EEG and eye tracking in archers: Pei 2022 dataset

This paper does not report an experiment with a result to discuss: it describes a collection of data made available to other researchers. Seventy-four people, 31 athletes from Shanghai and 43 university students with no competitive background, carried out a series of attention tasks on a computer while their brain activity (64-channel EEG) and eye movements (2,000 samples per second) were recorded together. Six of them are archers: three men and three women, three at National Master level and three at national first level. The other athletes are ten pistol and rifle shooters and fifteen modern pentathletes.

The few analyses reported are there to show that the data are usable. The ratio of alpha to beta waves, which the authors use as an index of attention (the lower it is, the more attention), was higher in the combined shooting and archery group than in the controls: 2.52 against 2.33 and 2.44 while fixating the target, 3.35 against 2.74 and 3.00 with eyes closed. The authors expected the opposite and explain it afterwards, suggesting that these athletes need less effort for the same task. The other analyses, the P300 wave in the vigilance task and eye movements with and without distraction, concern pentathletes and students, not archers.

The athletes came from Shanghai's competitive sports training centre and the controls from two universities. Before testing, everyone completed three questionnaires: sleep quality (PSQI), mood (POMS) and pre-competition anxiety (CSAI-2), which for the students became anxiety before an exam. Seated in a soundproof room, with the chin on a rest 80 centimetres from a 32-inch screen, they did two experiments.

The first, lasting 10 minutes, is a version of the Mackworth clock test: a red dot moves round a circle of 60 dots and a key has to be pressed whenever it skips one. The second, lasting 70 to 90 minutes, alternates rest with eyes closed and open; fixation for 2 and then 3 minutes on the centre of a scaled-down 10 metre air pistol target; fixation with visual distractors (a circle appearing at 24 positions around the target) or music (a pop song and a piano piece); and two short interventions, 5 minutes of meditation imagining the centre of the target between the eyebrows and a Schulte grid on an iPad. In the fixation tasks attention is estimated from how far the gaze strays from the centre of the target.

For archery the sample is close to token: six archers out of 74 participants, and in the analyses they are always pooled with the shooters. The target on the screen is an air pistol target, the tasks are done seated with the head still, and nobody shoots: what is measured is the ability to hold the gaze on a point, not aiming with a bow in hand, which lasts a few seconds under load and is done with one eye or both depending on the archer.

The paper is not a comparative study in any case. The brainwave data are group means with no statistical tests, and the result that ran against expectations is explained after the event. The P300 comparison uses 15 pentathletes and 15 students picked at random, and the eye movement example is a single pentathlete against a single student. The idea that a gaze closer to the centre means more concentration is also a working assumption of the authors, not something the study tests. The value of the work lies in the public data, which others can analyse.

For archers the paper changes nothing for now: it relates no measure to score, it does not compare stronger and weaker archers, and the six archers in it are never analysed on their own. The higher alpha to beta ratio in the shooting and archery athletes, the opposite of what the authors expected, fits the idea discussed in other studies that aiming demands less of the brain in experts; here, though, it remains a difference between means with no statistical test behind it.

In the registry the study counts for another reason: it is one of the few public collections of EEG and eye movement data recorded together in athletes from precision sports, national-level archers included. Anyone wanting to study how archers hold their gaze on the target has raw data ready to use, with a profile of each athlete: level, years of training and training load.