Our moral thermostat - why being good can give people license to misbehave

Blogging on Peer-Reviewed ResearchWhat happens when you remember a good deed, or think of yourself as a stand-up citizen? You might think that your shining self-image would reinforce the value of selflessness and make you more likely to behave morally in the future. But a new study disagrees.

i-6796fd2e10978049b77a127488b335ea-Shoulderangeldevil.jpgThrough three psychological experiments, Sonya Sachdeva from Northwestern University found that people who are primed to think well of themselves behave less altruistically than those whose moral identity is threatened. They donate less to charity and they become less likely to make decisions for the good of the environment.

Sachdeva suggests that the choice to behave morally is a balancing act between the desire to do good and the costs of doing so - be they time, effort or (in the case of giving to charities) actual financial costs. The point at which these balance is set by our own sense of self-worth. Tip the scales by threatening our saintly personas and we become more likely to behave selflessly to cleanse our tarnished perception. Do the opposite, and our bolstered moral identity slackens our commitment, giving us a license to act immorally. Having established our persona as a do-gooder, we feel less impetus to bear the costs of future moral actions.

It's a fascinating idea. It implies both that we have a sort of moral thermostat, and that it's possible for us to feel "too moral". Rather than a black-and-white world of heroes and villains, Sachdeva paints a picture of a world full of "saintly sinners and sinning saints".

In her first experiment, Sachdeva asked 46 students to copy a list of nine words that were either positive ("caring", "generous" or "kind"), negative ("disloyal", "greedy" or "selfish") or neutral ("book", "keys" or "house"). The recruits were told that they had signed up for a study on the psychology of handwriting, and they had to write a story about themselves that included all of the words they saw. They then completed a filler task, after which they were asked if they wanted to make a small donation to a charity of their choice.

Sachdeva found that the students who described themselves with positive words gave the least to charity - a measly $1.07. That was less than the average $2.71 donations of the group that used the neutral words, and about a fifth as much as the $5.30 contributions given by the negative-word group.

Of course, the volunteers' essays may not actually have affected their moral identity. Indeed, they had a tendency to use the positive words to describe themselves, but the negative ones to portray someone else in their lives. To control for that, Sachdeva repeated the experiment with another group of 39 students but this time, she randomly told them to write specifically about either themselves or someone they knew.

Among those who described other people, the nature of the words they used had no significant bearing on the amount of money they donated. But among the group who wrote about themselves, those who described themselves positively gave less to charity ($1.11) than those whose choice of words were negative ($5.56). It seems that a person's propensity for selflessness changes when their self-image shifts.

i-12b7099e9a36ac59d3727a29d9adbbc2-Smokestack.jpgA third experiment supported that idea. After completing the same task as before, 46 students were led to what they believed was a second unrelated study. They were role-playing as the manager of a manufacturing plant, which was facing pressure from environmental lobbyists to reduce the pollutants from its smokestacks using expensive air filters. Other managers had agreed to run them for 60% of the time.

Amid a smokescreen of general questions, Sachdeva asked the volunteers to say how often they themselves would run the filters for. Their answers showed the same trend as the first experiment.

Those who saw the negative words were extra-cooperative, running the filters for 73% of the time. The neutral group ran the filters 67% of the time. And the positive-word group were the least cooperative, running them just 56% of the time. They, in particular, were more likely to think that the plant's profits outweighed environmental concerns. However, when Sachdeva asked them to predict what proportion of the other managers would stick to the 60% agreement, the three groups gave similar answers. Again, it was their own self-image that mattered.

In all three studies, Sachdeva believes that her story-telling task psychologically primed the volunteers with positive or negative traits. They either wanted to cleanse themselves morally, or felt they had license to kick back a bit and let their wicked side out.

Other groups have found similar results before. In 1969, Merrill Carlsmith and Alan Gross found that people are more compliant to a researcher's requests if they had previously been forced to deliver painful (and fake) electric shocks to a (pretend) victim (but not if they just watched this happening). Their motive was to alleviate their own personal guilt, for they behaved in the same way even if the researcher was apparently unaware of their wrongdoing and even if their act of restitution had no impact on the shocked victim. I've also blogged before about situations where people will prefer cleaning products and will physically clean themselves if they remembered a past misdeed.

