1 min read: Behaviour is highly variable, therefore when we are observing behaviour we are likely to see variation over time. Regression toward the mean is the phenomenon that if a variable is extreme on its first measurement, it will tend to be closer to the average on its second measurement. If it is extreme on its second measurement, it will tend to have been closer to the average on its first.
This natural variation can skew the feedback we receive when attempting to encourage or discourage behaviour in others.
Perhaps you deliver some positive feedback after a good performance and the next performance you observe worsens. If you’re unaware of natural variation and the need to observe over time, this could punish your behaviour of delivering positive feedback, after all it seemed to make performance worse!
Conversely, perhaps you observe something highly undesirable and so deliver some critical feedback and the next performance you observe is much better. This could reinforce your behaviour of criticising, when in actual fact it was just natural variation.
In order to increase the reliability of the feedback on whether your attempts at reinforcement and punishment work, be sure to sample multiple performances over time.