2013 NCWIT Summit - Plenary III, "Habit Disruption: In Search of How People Change" by David Neal

August 14, 2013



Jane Margolis:  David, ah, Neal is a social and behavioral psychologist, specializing in behavior change. His academic research focuses on understanding basic mechanisms of habitual control in daily life, including the influence of environmental cues, planning, and self‑control. He received his PhD from the University of Melbourne, and completed his post‑doctoral training at Duke, where he was the director of social science research laboratories. David is published broadly, in the areas of behavioral change, attitudes, and motivational processes, and consumer and social decision making. His academic research has been featured in the "New York Times," "TIME Magazine," The BBC, NPR, "The Independent," and "The Australian,"

And additionally, he's been working with Microsoft on habit changing in technology use. He's worked with the Surgeon General on health behavior in the US Military, and with the Australian government on community‑based health behavior change campaigns. So I'm really looking forward to, um, to hearing his talk.

It very much builds on the investigation that we've been doing on stereotype threat, and, um, in the SSAB, the Social Science Advisory Board, we've been looking at the whole issue of social change. How does change happen and what is the relationship between systemic change, that must happen, and individual change, and how do the two converge?

So, I, We look forward to this talk, and, um, just know that a lot of us in the audience are wrestling with that kind of uh, tension, of these two different approaches to social change. So, welcome, David.


David Neal:  Wow! Um, hi, everybody! Um [laughter] I feel like I should sing now, but.

David:  Thank you so much for the invitation to come and talk to you today. Um, as Jane mentioned, I am a psychologist and the issue that I um, and Wendy Wood, who's my mentor, my collaborator in most of the work that I'll tell you about. The thing that we've really been interested in is how people can successfully change, especially really entrenched behaviors. Things that they've been doing for a long time. Things that are like second nature to them. Um, and, so what I'm going to try to do in this talk is sort of expand a little bit back from ah, gender and IT. And, look more broadly at the sort of, the psychological machinery if you like, that determines whether or not human beings ah at the individually level and at a group level are successful at changing their behavior. Um, but let me start by putting it in in context with some kind of data that I find kind of paradoxical.  And I'll see whether you feel the same way.

But Gallop has been, since the 1950s been doing this survey where they ask, whether you'd prefer to have a man or a woman as a boss. And so here's the data from the 1950s. Oh, sorry. Um, and what you can see is a steady decline over time in the number of people who say they would want a man for a boss, alright, and a steady increase in the number who say they would rather have a woman as a boss.

And now, the largest single group is people who say they have no opinion, no preference.  Ok? So, obviously, ah, haven't reached a state of full parody, but it's pretty evident that over the past 60 years, there's been a significant positive trajectory, right, in people's attitudes to women in leadership. Now, let's look at outcome data, alright.  So let's look at the actually percentage of women and men in these leadership roles in the United States.

And this is 2013.  Ok, and so we’ve reached about parody for all managers and professionals. This is using Department of, the Bureau of Labor Statistics, ah numbers. With managers there's still only 39 percent women, with all CEOs 27 percent, 16 percent for ah, F500 board members, ah, 14 percent for corporate offices, eight percent for top earners, and four percent for CEOs. Ok?  Ah, so and as you know, other domains of leadership like politics, the data is similar.

I think that the US currently rates 77th in women ah, elected to federal ah, legislatures around the world, which puts the United States I think just slightly ahead of Madagascar and the United Arab Emirates, um.


David:  So, the thing that is weird to me about this as a psychologist and also as a human being is, why would it be the case that there's been such positive shift on the sort of attitude, value metrics, but the outcomes haven't shifted much at all? So there's a kind of disjunction between the two. And of course, there are, there are many things that no doubt contribute to this.  Alright? So first of all, obviously the Gallup data is people's sort of self‑reported expressed conscious attitude. And maybe we, in fact we do know that people have unconscious attitudes as well that they may not be reporting on in those kinds of surveys.

Another contributor might be the fact that those data are from the general public, but maybe the attitudes of decision makers who are actually making these, these ah employment decisions, maybe the attitudes of those people are different.

