8 metres, under the guidance of an instructor. During every shot a 64-channel EEG recorded brain activity.

Scores rose: the mean per arrow improved by 1.18 points between the first and third days, the only significant difference between days, and the gain within a session, from the first 10 arrows to the last 10, was 2.3 points on the first day and 0.5 and 0.6 on the following two. One participant got worse on the third day because of muscle fatigue.

In the half second ending 200 milliseconds before impact, the authors measured, region by region, which hemisphere showed stronger activity in the theta, alpha and beta bands. A shift of theta activity towards the left hemisphere, compared with the first day, went with a better score in 7 regions, mostly central and parietal, with the strongest link in the supplementary motor area (coefficient 10.36) and the paracentral lobule (7.03). In the preceding half second, at rest and in the alpha and beta bands, no link survived correction for multiple comparisons. Power itself was not linked to score in any band and did not change significantly across the three days.

Sessions were divided into ends of 10 arrows, with breaks every 10 to 30 shots. To make the signal analysable, each shot followed phases called out by voice: bow down while looking at the target, bow raised without drawing, then draw and aim, with the moment of release left to the archer. Scores ran from 5 to 11, with the X counted as 11. The impact of each arrow was located on an audio recording synchronised with the EEG.

For each arrow the authors took three half-second windows: one ending 200 milliseconds before impact, to exclude the release and the recoil, one immediately before it and one during the rest phase. The signal was projected onto the cortex with a standard head model, since no individual MRI scans were available, and summarised by anatomical region. For each pair of mirror-image regions a laterality index measures how far activity leans to the left or right; its change from the first day was related to the improvement in score with a mixed model corrected for multiple comparisons. The coefficient reads as the estimated extra points for a complete shift from balanced activity to activity entirely on the left.

Eleven people over three days is a small study, as the authors admit, and there is no comparison group: lateralisation is related to improvement within the same group, so the result is an association, not evidence that the leftward shift produces the improvement or the other way round. The fact that the link does not appear at rest strengthens the idea that it concerns aiming, but does not prove it. One participant's session was lost through a technical problem.

The confidence intervals are wide (from 6.16 to 14.56 points for the supplementary motor area), and without individual MRI scans the location of the areas is approximate, as the authors themselves say. The shooting was also very far from an archer's: complete beginners, 8 metres, 14 lb, phases set by spoken commands. Some sentences in the discussion do not match the results: it says improvement was greatest between the first and second days, whereas the only significant difference is between the first and third, and that beta activity increased with practice, whereas power did not change between days. Finally, the means for the first 10 arrows, as low as 1.4, are incompatible with a minimum score of 5: they suggest that arrows missing the target counted as zero, which the paper does not state.

For anyone who shoots, the study suggests no change to technique or training. It describes what happens in a beginner's brain over the first three days of shooting: as the score rises, theta activity in the last half second before release shifts towards the left hemisphere, and those who shift more improve more. The authors interpret this as the motor system making the timing of movement control shorter and more consistent, but that is their hypothesis.

The most concrete finding concerns how a beginner learns: a great deal at first, with 2.3 points gained within the first session, and much less in the sessions that follow. The study says nothing about experienced archers, long distances or what happens after the first week, and it offers no measure that could be used in training.