Cognitive and attention tests in archery studies

One word, many instruments

In archery studies "concentration" covers different instruments: the time taken to press a button when a light comes on, the symbols crossed out on a sheet in a set time, the numbers found in order on a grid in sixty seconds, the delay a distractor adds to a response on a computer. These are performance tests, in which the archer performs a task and is scored on it. A questionnaire has its own methods page, and the "concentration indices" calculated by a commercial EEG headband, as in a 2023 Chinese study, are not tests. Findings in chronological order are in the line on attention and cognitive function; this page looks at the instruments.

Reaction time

This is the oldest measure. In the large American study of 1986 seated archers pressed a button when a light came on: 194.8 milliseconds in the weakest band, 173.5 in the strongest. The same study measured anticipation, stopping a moving light at the moment it reaches the end of its track, and across 127 archers the link with score was weak, r = -0.24 (the entry on anticipation timing). Later comparisons find no clear advantage for archers: 318 milliseconds in Japanese university archers against 321 in kendo practitioners, in a study mainly concerned with blinking, and 0.134 seconds on the ruler-drop test in young Turkish archers against 0.142 in pétanque players. The entry on reaction times is "reported".

The weakness of this measure is the noise within each person: in the fifteen young Ukrainian archers of a 2014 study the individual coefficient of variation ranged from 12% to 57%, sometimes more than the gap between groups. Nor should it be confused with the reaction to the clicker, which is measured with a sensor on the bow during real shooting: the entry on reaction time and skill level gathers both, and they need reading separately.

Pen-and-paper tests

Pen-and-paper tests are the largest group. In the Bourdon test and the d2, certain symbols are crossed out against the clock among others that look alike; in concentration grids the numbers are found in order on a hundred-square table; in Schulte tables and red-and-black tables scattered numbers are counted. They are cheap, and since 2020 they have been the most widely used: on the sixty-second grid of a 2026 Indonesian study forty-five juniors found between 14 and 29 numbers, with a correlation of 0.70 with their score over 36 arrows. The entry on concentration, attention tests and score is "contested".

The d2 yields several indices from the same sheet, and they do not say the same thing. In the forty-five traditional Turkish archers of 2025 the corrected total was related to score, r = 0.56 in the women and 0.47 in the men, while concentration performance, errors and the fluctuation rate were related to nothing: anyone who picks the index after seeing the data stands a good chance of finding something. Age matters too. In the forty children in Konya the Bourdon gave 55.1 at nine to eleven and 82.7 at twelve to fourteen before the course began: age alone was worth more than the roughly 25 points gained over four weeks (the entry on age and attention in young archers). So does the person giving the test: in a 2022 beginners' course d2 concentration started at 37, when in adults it normally lies between 150 and 180, and the test was given by the course coach.

Computer tests

On a computer, tests can separate components that blur together on paper. The attention network test distinguishes alerting, orienting from a cue and the cost of a conflicting distractor: 62 athletes from the Chinese national squad in archery and shooting responded in 508 milliseconds against 537 for provincial athletes, but the three networks worked the same way in both groups (the entry on attention and level is "contested"). On the Stroop test, which asks for the ink colour of a word naming a different colour, speed can mislead: in 2026 thirty-eight Chinese archers had an interference score of 42.8 against 48.1 for sedentary people because they completed fewer items, while making no more errors. Working memory is tested with tasks such as the N-back or the Operation Span, and in thirty-six Japanese kyudo archers the correlation with hits was -0.05 (the entry on working memory); in a Korean brain-imaging study twenty women archers did the visuospatial task as well as women who do not shoot, 47.9% against 40.6% (p = 0.29), but with more active frontal areas (the entry on frontal activation). A standardised test is also read against its norms: in a 2022 study of Indian children aged eleven to thirteen, the groups from all six sports sat well below the fiftieth percentile of the Vienna Test System, which suggests norms built for adults.

Tasks closer to aiming

In an Irish thesis from 2016 twenty-nine recurve archers bisected lines on paper with leftward errors of 0.26 millimetres against 1.76 for non-archers; in the computer version, however, the thesis's table and text say opposite things, and that result does not count (the entry on spatial attention is "contested"). In the joystick task of Gonzalez and colleagues, Olympic and club archers made fewer errors than non-archers but did not differ from each other (p > 0.999), so the test separates archers from non-archers, not levels (the entry on aiming without a bow). NeuroTracker is both a test and a form of training: in twenty Indonesian archers about 32 minutes of the task in front of a 3D television took the grid score from 11.2 to 19.7, against 15.6 for their teammates, and the score at 30 metres rose by 32.9 points against 19.7, in a questionable venue (the entry on perceptual-cognitive training).

The test that improves on a second go

All these instruments share one limitation: retaken a few weeks later, an attention test improves even if nothing has been done in between. In the randomised trial by Hussain and colleagues the inactive controls went from 910 to 811 milliseconds on the Stroop, no differently from those who had shot for six weeks; on the waiting list of the randomised trial of yoga mean response time fell from 630.9 to 614.1 milliseconds, while the network indices stayed put, and it was on two of them, orienting and conflict, that yoga made the difference. A before-and-after study without controls, such as the one on twenty-three Taiwanese archers by Wu and colleagues, who were faster on the Stroop after a month of mindfulness, therefore cannot separate the programme from the repetition. Timing matters as well: the sixty-five young Russian archers of a 2017 study did the tests after a competition and always in the same order, and the last one was the one they did best.

What to check

• which test and which index, and whether the index was chosen before looking at the data;
• whether the test measures speed, accuracy or both;
• whether a control group retakes the test;
• the age of the participants and the norms used to read the score;
• where and when the test is taken, and in what order;
• whether the score comes from competition or training, and whether experience and strength are allowed for, as in the 2026 study in which the link between the grid and accuracy vanished once arm strength was added.

How design and sample are weighed is explained in How we read a study; where to put attention during the shot, in the article on attentional focus.

@misc{campagna2026_enresearchmethodscognitiveattentiontests,
  author = {Campagna, Matteo},
  title = {Cognitive and attention tests},
  year = {2026},
  howpublished = {Research, matteocampagna.com},
  url = {https://matteocampagna.com/en/research/methods/cognitive-attention-tests},
  note = {accessed 2026-10-16}
}