06.Jul.2017 Yes, It Applies To You


Defence mechanisms around behavioural change

To effect behavioural change, you have to get people to accept that there’s something about their current behaviour that needs changing. And that’s when you’re likely to hit the ‘that doesn’t apply to me’ defence mechanism.

So, you tell people that 90% of drivers think that they’re above average, as an example of people overestimating their own abilities, hoping they’ll realise that they’re likely to be in the 90%. In fact – partly because people overestimate their own abilities – many people will think, ‘I’m in the 49% of drivers who justifiably think that, not the 41% who overestimate’.

There are two close but different statements here:

[1] ‘People sometimes think/behave irrationally or sub-optimally, and should examine this and change’

[2] ‘I sometimes think/behave irrationally or sub-optimally, and should examine this and change’

If you’re trying to get somebody to believe [2], [1] is much easier for them to accept. If they focus on that, excluding themselves, no behavioural change will take place.

I often see this in interviewing skills workshops, when discussing bias and the need to value processes over gut feelings. It’s easy to get people to accept the idea that, by relying on gut feeling, people often select people similar to themselves, or who make a very good first impression, rather than people who gave evidence that they’ll be good at the job. But the same people who just acknowledged that often argue very strongly that their own gut feeling is somehow different; that they’re an exception.

Why do people do this?

In their book ‘Mistakes Were Made (But Not By Me),’ Carol Tavris and Elliot Aronson identify two key factors behind people’s reluctance to examine their own mistakes: cognitive dissonance and confirmation bias.

We get cognitive dissonance when we try to hold two conflicting ideas in our head at once. It’s distressing to most people, so we try to avoid it. An experienced recruiter may feel they make good hiring decisions. But the idea that their gut feeling is unreliable conflicts with this. To avoid cognitive dissonance, they either have to ditch the idea that they’re good at hiring (hard), work out some kind of compromise between the two ideas (hard work), or discount the new idea (easy).

We exhibit confirmation bias when we watch for evidence that supports our viewpoint, and subconsciously ignore, forget or discount evidence that conflicts with it. So, when discussing the value of their gut feeling, a recruiter might easily remember the times their gut led them to a great hire, or ignoring it led to a bad experience. There will be some such examples because it’s not an exact science, and tossing a coin will give the right decision 50% of the time. They’ll fail to recall, though, the times a gut decision went bad. This selectivity is not deliberate. They’re not hiding evidence, they’re just not noticing or storing it in the first place, as it presents itself. So they don’t see any need for change.

I’ve focused on hiring for illustrations, but it’s the same with driving. Drivers remember all the times somebody else drove badly, but forget, don’t notice or downplay the times they did something non-ideal themselves. And believing you’re a good driver and believing you’re somebody who overestimates your driving skills causes cognitive dissonance. There are many other ideas out there where these defences may kick in. Any time we want people to change their behaviour because their current ways are non-ideal, avoidance of cognitive dissonance and confirmation bias may work against you.

How can we overcome it?

There isn’t one right way to get past this defence. But there are a number of ways more effective than just showing the common mistake made. The rest of this post suggests a few; I’d welcome further suggestions.

1: ‘Trick’ them into proving they do it

Use a question or short activity to make them demonstrate the problematic thought pattern or behaviour. Take care not to offend or embarrass, though. So, you could get them to answer two very similar questions or solve two very similar problems, with one key (disguised) difference. Would they go to a different shop, three-quarters of a mile away, to save £10 on a £20 item? Would they make the same journey to save £10 on a £1,000 item? If they answered differently in this example, they make value judgements relatively, not absolutely, like everyone else. Don’t tell them that other people answer the question badly, get them to answer it badly here and now. This can bypass confirmation bias, because they can’t easily discount evidence that happened in the room.

2: Assume they do it

Instead of ‘most people do this, don’t you think you do?’, try ‘everyone does this; how do you sometimes do it?’ This can work well in small groups or with some initial, frank examples of how you/some very experienced (specific) person does it. For example, Chimamanda Ngozi Adichie’s TED Talk, ‘The Dangers of a Single Story’ includes an anecdote about how she, despite suffering because of assumptions based on stereotypes, realised she’d done exactly the same to others. It’s a powerful way to get the ball rolling about how we all do/have done this sometimes. It helps with cognitive dissonance, because they can say ‘well, even people who are really good at this get it wrong sometimes’

3: Do it over time

Rather than asking them to accept right now that they do whatever it is, ask them to record examples of it between now and next time you meet. This can help address confirmation bias, but only if they’re engaged enough with the question to honestly consider it and look out for examples. This method works best with a relatively strong sales pitch on the idea that they could be subject to the issue, but you can impress that all you’re asking is for them to consider that they could be subject to this, not that they are.

Write a Comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>