Running through the shot in your head before you shoot it is advice every archer hears sooner or later: close your eyes on the shooting line, go over the sequence, see the arrow land in the gold, perhaps spend the evening before a competition imagining your ends. For anyone who shoots, the question that matters is whether any of this raises the score. The studies that have tried to answer it span forty years, from 1986 to today, and on that precise question they do not agree: the association says one thing, the first experiment with a comparison group says another, and the results that came later have not settled the matter.
Here I set out what has been measured, on whom and how firmly; each claim has its own entry in the registry, with its current status and sources.
What imagery means
In the studies the word covers different things. You can see the shot from the inside, through your own eyes, or from the outside, as if on video, and you can feel the draw in your muscles or the click of the clicker. Questionnaires also separate cognitive imagery, which is about technique and strategy, from motivational imagery, which is about goals, emotions and arousal. When a study says it trained imagery, then, the first questions are what the archers actually imagined and what else came with it, because in many programmes imagery is delivered together with relaxation or self-talk.
Archers who use more imagery score higher, in a snapshot
The first data on archers date from 1986. An American study collected 62 variables on 188 recurve archers, from beginners to an Olympic medallist, and related them to their average annual score. Among the 115 with complete data, those who reported using more visual imagery had higher average scores, with a correlation of 0.33, and the four skill levels also differed on this scale. In the same sample, however, control of the image (0.24) and its vividness (0.09) stayed below the threshold the authors had set: what correlated was how much the archers said they imagined, not how well they did it. The weekly hours they reported spending on mental training were not linked to score either, with a correlation of 0.15.
A snapshot of this kind cannot say what drives what: better archers may simply have more to imagine, or the two may grow together with years of practice. The same limit applies to a Turkish study from 2023 of 251 licensed archers aged 17 to 20, which found every form of imagery linked to task orientation, the habit of measuring yourself against your own progress, with correlations between 0.14 and 0.34. The highest of these amounts to about 11% of shared variance, so it is a weak tendency, and the only form that went with ego orientation more than with task orientation, imagery linked to emotional arousal, was also the least reliable scale in the questionnaire. In the registry the claim on imagery and task orientation is reported: a single study, on a large sample, showing an association.
The first experiment with a comparison group
The first intervention with a comparison group came in 1991, in Greece. Eighteen archers, nine from the national team and nine of the best who had not been selected, were split into two groups: the first did four weeks of imagery combined with relaxation, eight sessions plus daily practice at home, while the second carried on with technical training only. As the measure of score the authors used two indoor competitions of 60 arrows at 18 m, four weeks apart, so competition scores rather than training scores. The trained group went from 498 to 513 points and the control group from 434 to 454, and the analysis that accounts for the starting point found no difference between them (F = 0.11).
The study also rules out the easy explanations. On a dedicated test all eighteen archers were able to form images, none had ever followed any training in mental techniques, and of the nine trained archers seven said they had felt totally involved many times and two always. The null result, then, cannot be put down to archers who could not visualise or were not paying attention. Two other things weaken it: the groups were not formed at random and started from different levels, and four weeks may have been too short, as the authors themselves wrote; they could not lengthen the programme because the second competition was the last of the indoor season.
Since that year the claim on imagery and score has been contested, because the association pointed one way and the only controlled experiment the other.
The favourable results, and how much they weigh
The clearest favourable result published in a journal without warnings dates from 2025. Thirty-six Turkish beginner archers, aged 14 to 18, were split into three groups of twelve: for eight weeks, twice a week, one group added fifteen minutes of guided imagery of the shot sequence to their training, one added fifteen minutes of slow alternate-nostril breathing, and the third simply trained. Over nine arrows at 18 m the imagery group gained 4.92 points and the breathing group 2.25, while the control group lost 0.42, and imagery separated from the control group significantly.
The Turkish result has three limits that matter: the groups were formed in sequence rather than at random, the archers were beginners, who improve at many things in eight weeks anyway, and the score rests on just nine arrows. The most curious detail, though, concerns the questionnaires. Cognitive imagery, the kind about technique and strategy, did not move in any group, and specific motivational imagery rose both with imagery training and in the control group, so by the end of the programme the three groups did not differ on any imagery scale: the score went up without any change, relative to the others, in what the archers said they were doing in their heads. An Indian study from the same year did find cognitive imagery rising after twelve weeks, but it appeared in a journal with questionable editorial standards, which does not decide a status in the registry. That leaves the Turkish study, on beginners and with groups of twelve, which is why the claim on the effect of training on cognitive imagery is contradicted.
