If you’ve been reading this blog for a while, you probably know that I’m fascinated by findings that show just how little we know about ourselves. Most of what’s going on in our heads occurs below the level of awareness, and behind the often impenetrable barrier of the unconscious. Often when we’re asked to make judgments, explain our actions, or assess our current motivational or emotional states, we’re pretty much just guessing, and using what, from a third-person perspective, often seems like the least relevant information to do so.
One great illustration of this came in a classic experiment by Norbert Schwarz and Gerald Clore(1). They randomly called people in the university directory on either rainy or sunny days, told them they were doing research for another university, and asked them to rate their current mood and their satisfaction with their life “as a whole these days.” Previous research had shown that the weather can affect people’s current moods, with rain making people less happy, and sunshine more happy. So Schwarz and Clore expected people called on the rainy days to be less happy, currently, than people called on sunny days. Schwarz and Clore were able to replicate this effect, with rating themselves as called on sunny days rating themselves as being happier (m = 7.5, on a scale of 1 to 10 with 1 the least happy and 10 the happiest) than people called on rainy days (m = 5.4).
But the weather shouldn’t affect life satisfaction, right? Oh, but it did! In their control condition (more on that in a moment), people called on sunny days were more satisfied with their lives (m = 6.6, again on a 10-point scale) than people called on rainy days (m = 4.9). This suggests that people are using their current mood, which is largely caused by temporary external factors (in this case, the rain), to evaluate their life satisfaction. But it gets better.
In two conditions, the interviewer caused the people to think about the weather before asking them the mood and life satisfaction questions (in the control condition, the interviewer didn’t say anything about the weather). In one of these conditions, the weather prime was indirect: they just began the conversation by asking how the weather was. In the other, it was direct: they told the respondent that they were interested in how the weather affected people’s moods. In both cases, the effect of the weather on mood remained. That is, people still felt happier on sunny days than rainy ones. But the effect of the weather on life satisfaction disappeared. The ratings for people called on rainy days (6.71 and 7.07, in the indirect and direct prime conditions respectively) were almost identical to those of the people called on sunny days (6.79 and 7.21). Further, when Schwarz and Clore divided people by whether they were in a good or bad mood, they found that the effect of making them think about the weather only held for people in bad moods. That is, if they were in a good mood, they rated their life satisfaction as being high regardless of whether they were induced to think about the weather, but if they were in a bad mood, thinking about the weather caused their current mood to influence their life satisfaction rating less, and they gave higher satisfaction ratings.
What the hell, right? Here’s how Schwarz and Clore explain their results (p. 520):
[B]eing in sunny versus rainy weather influenced subjects’ reports of general well-being. However, this influence was not direct. Instead, it appeared to occur only insofar as these factors affected subjects’ moods, and these moods were considered to provide reliable information about well-being. Subjects appear to seek personally irrelevant explanations for an unpleasant mood state when such explanations are available. As a consequence, they then do not use their mood as information about their well-being. When in a good mood, however, subjects do use their mood as a basis for judging the quality of their life regardless of the availability of alternative explanations for the mood.
In other words, when we’re doing something difficult like rating our overall life satisfaction, something that doesn’t come very naturally, we tend to rely on our current mood, even if that mood is influenced by factors that are, for the most part, irrelevant to our overall life satisfaction — like the weather. But if we’re made aware of the influence of irrelevant factors on our current moods, we realize that those moods aren’t good indicators of our life satisfaction, and discount them when assessing that satisfaction, particularly when the unreliable indicator suggests dissatisfaction.
This finding seems particularly relevant these days, when research about happiness and life satisfaction is all over the news (e.g., “findings” suggesting that men are happier than women). If we’re just not very good at assessing our life satisfaction, and have to rely on our current mood, whatever its causes, to do so, then such research, at least to the extent that it relies on self-report (and it mostly does) is pretty much worthless.
1Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social
Psychology, 45, 513-523.