Of course, there are no doubt generational effects where maybe it takes the outcomes longer to catch up. An attitude can change quickly, ah, someone's movement up the corporate ladder is a ah, much more time‑lagged phenomenon. Alright?

And then of course there’s other sort of structural variables like ah, traditional gender roles, and ah workplaces and organizational roles that are structured better to men's lives than to women's lives. So all of those things no doubt are important, and I don't want to ah gloss over those in any way. But what I want to do today is to focus on what I think might be one other contributor to this disjunction.

And that is the idea that organizations are made up of individuals, individuals are creatures of habit, and we have myriad sets of routines, rituals that have accumulated over time and may partly explain why we have such a hard time marrying up our attitudes and our values with the behavioral outcomes that we care about.

And so, here's a kinda little graphic um, that sort of brings this into relief. So, so the sort of basic thesis that I'm going to argue today is that there is this large gap between our attitudes and our intentions, our values, and the behavioral outcomes. And that one of the things that causes that is habits that we've accumulated over time at an individual level, at an institutional level, at an organizational level. Ok.

So to get a little bit more psychological about this, I want to eh introduce to you this idea that in a sense we have two selves um, and I'll kind of run with this idea throughout the talk. One self we have is the, what I'm going to call the "intentional self." So this is the part of ourselves that is guided actively by our attitudes, our goals, our values.

It's the more conscious side of ourselves. It's the side of ourselves that has knowledge that we can easily talk about and verbalize, and it's the sort of eh control, the sort of knowledge that can change pretty quickly just by giving someone a rule or by choosing to take on a rule yourself. So that's the kind of intentional system.

And then on the flipside, we've got a habitual system, alright.  Which we also rely on from time to time to drive our behavior. And that system is guided very differently from the intentional system. It's driven by cues. We'll talk about what those cues might be um later on. It's less conscious.

Most of the time, we're not thinking about what we're doing when we're acting out of habit. It's not verbalizable. Just like you know all of us perhaps can ride a bicycle, but if someone said, "Could you explain how you do it?" It's very hard to verbalize that kind of implicit knowledge.

And the final piece is that habit knowledge changes very slowly, and only through experience. So whereas you can give a person a rule to change ah, to change the intentional self, the habitual self really only changes in the long term through sort of slow, gradual accumulated experience.

And one of the really important things to note is that these two sort of sides of ourselves are ah often locked in a, in a battle to control behavior, and the sort of balance of power switches the more we've done something. Right?  So initially, if you're in a new role, in a new place doing a new thing, the intentional system will be dominating control of your behavior.

But then the more you perform a behavior, the more familiar you are in an environment, the more opportunities there are for you to have accumulated habits, and the balance of power will switch over so more of what you do is controlled by this habitual self. Ok?

And so this, I think underlies some of the problems we see in getting individuals and groups to bring their behaviors in alignment with their attitudes and intentions. I want to use a little visual analogy to make this point.

So you, you may have seen this visualization before, um.  And so if I asked you which one of those tables is longer, which would it, which would it be?

Male Audience Member:  The one on the left.

David:  Can anyone see anything other than the one on the left being longer than the one on the right?

Female Audience Member:  They are the same length.

David:  Ok, they are the same, they are the same, but what is your visual experience? Does everyone see the one on the left as longer? OK. Let me help you out by putting a ruler on them.  Ok?  And now let me put the ruler next to the other ruler.  Right?  To try and convince you that actually they are the same length. Right?

And then take that away and if, unless you are different from me, you still can't help but see the one on the left as longer than the one on the right.  Ok?


David:  So, this, this I think, as an analogy is, ah, is a way of thinking about this battle of wills between the intentional self and the habitual self.  That in this case here, your sort of rational conscious mind knows that the tables are equal length, but the visual processing system is so automated ‑‑ is so reflexive ‑‑ and it's the one that's controlling what you actually see. Alright? So you can't use your sort of conscious, rational self to make yourself see these tables as they actually are. You’re forced to see them as that unconscious, habitual vision system, vision system, forces you to see them.

And so what I'm going to argue is this, to some degree, is analogous. This situation is not quite so bad, but to some degree, analogous to the problem we have when we try to just focus on changing hearts and minds when people's behavior is stored in this other system. Ok.

