Don't touch! On the mixed effects of avoidant instructions

What happens if you tell a golfer not to over-shoot a putt? Does it make them more likely to overshoot (an ironic effect, like the way suppressing thoughts of white bears actually leads to bear-based thoughts) or does it provoke over-compensation - putts that are particularly short? The same question could be asked for similar situations in other sports and also for movement instructions in the psychology lab.

Past research has produced mixed results - sometimes ironic effects are observed, other times over-compensation. To find out what's going on, Christopher Russell and colleagues conducted a lab experiment in which 40 undergrads repeatedly traced an imaginary straight line between two on-screen dots with a computer mouse. On key trials, they were given an additional instruction not to move to the left (or right, etc) of the line.

The effect of this additional instruction was far from uniform across the participants. The largest sub-grouping over-compensated. So if the additional command was 'do not move to the left', these 26 participants actually traced a movement that was further to the right of the straight line than when no such additional instruction was given. However, there was another group of participants who demonstrated ironic effects - the instruction to avoid going left made them more likely to do just that. Although the ten participants in the latter group scored higher on trait and state anxiety, their errors were actually smaller than the over-compensators.

What about the effect of an extra distraction task? When participants had to trace straight lines while simultaneously holding in a mind a 7-digit number, the effect of an additional 'do not go left/right' instruction was reversed for about half the participants - if they'd shown ironic effects in the absence of the memory task, they now over-compensated, and vice versa. Curiously, for a minority of participants, the extra burden of the memory task actually had a beneficial effect - cancelling out their earlier ironic response to the additional task instruction, so that their mouse movements ended up straighter.

So, where does all this leave us? Although the theoretical implications are beyond the scope of this item, there are some take-home practical lessons. 'The avoidant instructions used here deliberately resembled those used by coaches to direct their athletes and by athletes in their self-talk,' Russell and his colleagues said, 'to show how ironic outcomes and over-compensations can be unwittingly inflicted in sporting contexts.'

Researchers should also take note that the avoidant instructions they give to their participants could have important, unpredictable effects that vary from one participant to another, and from one condition to another, depending on concurrent task demands. 'What experimenters often assume to be random error in their data may actually be explained by ironic and overcompensatory responding styles,' Russell's team said. 'In any event, experimenters may minimise their biasing influence by emphasising to participants what is to be achieved while neglecting to specify what should be avoided.'
_________________________________

Russell, C., and Grealy, M. (2010). Avoidant instructions induce ironic and overcompensatory movement errors differently between and within individuals. The Quarterly Journal of Experimental Psychology, 63 (9), 1671-1682 DOI: 10.1080/17470210903572022