Swing thoughts: where your attention should sit

Arms or club: two words apart

The choice is narrower than it sounds. Nevo, Halperin and Ziv gave putters one of two instructions: concentrate on the pendulum-like movements of your arms, or on the pendulum-like motion of the golf club. One noun swapped. Attention aimed at a body part is an internal focus; at the club, the ball or the target, an external one.

Golf is where the idea was first tried outside a laboratory. Gabriele Wulf, Barbara Lauterbach and Tonya Toole had 22 people with no golf experience hit 80 pitch shots with a 9-iron, half attending to the arm swing and half to the club swing. The club group was more accurate in practice and again a day later. Barzyk and Gruber's review records the family: 6 to 30 participants, 60 to 80 shots, a target 15 m away.

The constrained action hypothesis

Wulf's explanation is that an internal focus interferes with something already running unsupervised. Steering your own limbs constrains the motor system; pointing attention at the club leaves it to organise itself. That is the constrained action hypothesis, and the experiment behind it was not a golf shot.

Participants balanced on a moving platform while responding to a tone by pressing a button as fast as they could. The external-focus group balanced better and pressed faster. Quicker on the side task means the main task was consuming less attention: an internal cue does not hold worse information, it charges you attention the movement needed.

Kuhn and Taube set out four lines of evidence: postural correction frequency, dual-task costs, muscle activity, oxygen consumption. The muscle-activity strand is thin, pooled from 12 studies and 216 subjects. Pelleck and Passmore's golf putt is the closest trial of turn my shoulders. Accuracy, muscle activity and kinematics all came out worse when attention went to technique-relevant upper-body mechanics than to lower-body mechanics irrelevant to the stroke.

What the putting numbers were

≈4 cmArms cue against club cue, first day, 2 m putt

Thirty students putted from 2.00 m to a regulation 10.8 cm hole, 15 putts under each instruction, on four days across roughly a week. The pre-registered analysis found no effect of instruction on absolute error, F(2, 58) = 2.35, p = 0.10. Practice moved it: mean error fell from 35.17 cm on day one to 23.91 cm on day four.

An exploratory day-by-day look found the one difference. On day one the arms cue left the ball 38.48 cm from the hole and the club cue 34.19 cm, while a third instruction mentioning no technique at all, make sure you are standing at the correct distance from the golf ball, left it 32.84 cm away. The control beat both, and by day two the gap had gone.

How firm any of this is

g = 0.01External-focus performance effect, corrected for publication bias

Seven earlier meta-analyses all reported external-focus superiority. Brad McKay, Michael Carter and colleagues reanalysed those data in Psychological Bulletin with models that allow for publication bias, and found moderate to strong evidence of it everywhere. Corrected mean effects came out at g = 0.01 for performance and 0.15 for retention, with Bayes factors favouring the null.

The golf work is fragile from another angle. The experiment quoted most often as proof that experts benefit too, the second half of Wulf and Su's study, had six golfers in it, at a statistical power Barzyk and Gruber put at 0.30. Perkins-Ceccato and colleagues found the direction reversed by skill: better golfers did well with an external focus, weaker ones with an internal one.

Nor do external cues agree among themselves. Bell and Hardy gave 33 skilled golfers 100 pitching-wedge chips from 20 m, and the group aimed at the distant effect of the shot beat both the club-focused and the internal group. Direction is the safe part of this literature. Magnitude is not.

How many, and what survives the first tee

4.25 and 2.81Spoken thoughts per full shot, 4-handicap and 20-handicap golfers

Nobody has counted the thoughts inside a swing, because nobody can see them. Thoughts per shot have been counted. Whitehead, Taylor and Polman had golfers speak aloud through six real holes: the 4-handicappers averaged 4.25 spoken thoughts per wood or iron shot against 2.81 for the 20-handicappers, and 2.19 against 1.47 on putts. Most of the extra ones were planning and club choice, before the club moves.

The competition round is where that changes. Eight low handicappers played a practice round, then one with money on it and a league table going up in the clubhouse. Technical self-instruction during putting rose from 1 per cent of putts to 9, and the players who shifted most scored highest on a reinvestment questionnaire. Pressure fills the head rather than emptying it.

Aiken and Becker's chipping study is the nearest thing to permission for both: a group using an internal focus while preparing and an external one while swinging beat the pure internal group, though only once participants who ignored the instruction were dropped. Pebble Swing reads joint angles out of the frames and the backswing-to-downswing ratio out of the motion, so the position you are changing gets checked between rounds instead of mid-shot.