So let me just spend a little time convincing you that this, that phenomenon isn't something that just works with visual illusions. It actually plays out in real people's lives, um.

So I’m going to show you some data. This is from ah a big study called a meta analysis. It is basically a study where the authors, ah get together a whole bunch of other studies and compute the average effect from those studies.

And the studies in this meta analysis all had one thing in common, and that is they were interested in trying to predict what a person would do in the future based on what they said they wanted to do ‑‑ their intentions, their preferences ‑‑ and to have its strength through the behavior. Meaning how long had they been performing that behavior and how frequently do they perform it?  Ok?

So you've sort of got these. You can think about this as we want to see whether the habitual self or the intentional self, who has control of the reigns in predicting what someone does in the future?

And so what you can see here is that, first of all, for behaviors that people don't perform very often ‑‑ so things like getting a flu shot or registering for class or donating blood ‑‑ the intentional self controls what people do.

Ok?  So these are correlation, coefficients so that basically tells you that there is a much stronger relationship between intentions and future behavior for these actions that we don't do very often. Alright?  But now let's look about, let’s look at the pattern for behaviors that people perform really frequently, and especially if they do them in the same environment.  Ok?

So things like exercise and seat belt use. Suddenly, the balance of power gets flipped around.  Alright?  And the habitual self and that basically just means the sheer number and and contextual stability with which you've done the action in the past.  Alright?

Have you done it a lot? When you do it, do you do it in the same environment? That is the thing that is predicting whether you will do it in the future and whether you say, "Yes, I'm absolutely going to go and donate blood," or, "No, I'm definitely not going to donate blood." That contributes very little predicatively to rather you go and donate blood.  Ok?

So this is also, as you can imagine a major problem when you want to try and get in either as, with an individual or with a group of people to change behavior. Alright?  And so what I'm going to show you now is data from intervention studies that have ah, taken the behavior and tried to convince people through a sort of intentional pathway to providing lots of information, lots of facts and trying to get people motivationally on board with the idea of changing their behavior. Ok?

And so in this study, what Reverend Sharon did was ah broke up the studies into ones that were trying to tackle things that we don't do very often. Ok?  So things like  um class registration or blood donation versus things that we do really often, ok,  like exercise. And then what they asked was, "Well, how effective were these interventions?" The ones that were trying to target people's motivations their intentions, their knowledge. How effective were they for these two different classes of studies?

And what they found was that for the really infrequent behaviors, stuff people don't do very often, intention based interventions work. Ok?  So this is what's called an effect size. And a, anyone who would want an intervention and get an effect size of of .74 would be really happy. They'd be like, "This this worked. This intervention changed people's behavior."

On the flip side, for behaviors that people perform really frequently in the same context, the effect size was dramatically smaller. So, so the basic takeaway is that, you know, trying to tackle people's intentions, arming them with information, getting them motivationally on board, will work if the thing you are trying to change is something that people do relatively infrequently.

But if they do it all the time, been doing it a long time and in the same environment, changing intentions, changing, providing knowledge will often have a pretty small effect. Ok, so ah you should be thoroughly depressed at this point about the scope to change people's behavior, that was my goal.

Umm, but what I want to pivot to now is well, is there anything you can do about it, right?  The balance of power gets switched to habit and then you can't talk to people anymore. You can't provide them with information. You can't motivationally engage them. It has little, little impact.

So what can you do? Fortunately, there are things that you can do and that's what most of our research has has focused on.

So the first idea I want to introduce to you is that context change ‑‑ and these can be both small changes and large changes ‑‑ can ah, upset, can disrupt the cues that keep people locked into doing what they've done in the past and create a little window of opportunity for intentions to regain control of behavior.

So here’s is a little suggestive evidence, um ah, just just to support this. So this was a study that was done in Europe. And ah what they were interested in was ‑‑ the ah this is sort of environmental ah study ‑‑ looking at whether or not people's concern about the environment predicted whether they used their, their car or not, the percentage of the time they used their car.

And so these are the results they found for the average person, which is kind of sad. Basically, what it shows is that ah, actually people who had high concern for the environment ‑‑ that's the green bar ‑‑ had slightly higher car use than people who had low concern for the environment.  Ok?

