Tuesday, November 20, 2012

Not much more than an excuse to see an amazing goal

If you want to maximize your chances of success today (unless you're competing in sports), you'd better banish the color red from anywhere around you, because red is associated with failure.  These results are reported in a paper published in 2007 ("Color and Psychological Functioning: The Effect of Red on Performance Attainment," Elliot et al.), but we just heard about them on a repeat of the Nov 16 BBC radio program, "The Why Factor," on which a panel discusses the cultural meanings of the color blue.

From the paper abstract:
Red is hypothesized to impair performance on achievement tasks, because red is associated  with the danger of failure in achievement contexts and evokes avoidance motivation. Four experiments demonstrate that the brief perception of red prior to an important test (e.g., an IQ test) impairs performance, and this effect appears to take place outside of participants’  conscious awareness. Two further experiments establish the link between red and avoidance motivation as indicated by behavioral (i.e., task choice) and psychophysiological (i.e., cortical activation) measures. The findings suggest that care must be taken in how red is used in achievement contexts and illustrate how color can act as a subtle environmental cue that has important influences on behavior.
Presumably this is known to psychologists, but we didn't know it -- there's a lot of empirical research done on the effects of color on performance and productivity, and some older theoretical work on color and psychological functioning.  A 1942 paper described physiological effects of certain colors in psychiatric patients, with, e.g., red and yellow being experienced as "stimulating and disagreeable" and serving to "focus the individual on the outward environment, whereas green and blue are "quieting and agreeable and focus individuals inward."

Apparently not much work was done on this between then and the work reported in 2007, and replicated a number of times since then (though much of it by Elliot and colleagues).  Various hypotheses about the effect of color on mood had been proffered, but based, according to Elliot et al., on suspect premises and not rigorously tested.  Naturally enough, people have tried to find differential effects of pink vs blue, but with no significant results.  So, in 2007, the time was ripe for a rigorous test of the effect of color on performance.  (The things people can get funding for.)

The idea they tested, in six different experiments, was that red is culturally associated with danger -- stoplights, fire alarms (fire alarms?), and warning signs -- and red marks on school papers are associated with the psychological danger of failing, so that exposure to red will elicit failure.  The idea that the association becomes causation seems a bit of a stretch, and even a logical fallacy, but apparently the psychological literature is replete with examples of just this; exposure to a negative object associated with failure causes fear of avoiding failure, which produces anxiety, which impacts performance -- i.e., causes failure. And of course this is all subconscious.

Well, so much for logical fallacies, apparently.  Each of the experiments, described at length in the paper, "provide strong support for our hypothesized effect of red on performance."  Not only did they measure the effect of color on achievement test results, but also, with EEG, discovered that different parts of the brain were activated after red vs green exposure.  

On the 20 question IQ test, the 15 participants in red conditions did significantly worse than those in green conditions; the former got an average of 10 or so questions correct, while the latter got an average of 13+ correct.  This same basic trend was observed for each of the experiments.  Not, it must be said, wildly different, but a trend.  Elliot et al. suggest that their findings have social consequences as achievement tests are "filtering devices in society."  They propose that cultural influences, such as our response to color, on achievement are profound. 

If these results are true -- and it's hard to tell from the paper whether the effect on IQ scores themselves would be significant -- and if IQ scores are so readily influenced by psychological factors, then what are these tests actually measuring?  Most arguments about IQ have to do with 'racial disparities,' the difference in mean scores by race, and whether that is innate or cultural, but if, e.g., a test proctor is wearing red, and red does have an effect, it would be on everyone in the room, not just one race.  That is, everyone in cultures for whom red is a danger signal.  If this effect is real -- and if it is, this doesn't mean that the explanation offered in the Elliot paper for it is necessarily correct -- these results would point to an inherent instability in IQ scores due to unmeasured confounders, whether or not you believe IQ itself is real. This, of course, is something that plagues many many types of studies.

