Last Thursday I stumbled across an article: Thinking Too Much: Introspection Can Reduce the Quality of Preferences and Decisions. The article described two experiments. The first concerned itself with jam (or more specifically, how introspection affects our jam-related decision making). During the jam experiment, participants had the tough task of tasting different brands of strawberry jam. The experimental condition then filled out a questionnaire about why they liked or disliked the jams whilst the control condition filled out a non jam related questionnaire. Both conditions then rated how much they liked/disliked each of the jams. Finally the ratings given by the participants in each condition were compared to the ratings given by a panel of jam experts.
It turns out that those in the experimental condition (who wrote down why they liked or disliked the jams) were more likely to disagree with jam experts than the control condition. The conclusion being that asking people to inspect their own thinking can impede the quality of subsequent decisions. In a market research context, this sort of behaviour could have led to the conclusions which supported the launch of the much maligned New Coke in 1985. In blind taste tests, people preferred the sweeter New Coke. But when they knew what they were drinking, they wanted the original—a decision that couldn’t survive introspection.
The jam experiment isnt the only one to look at this effect, there are a number of studies that have investigated the effect of introspection on decision making (Tordesillas, 1999; Wilson, Kraft & Lisle,1990; Wilson, Kraft & Dunn, 1988). These studies throw up some interesting challenges for market researchers investigating how or why people make decisions. The main concern is that asking people to reflect on how they reach a decision affects their subsequent answers. We shouldn’t stop asking these questions entirely—the introspection effect is strongest when people analyze decisions that would normally be made intuitively (Hardman, 2009). However it does highlight that we need to be careful, because the way we conduct our research can often effect our results in subtle ways which we may not be aware of.
So the jam experiment has given me food for thought, I’ll think twice when I’m next tempted to ask an “and why do you feel that way?” type question in a survey or interview. There are also some rather broad unanswered questions which require some brain time (or someone who knows more about decision psychology than me): what is the best way of investigating why intuitive decisions are made, if we subscribe to a two process model of intuitive vs analytic decision making how do we distinguish which process is used and most importantly, is there any jam in the fridge, all this writing about jam has led to a huge craving for jam sandwiches.