So not, not a very happy making result, um. That’s the average person. But then they found this other group who had their pattern flipped around, ok.  Where if you have high concern for the environment, you're in a green bar there, you use your car much less than if you have low concern for the environment.

And so, who are these people? Well, turns out they're people who are the same as the average people, but people who recently moved house. Ok? So the mere fact of moving to a new environment, disrupting all of those cues associated with what you've done in the past was enough, in a study, to move people's behavior to be more in alignment with their values and their attitudes.

So, this study's really interesting, it doesn't specifically look at, at habit. So, we've been eh, in our work, trying to drill down into this a little bit more um, sort of scientifically in a controlled way, and I want to share a couple of studies with you on that.

So, I want you to eh, engage in a little sort of imaginary exercise with me and and see whether you do the same thing that I do.  But, imagine that you’re, you've gone to the movie cinema. It's a, it’s a Saturday night. Um, and you buy some popcorn on your way in. And you sit there and the trailers start and before the trailers are finished all the popcorn is gone. Because if you're like me you've been mindlessly shoving it into your face.

And, ah and all the people around you are doing the same thing. And ask yourself the question, why on earth would we, why do we do that? Who’s who’s controlling our behavior when we engage in that? Is it the intentional self? Is it just that we're so overwhelmingly ah influenced by the delicious, irresistible taste of the popcorn that we've got no choice but to eat? Or, or is it instead something about the environment and the ritual and the practice of being in that setting and and an association with with eating that particular food?

So that's kind of what we were interested in in this study. And ah, so what we did was, eh we got a bunch of people and then we randomly assigned them to do one of two things. Watch a movie, watch movie trailers in a cinema, just like the scenario I just described to you there, or, the control group watched music videos in a large, darkened room.  Ok?  So think about these two groups of people. They're both sitting in large, darkened rooms, but one of them it's an environment that the mind associates with consuming popcorn, that is a movie cinema, and in the other it's not. OK?

And then, as social psychologists like to do deceptive, manipulative, mean things to people, um, we gave...


David:  We randomly assigned, randomly assigned people to get a box of popcorn. And for some people it was beautiful. It had just been popped in half an hour ago. It was warm. It was buttery. It was delicious. And the other half of the people, the popcorn had been sitting in my office for a week, um. [audience chatter]

David:  In sanitary conditions I hesitate to quickly add. Em, so people are randomly assigned to get this fresh or this horrible style popcorn. And they're in one of these two environments. Ok? And so, and at the end of the study, we we got a measure of whether you are the sort of person who always buys popcorn when you go to the movies or rarely do that. So there's a measure of kind of a whether you're kind of a habitual popcorn eater um or not. OK.

So first of all just to ah, to reassure you um, that the manipulation kind of worked. Ah, people do indeed find fresh popcorn nicer than stale popcorn. I won't be winning any Nobel Prizes for that finding.


David:  But it checks the box that the people noticed that the stale popcorn was not very nice. But then the question is, ok people disliked the stale popcorn, but what did they do behaviorally? Ok?  So let's look first at that group of people who were in that environment that for some people will be associated with consuming popcorn. Ok?  So sitting in a, in a cinema watching movie trailers.

And let's look first of all, these are the people who are the non‑habitual eaters, who said no I don't actually buy popcorn that that often when I go to the movies. Sometimes yes, sometimes no. Well they acted as you would expect, rationally right? They ate most of the popcorn if it was fresh. And they didn't eat most of the popcorn if it was stale. Ok?

Now let's look at the habitual eaters.


David:  They ate exactly the same amount, exactly the same amount regardless of whether it was fresh or stale. Ok. But now the interesting thing is, what happens when you take those people and you put then in an environment that isn't for them associated with eating that food, ok? So let's look at the control condition now. So first of all here are the non‑habitual eaters. They behaved basically the same as they did in the other study. Ate it if was fresh, didn't eat it if it was stale.

But now let's look at the habitual eaters. So now suddenly the habitual eaters are eating it if it was fresh and not eating it if it was stale. Ok? So just shifting these people out of the environment that is associated with their routine was enough to restore control to their preferences, their attitudes. And they were like, "I'll, I’ll eat it if it's fresh and if it's not I won't eat it."

