science tumbled

(pretty pics / longer stories / ask)

Why a classic psychology experiment isn’t what it seemed

Priming is a psychological phenomenon in which being exposed to a word or a stereotype can make us more likely to later act according to the prior stimulus, even if we have no conscious recollection of it. For example, people are more likely to complete a word stem like “TH” with “think” if they were previously exposed to that word. One widely cited study published in 1996 found that subjects exposed to words relating to old age in a word-scrambling task would walk more slowly down a hallway after completing the task. Slowness, of course, being a trait stereotypically associated with the elderly. The authors took this to be evidence that the word-task had primed the subjects to act old. But a new study led by St├ęphane Doyen presents evidence that it may all be in the heads not of the subjects but of the experimenters.

In science, experiments are often double-blind, that is, neither subjects nor experimenters know which one of the possible conditions the subject is in. This is to prevent the unconscious expectations of either subjects or experimenters from influencing the outcome. Doyen attempted to replicate the 1996 findings, but devised a very strict protocol to ensure that the experiment was indeed double-blind, as expected. This experiment failed to reproduce the original findings. No one showed evidence of priming. However, Doyen then did another experiment, this time manipulating the experimenters’ expectations. Half of the experimenters were told to expect that the primed subjects would walk more slowly. The other half were told the subjects would walk faster. Lo and behold, that is exactly what Doyen observed. When the experimenters expected primed subjects to walk faster, they did; when they expected them to walk slower, they walked slower. (The actual measuring of walking time was done automatically using infrared sensors.)

The priming was real, but it had nothing to do with the unrelated word-task. It was all about what was going on in the experimenters’ heads, and how the subjects picked up on it and acted accordingly. Which, as Ed Yong points out, isn’t so different from Clever Hans, the horse who mastered arithmetic.