The other favourable studies have more fragile designs. The Indian study just mentioned is randomised, with thirty national-level archers over twelve weeks, and gives the imagery group a 17-point advantage over the group that only trained; of the shooting test, however, it says only that the regulation number of arrows was shot, and the journal has questionable editorial standards. Another Indian randomised trial, from 2026, takes ten archers from 85.1 to 100.5 points in four weeks, against a control group that went from 83.5 to 84.3, with a shooting test that is not described and again in a questionable journal. In Malaysia, finally, in a study with no control group and again in a questionable journal, imagery alone did not move the score at 70 m (p = 0.153) and only did so when combined with self-talk.
The registry reads these studies with two rules: a paper in a journal with questionable editorial standards does not decide the status of a claim, and a before-and-after study without a control group never takes a claim beyond limited evidence. Once the rules are applied, the picture looks much as it did in 1991. The largest gains come from beginners, whose learning of the technique adds to any effect, and from the weakest designs, while the Greek experiment, measured over two competitions, still has no answer of equal weight. The effect of imagery on score remains contested.
Mental training as a whole
If the question is widened to mental training in general, meaning mindfulness, relaxation, breathing, autogenic training and imagery, the controlled studies on archers in reliable journals number four and split two against two. In favour are the Turkish beginners and twelve archers from the Hong Kong squad who, after twelve weeks of mindfulness, gained 11.3 points over 72 arrows while their matched controls dropped three. Against are the Greek experiment and sixteen Ukrainian archers randomised to autogenic training, who after a single week could not be told apart from the controls, although the study does not publish the score data. The claim on mental skills training and score is therefore contested too.
The evidence suggests a hypothesis: the two null results come from the shortest programmes, one week and four, and the two favourable ones from eight and twelve. A meta-analysis of seventeen Korean studies on archers points the same way, with an average effect of 0.58 for programmes longer than twelve weeks and 0.39 for shorter ones, across anxiety, mental skills and performance taken together. It is a comparison between different studies, though, and nobody has ever tried two durations on the same archers, so the claim on the length of mental training is reported; the 2026 Indian randomised trial, in a questionable journal, also finds a large gain after only four weeks. I have written about mental work more broadly in the article on mental skills training.
Why a group average can show nothing
Experiments compare averages, and an average can hide archers who respond in opposite ways. One clue comes from interviews with the eight archers of the Italian Olympic team a month before Atlanta 1996. Describing how they approach competition, three say they see themselves from the inside, two that they see the arrow landing in the gold, one that they see themselves from the outside, four talk themselves through the image, and one avoids imagery because it makes them feel unsure. The study does not measure score, but it helps explain why a one-size-fits-all programme could help some archers and disturb others, leaving the final average unchanged.
Where top archers put imagery
There is also the question of how much room imagery takes up when high-level archers describe their own shot. In 2025 sixty-five Korean archers, eighteen-year-olds with eight years of competition behind them and an average of 327 out of 360 at 70 m, answered open questions about how they had shot an X. In the preparation phase the most frequently mentioned category was confidence, with 25 mentions, followed by loss of self-consciousness, a head clear of thoughts and of the score, with 17. Psychological regulation techniques were the least mentioned category: 9 mentions in all, of which 6 concerned concentration, 2 imagery and 1 deep breathing.
The Italian interviews of 1996 point partly in another direction: there, slow breathing was mentioned by a single archer, as among the Koreans, but imagery was one of the main competition strategies. The two studies did not ask the same thing, however, because the Italians were asked directly whether they rehearsed the shot in their minds, while the Koreans freely described a single successful shot. On breathing the two studies agree, on imagery they do not, which is why the claim on classic mental techniques among high-level archers is contested. It remains a count of mentions, which tells you what comes to mind when an archer describes their best shot, not how often they use a technique or whether it helps them.
