Emotion and score in one archer: Johnson 2007 study

The study follows a single 25-year-old Brazilian archer, ranked second in her country, during competition. After each end she said aloud how activated she felt and how pleasant she felt, on two scales from 1 to 9, while a heart rate monitor recorded her heart rate on every arrow. For each distance (18, 30, 50, 60 and 70 metres) the authors looked for the range of arousal, pleasantness and heart rate in which a good end was more likely than an average or poor one.

At 18 metres her best arousal zone ran from 5.0 to 7.7, peaking at 6.3 (a 55.4% probability of a good end). Below 3.7 a poor end became the most likely outcome, and it did so again above 7.7: for this archer at this distance the curve had the inverted-U shape the authors expected. The zones, however, changed with distance. For arousal they were 5.7 to 7.4 at 30 metres, 3.9 to 5.1 at 50, 4.1 to 4.6 at 60 and 5.0 to 6.6 at 70; for heart rate 115 to 128 beats per minute at 18 metres, 135 to 148 at 30, 117 to 125 at 50, 120 to 122 at 60 and 119 to 132 at 70. The width of the arousal zone narrowed from 18 to 60 metres (from 2.7 points to 0.5) and widened again to 1.6 at 70.

The chart of the indoor competition also shows how her state shifted over the day: she started below her zone (arousal 4), entered it at the fourteenth end for two ends, spent five ends above it and twelve below it again, and finished with 28 consecutive ends inside her best zone.

For arousal and pleasantness the authors used Russell's affect grid, a 9 by 9 table in which one axis measures how activated you feel and the other how good you feel. The archer reported it aloud, quickly, at the end of each end of three or six arrows, having practised it in training. Heart rate came from a Polar Vantage NX strapped to her chest, recorded remotely at the moment of each shot and then averaged over the end; the score was noted alongside.

The data are 60 ends at 18 metres (6 rounds of 10 three-arrow ends), 96 at 30 and at 50 metres (8 rounds of 12 three-arrow ends) and 48 at 60 and at 70 metres (8 rounds of 6 six-arrow ends). Each end was classed as good if it was more than half a standard deviation above the mean, poor if it was half a standard deviation below, and average otherwise. Average and poor ends were then split according to whether arousal (or pleasantness, or heart rate) was above or below the mean of the good ends, giving five categories. On these, for each pairing of distance and measure, an ordinal logistic regression produced the probability curves: fifteen in all, from which the zones are read.

Two members of the research team were part of the archer's training environment. The discipline is not stated; the distances are those of the women's FITA round of the time.

It is a single case, and the authors say so: the method is meant to describe that athlete, not to generalise. The text speaks of two competitions, one indoor and one outdoor, but the numbers (six rounds at 18 metres, eight rounds at each outdoor distance) point to several events, and how they were spread is not explained. The rule for classifying ends is also described confusingly: the sentence refers to the means of arousal, pleasantness and heart rate, while what is meant is the score. The half standard deviation cut-offs are the authors' choice, and they accept that others could have been used.

There is also a question of direction. The archer reported how she felt at the end of the end, when she already knew at least part of its result: a bad end can make you feel worse, so the link between state and score could run the other way too. The study does not try to predict later competitions and does not check the zones on new data. Some zones are extremely narrow (at 60 metres the best heart rate runs from 120 to 122 beats per minute), with only 48 data points per distance. Finally, the percentage differences between zone widths are computed against different baselines, so some of them (220% between 60 and 70 metres) look larger than they are. As for level, the title calls her an international archer: she was second in her country, competed abroad and was in her third year of archery.

For anyone who shoots, the most interesting finding is that for this archer there was no single right level of arousal. Her best zone changed with distance and was narrower at 50 and 60 metres than at 18, and within her good zones her heart was beating at 115 to 148 beats per minute: nothing like resting calm. The idea of one ideal state, valid at every distance, finds no support in this archer's data.

The study also suggests a way of watching yourself: noting after each end how you feel and what you scored, and looking back later at where the best ends fall. It takes a lot of data per distance, though (here 48 to 96 ends), and the result belongs to whoever collects it. The question of direction also remains: knowing that good ends come with a certain level of arousal does not mean that getting to that level will raise the score. No intervention was tested, and the authors name it as the next step.