The persistence of gene-worship in the face of these facts is somewhat striking.  A gene devotee would say that we knew that not all of a trait is genetics.  The 'heritability' or fraction of the variation that is due to genetic variation is usually around 40%.  This doesn't mean that a particular genotype accurately predicts a particular trait, like IQ, but it means there are other factors.  Perhaps seeing red is one of them.  This should pull some of the obsession away from genetics, because we don't know these factors well, there are many of them, and they may change rapidly in our culture, unlike genes, so that the predictive power of genes also changes correspondingly since genetic effects are mediated by these other, changeable, factors.

In any case, a quick review of the literature suggests that in sports red is associated with success -- apparently this is well-known.  In competitive sports, that is; red swim suits are good, and red soccer shirts, red t-shirts for any sport.  Red either makes you win, or makes your opponent lose, depending on your interpretation of the studies of this.  If the latter, then clearly the competitor sporting red just can't look down.

Hey, it worked when Ibrahimovic was playing for Barcelona!

Zlotan Ibrahimmovic, when he played for Barcelona

But here, Ibrahimovic is in gold, and the goalie in red.... (and, by the way, he's making what some have said is the best goal ever scored.)

 Enough said?


  1. I'm just going to provide three examples that came to mind on this topic of color and psychology or behavior. I have no overarching point other than to share and perhaps to suggest that reading (or rereading) Lakoff & Johnson's "Metaphors We Live By" when grappling with these questions might be useful.

    First, I recall the year my high school library had their tabletops resurfaced from a pale creamy yellow to a bright red. That simple change not only upset the reading teachers and librarian but also disrupted the concentration of students (including me) trying to study at those tables (previously passive attractions that became active distractions).

    Second, restauranteurs have long been interested in the effects of color on mindset and behavior, though I don't know what credible scientific evidence there may be on the point. Why do fast food restaurants have bright reds and yellows (as opposed to pastels, cool hues, or even neutrals)? And why do relaxing coffee shops look so different? Red and other "high energy colors" allegedly cause blood pressure to rise, makes customers hungrier, makes customers order more food and eat more quickly, etc. Now we are seeing a transition, a re-branding of restaurants to use more green to trigger a whole host of emotions/thoughts (healthy, natural, sustainable, etc).

    Finally, I recall interviewing advice that was pervasive in the law school setting if nowhere else. We were warned not to wear anything but navy, black, or tan/khaki suits. The idea is that grey signified seniority and would challenge the authority of the interviewers and that other colors would be dangerous because it was hard to interpret whether color would send a message of confidence or arrogance. This would extend to tactics for representing clients at mediations or arbitrations (red shirts or ties when you want to portray a strong, aggressive position; wear cool colors to portray your position as more amicable, willing to compromise, etc). I'm sure you can find many conflicting reports on the do's and don't's.

    Causation is so tricky when we are thinking about behavioral traits. There are constant feedback loops and complex interactions (both on the developmental timescale and evolutionary timescale). Can we tease apart whether seeing red (no pun intended) really triggers the emotions/thoughts/physiological responses or whether red is simply the arbitrary color we've assigned to so many aspects of society that are emotionally charged (proscriptions like stop lights/signs, warning signals, etc) and trigger those responses without any real contribution by the arbitrary color selected? And I guess I question whether it really make a difference now whether it is the color red by virtue of being the color red or red by virtue of its pervasive associations with emotionally-charged aspects of society. Is there a way to reverse that process of our (my apologies for bringing it here) implicit prejudices and biases regarding the color red?

    1. Absolutely. I don't disagree at all that colors may have strong effects, and possibly and perhaps even probably for the reasons you suggest. As always, my greatest problem is with whether it's possible to demonstrate any of this with the kinds of study designs and methods we currently use. And that interpretation of results can fly off in any direction, too often based on pre-conceived notions of what's being tested or what results are going to mean. It's perennially of interest to me that so many studies that ask the same question so often get diametrically opposed results. And even if red does have an effect, does it make a player play better? Or make his/her opponent play worse?

      But the real question is, what about that goal?? Some have said that the goalkeeper is too far forward for it to really be as spectacular as it looks at first blush. But then, perhaps disagreement on whether or not this is one of the all-time great goals is as good an example as any of why 'evidence' is so hard to interpret. Here it's right before our eyes, but we could well all see it differently.