Sachdeva also cites several studies which have found that ethical behaviour provides a license for laxer morality. People who can establish their identity as a non-prejudiced person, by contradicting sexist statements or hiring someone from an ethnic minority, become more likely to make prejudiced choices later.

There are many potentially fascinating ways of expanding on this study. For example, it would be interesting to see if asking people to remember many instances where they behaved ethically would produce a stronger license to misbehave than recalling just a single good deed. 

Even better, you could see if changing a person's self-image would affect their tendency to cheat in psychological games. That would tell us whether moral licensing gives people an excuse to avoid actively doing good deeds, or whether it actually increases the chances of immoral behaviour, perhaps by lowering the bar for what is deemed acceptable. Do people just avoid being good or would they actively be bad?

Sachdeva is also interested in the types of situations where people seem to break free of this self-regulating loop of morality, and where good behaviour clearly begets more good behaviour. For example, many social or political activists drop out of their causes after some cursory participation, but others seem to draw even greater fervour. Why?

Sachdeva has two explanations. The first deals with habits - many selfless actions become more routine with time (recycling, for one). As this happens, the effort involved lessens, the "costs" seem smaller, and the potential for moral licensing fades. The second explanation relates to the standards that people set for themselves. Those who satisfy their moral goals award themselves with a license to disengage more easily, but those who hold themselves to loftier standards are more likely to stay the course.

Update: Janet Stemwedel has a typically eloquent response to this study, and what it means for scientists. I'd highly recommend a read.

Reference: Sachdeva, S., Iliev, R., & Medin, D. (2009). Sinning Saints and Saintly Sinners: The Paradox of Moral Self-Regulation Psychological Science DOI: 10.1111/j.1467-9280.2009.02326.x

More on morality:

Subscribe to the feed

More like this

Earlier this week, Ed Yong posted an interesting discussion about psychological research that suggests people have a moral thermostat, keeping them from behaving too badly -- or too well: Through three psychological experiments, Sonya Sachdeva from Northwestern University found that people who are…
Throughout our language, the vocabulary of physical cleanliness is also used to describe moral cleanliness. We describe saints as pure and thieves as dirty; consciences can be clean and sins can be washed away. But more and more, psychological studies tell us that these concepts are entwined in a…
The English language is full of metaphors linking moral purity to both physical cleanliness and brightness. We speak of "clean consciences", "pure thoughts" and "dirty thieves". We're suspicious of "shady behaviour" and we use light and darkness to symbolise good and evil. But there is more to…
The more moral you believe yourself to be, the less moral you may be inclined to act, according to a new study in Psychological Science. Psychologists evaluated the moral self-image of subject participants and then presented them with a variety of scenarios in which they were asked to donate money…

Just think what will happen when a whole generation brought up to maximize "self-esteem" achieves its full social power.

We're doomed, I tell you, doomed!

By Pierce R. Butler (not verified) on 06 Apr 2009 #permalink

Charities in the U.S. have noticed that their support goes down under Democratic Administrations and up under Republican ones. Could these findings help explain this phenomenon? People may be ascribing selfless and selfish attributes to the parties and therefore being "let off the hook" by the former? Would be interested in your view, Ed (here or on Twitter).

That is fascinating. I have the same question as the first person who commented, wondering how the emphasis on our kids' self-esteem will then affect their moral choices. I have 2 kids, ages 7 and 10, so this is a practical question!


The people who feel "less moral" are reminded of this and give more to charity as if it were some kind of penance.

Those that feel "more moral" are self-righteous and they think that they already do enough.

I've been a guinea pig for a couple of (completely unrelated) psychology studies where I've been primed by writing a negative essay about myself. Doing something like that - especially if you're expecting someone else to be reading it and judging you by what you've written - can be quite upsetting and make you feel pretty crap and ashamed about yourself. So I'm not at all surprised that the negative group took the opportunity to give themselves a moral boost.

I've never managed to get into a positive group - interesting that the effect sizes seem to be large there as well. Any chance of p-values? I've no access to journals atm.

Typos, I think (feel free to delete this bit): 6th paragraph - "$1.07...over five times less than the $5.30 contributions given by the negative-word group."