What happens in the body while you imagine
Some studies have looked inside imagery rather than at the target. The oldest is Japanese, from 1964, and concerns kyudo: fifteen university practitioners imagined their own shot while their electrodermal skin response, an index of arousal, was recorded. In the four experts, from second to fourth dan, no clear changes appeared, while in the eleven less experienced practitioners the response was present in most cases and absent, or nearly so, in only four. In kendo the same study found the opposite link, which suggests a trait of the discipline more than of experience. The claim that experts react less when imagining the shot is reported, which is the most the registry allows for a claim resting on kyudo alone.
The second finding bears on a very common piece of advice, to picture a calm place to release tension before shooting. In a Japanese study from 2004, seven kyudo archers sitting at rest with a sensor on one finger imagined a relaxing scene, and the small vessels in their fingers narrowed instead of widening, down to 77.6% of the resting value, more than in kendo practitioners and untrained office workers; only the kendo masters, with around thirty years of practice, widened them, up to 107.1%. With a tense image the kyudo archers dropped to 49.6%, and in neither case did heart rate or breathing change significantly. It remains possible that the narrowing comes from the effort of imagining rather than from the content of the image. These are seven people, with no measure of shooting, so the claim on calming imagery is reported: for an archer who pictures a beach before an end it is a warning, not a rule.
On the brain the data do not match. In a 2010 magnetic resonance study, eighteen elite archers imagining the shot activated little more than the supplementary motor area, while people who do not shoot lit up a wide network; in 2014, with twelve international medallists, the same Korean group did not find a smaller network, and the claim is contested. In six Korean national team archers, finally, eight sessions of guided imagery with neurofeedback raised the reported clarity of their images, kinaesthetic imagery from 8.5 to 12.2 on average, but without a statistical test or a comparison group, and without looking at score: the claim on imagery ability is reported.
Imagery against target panic
Imagery also appears among the remedies for target panic, usually alongside going back to the boss at short range. The only study is Indonesian: twelve para-archers, from beginners to elite, did twenty sessions over five weeks of specific bow drills combined with imagery. According to an observation sheet filled in by eye, the number who kept their symptoms under control under pressure went from three out of twelve to five after two weeks, and to ten by the end. There was no group continuing normal training, however, the judgement was the observer's rather than a measurement of the release, the study does not separate the drills from the imagery, and one archer finished below their starting point. The claim on drills and imagery for target panic is reported. I cover the problem in general in the article on target panic.
What I take from this for the shooting line
The registry does not allow us to say that imagery raises scores, nor that it is useless. It allows us to say that the studies on archers have pointed in opposite directions for more than thirty years, and that the most favourable results come from the weakest designs or from beginners. From this I draw three points for anyone who uses imagery or wants to try it, knowing that none of them has been tested as such.
The first concerns how to judge it. An archer who decides to add imagery to their training can treat it as an experiment on themselves: keep everything else the same for a few weeks, record training and competition scores before and after, and remember that a beginner improves anyway, with or without images. The Greek experiment shows how easy it is to credit a method with fifteen-odd points that also reach the group not using it.
The second concerns content. In the programmes studied, imagery of the technique and of the sequence, from the hand on the grip to the release, is something different from a calming image, and the only physiological data on calming imagery in archers, which come from kyudo, show no relaxation response. Anyone who uses an image to calm down before an end would do well to check whether it actually calms them, by looking at how they shoot and not only at how they feel.
The third concerns the differences between archers. Among the Italian Olympians of 1996 one avoided imagery because it made them feel unsure, and in the accounts of the 65 Koreans of 2025 imagery comes up twice in the preparation for the shot. An archer who does not find imagery useful has no study obliging them to use it, and one who finds it helpful has no study advising against it.
What is missing
The study that would settle the question is missing: a randomised trial on trained archers, with an official competition score as the main outcome, a comparison group receiving the same attention without the mental content, and a second measurement a few months later. It would also be necessary to separate imagery from relaxation and self-talk, which it almost always comes with, and to compare different durations, to test the idea that four weeks are too few. Above all, it would mean looking at individual archers and not just the average, because the interviews suggest the same programme could help one archer and disturb another. Until then the question we started from remains open, with studies on archers pointing in opposite directions.