Who you (think you) are shapes what you think

(featured image credit: Settergren)

We have multiple identities, and see the world accordingly

Say you spent your youth in Scotland, but have been living in London for over 20 years. Are you a Scots person living in London, or a Londoner with Scottish roots? That may matter more than you might think.

The renowned British sitcom Yes, Minister and its successor Yes, Prime Minister remain, more than 30 years after the last episode aired for the first time, an example not just of superb comedy, but also of splendid depictions of cognitive phenomena in the wild. One of the finest illustrations of framing (the way in which a particular piece of information or proposition is formulated in order to elicit a given response) is an exchange between the cunning civil servant Sir Humphrey Appleby and PM Jim Hacker’s Principal Private Secretary Bernard Woolley.

yesministerIt concerns the possible reintroduction of military conscription or national service. An opinion poll commissioned by his party has convinced the PM it is popular with the electorate. Sir Humphrey explains how a new opinion poll would produce the opposite result, simply by framing the key question differently.

This is a bit different from the more common applications of framing we encounter, in particular when we go shopping. For example, ‘Buy one, get one free’ offers are more appealing than a 50% discount (even though both offer the same advantage, and even though you need to purchase twice the amount to benefit!). In fact, we are rather sensitive about how numbers are framed in general. In a classic paper, Amos Tversky and Daniel Kahneman describe an experiment in which people had to choose between two interventions, (a) and (b) to counter a disease which would kill 600 people. They divided the participants in two groups:

  • Group 1: (a) 200 people will be saved, (b) 1/3 probability that 600 people will be saved and 2/3 probability that nobody will be saved
  • Group 2: (a) 400 people will die, (b) 1/3 probability that nobody will die and 2/3 probability that 600 people will die

The two versions of (a) and (b) represent identical outcomes, but in Group 1, where the options were framed as ‘lives saved’, 72% chose option (a), while in Group 2, with the ‘people dying’ frame, only 22% chose option (a).

It’s not just the presentation

As the dialogue between Sir Humphrey and Bernard suggests, some clever sleight of hand can manipulate us into changing our preferences, even if there are no numbers involved. But we would be mistaken if we believed that how we think about prices, combating disease, or military conscription only a matter of how the information is presented to us by others.

Jenny Xiao and Jay Van Bavel, two American psychologists, examined how our social identity can influence implicit attitudes. They divided participants arbitrarily in a red and a blue team, and assessed their implicit evaluations towards their own in-group and the out-group. In a second experiment, they first told participants whether their team was competing or cooperating with the other team before the evaluations, and in a final experiment they tested whether the implicit group evaluation changed within an individual who was reassigned to the other team.

The researchers found the participants quickly developed implicit preferences for members from their in-group, relative to the out-group. When the intergroup context was given as competitive, the results were similar, but when participants were told that their team was cooperating with the other team, there was no implicit preference. Most strikingly, when people were switched to the other team, they instantaneously favoured their new in-group (i.e. people who were in the out-group mere moments earlier). This effect was more pronounced for people who feel a stronger need to belong.


Is this your in-group? (photo: eak_kkk)


Implicit preferences have traditionally been assumed to be in the realm of the unconscious, and the emotional – hard coded almost, and hence hard and slow to modify. This research puts this into question.

While these are laboratory experiments, the results suggest that the identity we ‘feel’ at a particular time may strongly influence implicit, automatic preferences more generally. Favouring in-group members can mean that we are more likely to accept and share their opinions, or support their proposals. It can also mean that we might be more tolerant to their transgressions of moral values. But it seems that this may shift quickly if we change in-group – if we activate a different identity. And of course, we do have different identities, each with different in-groups and outgroups. Members of our family or strangers, people who share our nationality or foreigners, divisions along racial lines, level of education, social class… the list goes on.

In a tweet Jay Van Bavel, one of the authors, observes why this may mean efforts to reduce implicit racial bias may fail. About a year ago, an incident in a Philadelphia Starbucks branch caused a lot of commotion. Staff had asked two black men to leave the premises, apparently because they had used the toilets without ordering anything. When they declined to do so (they said they were waiting for a friend), the police were called, and when they continued to refuse to leave, they were arrested for trespass. (At that moment, the friend they were waiting for turned up). Starbucks boss Kevin Johnson apologized profusely and announced the company would deliver training to reduce unconscious racial bias to 175,000 employees.

Alias ourselves

There seem to have been no further similar incidents, but that doesn’t allow us to conclude the training has worked. Research by Calvin Lai, a psychologist at Washington University, Saint Louis, and colleagues, found that 9 interventions of the kind used in the Starbucks training (of which one was a sham) did indeed briefly reduce implicit preferences. However, after a delay of at most a few days, the effect had worn off. This supports Mr Van Bavel’s doubts.

Earlier work by John Turner, a British social psychologist, and colleagues, provides a context for these observations. The so-called self-categorization theory distinguishes between a ‘personal’ identity (how we see ourselves different from other members of our in-group) and a ‘social’ identity (how we define ourselves through shared similarities with members of certain social categories in contrast to other social categories). And that social identity can vary considerably with the social context: we can see ourselves as members of a wide range of in-groups.


Watch out, there’s Sir Humphrey about (image: youtube)

How we perceive and interpret the world depends on what identity we adopt at the time. So what? A first take-away is that, contrary to what it may feel like, we often don’t have a fixed view on matters. We may not even be consciously aware of the identity that is inspiring our opinion. Maybe it would be good if we took both our own views and that of others with a pinch of salt. We should also treat with suspicion anyone who, in the style of Sir Humphrey, tries to persuade us for or against a cause by appealing to a particular identity.

We can use this insight in situations of conflict too. Disagreements with others can look very much less dramatic when you discover a shared identity – imagine how a heated conversation might change when you find your opponent has, like you, teenage children, or a love of Shakespeare.

But perhaps the ultimate take-away is that it is not so much who we are that shapes what we think, as who we choose to be. Better choose wisely.

About koenfucius

Wisdom or koenfusion? Maybe the difference is not that big.
This entry was posted in Behavioural economics, Cognitive biases and fallacies, Psychology, Uncategorized and tagged , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s