11th paragraph - "Those who saw the negative words were extra-cooperative...And the the negative-word group were the least cooperative, running them just 56% of the time."

By Charlotte (not verified) on 06 Apr 2009 #permalink

Pierce/Lillian - I think that's a *really* interesting perspective on this, although I suspect that self-esteem is a much broader concept than a person's assessment of their own morality. I think in terms of taking lessons from this on living our lives, it would be the stuff mentioned in the last paragraph - set the highest possible standards and be prepared to repeat behaviour until it becomes automatic.

Arjay - again an interesting theory, but I don't think you could say anything about it based on this research; this is very much about how *your* own perception of *yourself* affects *your* behaviour.

Ambivalent academic - I think that's pretty much what the authors thought too.

Charlotte - I'll send you a copy of the paper; thanks for the editing ;-)

Haven't there been similar studies (to the first one) in the past? I don't necessarily mean in design, but I thought it was already well known that people that feel better about themselves have less need to offset that by acting externally.

This is still a really interesting article though. I wonder about the generalizability, or perhaps more to the point, the impact of individual differences.

I would take myself as a case example: I tend to have a very high moral view of myself (punctuated with certain gaps when I think about my failure to adopt vegetarianism, which I think is more moral). Overall, I think I do mostly good, and alot of it. I work in research, volunteer, donate to charities (mostly animal based ones, but I work with people all day - it's a balancing act), and am planning a career working with low-income children with developmental disorders. I think I am a pretty good person.

Now given a random meeting in the street (say some college students looking for donations for a local environmental project). I might be less likely to donate to that, since I've already decided on many things that I want to do to help people, and I allocate my resources in a way I enjoy. Overall, I would think of myself as a good person and I think somewhat objectively I would be too.

I wonder if there is a group within the study who is both moral and would "act less morally" as defined by the study parameters. If the study was designed to be like the supposed random environmental project meeting (and it sounds that way, although I have not read the original text), I would wonder about the validity of the measure.

What about the idea that many who really don't care about others still think very highly of themselves?

By InventedHandle (not verified) on 07 Apr 2009 #permalink

Think about the religious stuff. People are made to believe they are born with sin. Will that negative self concept make them be more cooperative than others?

By flyinglion (not verified) on 08 Apr 2009 #permalink

I think it's dead wrong on 2 counts:

It assumes people value giving to charity in the abstract - I don't, for instance, it's usually a game, politics or religion are often involved, and I live in the US, where the elites are starving the beast - us - and scolding us for not replacing basic social services with our money while they make Scrooge McDuck look like a pauper.

It avoids what Mark Twain talked about, the long term vantage, where self esteem FOLLOWS a long period of actually virtuous activity - If the above commenters think people without self esteem are generous and helpful, then indeed we will learn something the hard way.

Actually, I think it's nonsense on many counts, but those two will do for starters. fail.

By Marion Delgado (not verified) on 08 Apr 2009 #permalink

I'm with Marion; I'd want to look at effects over the long term. It is at least prima facie plausible that thinking either well or ill of yourself over the long term would tend to adjust your "thermostat" (if we must use that metaphor) up or down, so while making someone feel badly about themselves might produce better behavior in the short term, it might be counter-productive in the long run if it eventually lowers the thermostat. Indeed, looking at the behavior of people who have a long history of being told they're worthless seems to suggest that this is more than just a theoretical possibility.

"it would be interesting to see if asking people to remember many instances where they behaved ethically would produce a stronger license to misbehave than recalling just a single good deed."

"In 1991, Norbert Schwarz found that people who were asked to remember 12 examples of being assertive rated themselves as being less assertive than those who just had to remember 6 examples. It seems that because people had trouble in bringing a dozen examples to mind, they reasoned that they must not be very assertive after all!"

Our family has been discussing how imposing a moral code (through rules) is necessary for younger children - until they learn to 'do the right thing' simply because it *is* the right thing to do. This study will add a new layer to the discussion.

I can just hear the scenarios: "I don't have a curfew because I've learned to come home at a decent hour, but because I'm sooo responsible, I can drink underage."

Does this extend to why, after a good gym workout or yoga session, I really want to light up a smoke?