Many a times, researchers have their own personal agendas and its very human to fall in to the temptation to interpret study results or spin them to suit ones long term subject matter and expertise. This is a trap in which Joe Forgas et al fall when they report in JESP that happy people are selfish and sad people are fair. They have a long research interest that goes aka sadness is beneficial for you and every result has to fit in that model.
In this latest study they use the behavior in the dictator game as a proxy for selfish behavior. the classical dictator game consist of giving a sum of money to a person and asking them to divide it between themselves and another human being any way they deem fit. If an agent is rational , he should be purely selfish (there are assumed to be no future/past interactions and no reputations to maintain/ cultivate) . As is the normal finding, humans normally give upto 50 % of their share to another person in the dictator game although there are no obligations. As such , dictator game is indeed a good measure of internal selfishness of a person.
What Joe et al do is to induce good(happy)/bad(sad) mood in their subjects and then ask them to play a version of the dictator game. So far so good. However in their version of the dictator game , one is not given a sum of money to divide amongst oneself and someone else; but they are given 10 raffle tickets- each raffle ticket increasing the odds of winning a lottery of 20 $. Now ,here is where I think they have blundered an confounded the results. they have introduced odds and probability thinking in the scene and everyone knows that a normal person prefers a sure sum of money (10 $) over a chance of winning equivalent sum (50 % chance of 0$ and 50 % chance of 20 $). Both utilities are mathematically equivalent but we are all normally risk-averse and prefer the assured sum. However, and this a big however, happy people are more risk prone and may prefer a chance much more than an assured sum. In sad mood things would be reverse.
Its not as if Joe et al are not aware of the happiness-risk proneness link, but they somehow ignore it and let it confound the results. To quote form the paper:
Happy mood may also function as a motivational resource (Trope, Ferguson, & Raghunanthan, 2001), allowing happy individuals to accept greater risks. These findings suggest that happy mood should promote a more confident, selfish allocation strategy, while negative mood should trigger more cautious, fair allocation.
Readers will immediately see where I am coming from. I am a huge fan of seeing happiness/ sadness in terms of environmental risk and safety and as motivational focus- promotion versus prevention focus. It is thus pretty obvious to me that a (or say 10) raffle ticket (a chance of winning 20 $) does not have the same value for a sad person a sit has for a happy person who is more in a risk -prone frame of mind. thus, it seems obvious that a happy person will adhere more utility to the raffle tickets and may not that readily part with them; while a person in a sad mood may think that chancy bit of paper as worthless and be more willing to share it with others. I challenge Joe et al to repeat the experiment with real money and not waffle tickets and then draw any conclusions.
Whats more in the first experiment (described above) the selfish tag on happy people was due to the fact that they did not share that much with out-groups. If out-groups (strangers ) were not present perhaps the results would not have been significant based on ingroup data alone. If, somehow, being sad broadens your vistas and makes you treat outgroups (strangers) the same as ingroups, then this would again confound the result and invalidate the conclusions reached by the authors. Of course this thesis that being sad ,makes you more open to strangers flies in face of the study I covered yesterday that sad people prefer familiarity; but it is something to think about and design experiments to rule out.
Experiment 2 suffers from the same methodical flows. Experiment 3 tried to prove that when social expectations about being fair were relaxed, then happy people gave free rein to their selfishness and became selfish, while sad people remained fair and followed external norms of fairness. This is purportedly to relate it to the internal focus of selfish people and external focus of sad people and fit in a larger framework, but again it fails to convince me. At the outset let me clarify that I do adhere to happy people have internal focus while sad people are more driven by external norms. However the experiment they did supports my thesis sand not theirs.
In experiment 3, they manipulated the perceived social norm of fairness by revealing to subjects the behaviour of some hypothetical earlier participants in the dictator game. some were in the ‘fairness is the norm’ condition (more fair splits be earlier hypothetical participants), others were in the ‘fairness is not the norm’ condition- I like to call this ‘unfairness is the norm’ condition.
Now there are two caveats to this. First not everyone decides whether to act fairly or not based on existing norms. As per Kohlebrg’s or other modern developmental theories one sense of morality in an earlier stage may be driven by norms , but at later stages is determined by internal values and internalized norms. thus it is wrong to apriori believe that if people don’t act selfish it is because of pressure of the social norm of fairness. This is the position that study authors take and this is not necessarily true. Second, even if one grants that one works under social norms, it is false to believe that the social norm is fairness; with the myriad misinterpretations of darwins theory selfishness has become the de facto social norm. Thus, one can as legitimately claim that selfishness is the existing norm that goes undermined by manipulations of experiment 3 and when primed with fair prior dictator behavior, gives free rein to mood effects to take place; while in the second condition where selfishness norm is reaffirmed there are no mood effects.
This hair splitting is important because in the third experiment they did not find a main effect of mood/ prior fairness norm on dictator offerings, rather there was an interaction between the mood and prior priming. They found that when prior exposed to social norm of fairness there was no difference in happy and sad condition; the difference was there only in ‘selfishness is norm’ condition. One can thus, also interpret these findings as ‘selfishness is norm’ – in that selfishness norm when chances are involved happy people make more risky choices than sad people; however when the norm of selfishness is undermined, internal values like being fair (yes there is considerable literature that being fair is more natural and internally driven than being selfish) takes hold and make seven happy people who value the raffle tickets a lot to become more fair and altruistic and share their tickets to the same extent as sad people do normally.
Its not as if they haven’t considered the dilemma of why the norm and undermining of it should be one way only. To quote:
Why did sad people not simply follow the norm – fairness or selfishness – and happy people follow their own internal state (i.e., ignore the norm and act selfishly) in this study? It is likely that information provided about the selfish behavior of others, being socially undesirable, could not invoke an acceptable, alternative shared social norm, and so served merely to undermine the powerful norm of fairness, allowing full scope for mood effects to occur. In contrast, information about the socially desirable, fair behavior by others served to reinforce a powerful existing social norm and so constrained mood effects, as found here.
So in summation, I am not convinced, I still believe the results they got are due to the happiness as increasing risk proneness effct. But I agree broadly with their thesis that sadness also has adaptive value and happiness should not be seen as all rosy and sadness all bad. The bad effect of extremes of euphoria/ mania are well known, to complement lets hear what they have to say of the good effect of sadness. I’ll like to end with their own quotes on this matter.
Interestingly, our results further challenge the common assumption in much of applied, organisational, clinical and health psychology that positive affect has universally desirable social consequences. Together with other recent experimental studies, our findings confirm that negative affect often produces adaptive and more socially sensitive outcomes. For example, negative moods can improve the detection of deception (Forgas & East, 2008), reduce judgmental errors (Forgas, 1998), improve eyewitness accuracy (Forgas, Vargas, & Laham, 2005), and improve interpersonal communication strategies (Forgas, 2007). The present experiments confirm this pattern by demonstrating that mild negative moods also increase fairness and sensitivity to the needs of others.
Tan, H., & Forgas, J. (2010). When happiness makes us selfish, but sadness makes us fair: Affective influences on interpersonal strategies in the dictator game Journal of Experimental Social Psychology DOI: 10.1016/j.jesp.2010.01.007