Ok, so that's context change. But often in an, in an organization it's not possible to go and make everyone move their desks or move to a different location, or. Context change is not always a practical thing to do, either at an individual level or at an organizational level.

So what are some of the other things that you can do? Well, if you can't change the context we've also looked at the idea of um doing something to disrupt the action sequence. OK? One of the other things that fuels habits is that people ah, end up with a kind of sequence of actions that get tied together. One step in the process triggers the next step in the process.

Imagine yourself getting into the car in the morning, that there are actually a whole bunch of different actions you perform totally mindlessly but you do them perfectly and in sequence. One thing triggering the next in the sequence.

So, if you can't change the environment you can do certain things to alter the decision process in some way. Add a step in. Mess up the normal sequence. And, at least we hoped in this study to find, ah to see whether that would also help people um overcome the influence of habit and act out of intention.

So, What we did was ah, again got a bunch of people, put them all in the movie cinema, and then randomly assigned people to eat either with their dominant hand or to eat with their non‑dominant hand. Ok? So you're literally just moving the box of popcorn from one hand to another. OK? That's all you're doing. So you're using a different hand to eat.

But it was enough we thought, maybe, to just mess with the action sequence enough to give people a little window of opportunity to do something different. And again we gave them stale or fresh popcorn.

So first of all, here are people eating with their their dominant hand, so this is the sort of the normal, non‑disrupted way of eating. And we replicate the effect we saw in the previous study. That is that if you're a habitual eater, you eat, in fact in this study people ate slightly more of the stale popcorn than the fresh popcorn, em.


David:  But again, they're acting out of habit. There is no influence coming from the freshness of the popcorn. But then here’s folks who, here’s folks who switched the hand they used to eat. Ok? So now suddenly you've got a rational, sensible gap between the fresh and the stale popcorn, people eating it if it's fresh, not eating it if it's stale. Ok.

So, the sort of interim summary of this would be that, um there are ways to put the intentional self back in control. And one of the things we can do is get in and somehow mess with the environment. Alter the environment. Remove cues that have been ah associated with triggering the old way of doing things.

And so just to sort of think practically for a second about ways in which this might be you know scalable, we often think about um avoiding making changes to organizational practices when many other changes are taking place. But I think one of the implications of this type of thinking is that, actually, it can be a good thing to try and make new changes either with or just after other major structural changes have taken place.

Especially if they involve moving people out of environments that they've been stuck in for a long time, physically changing the environment in some way.

So perhaps timing ah, interventions to occur shortly after those kinds of major structural events might actually be a good thing rather than a bad thing, which might be the standard current thinking.

Or instead we can do something to disrupt the familiar action sequence. OK, so that could be, obviously here we were thinking about very simple, low‑level behaviors, but within, within a team it could be ya know doing something to alter the normal sequence of decision making. Adding a person in, changing the order of a step, might be enough to dislodge people from just ah, running through the normal, habitual pattern of thinking and instead do something else.

One I think important note here is, throughout this research, is that these sorts of habit disruption techniques only work if the intentional self is already aligned with the desired change. Right? So we saw in those, in the popcorn studies for example that um, making people ah, moving them to a new environment or making them switch hands didn't drop the total amount of food people consumed. It just made them sensitive to whether the food was fresh or stale. Ok?

So in other words, if people's motivations and intentions are still pushing in the "wrong direction" then messing with the environment, messing with the action sequence, won't help.

So we really are, we really emphasize in this work that um, not that these techniques are intended to replace strategies based on information, motivation, um and intention change, goal setting, but to augment them. And so if you can successfully change people's intentions ‑‑ and ‑‑ disrupt their old ways of doing things, that's, that’s the optimal sort of strategy to take.

Ok, so in the, in the um time I have left, I wanted to just share with you um three more simple and interesting ways that we've come across either in our work or in other peoples' work that people can kind of escape the the tyranny of this habitual self when it's not working for them.

Um, and the three ideas are, that the first one is the ah, the first one is the question of whether or not just by being mindful, just by being vigilant in our thinking, consciousness‑raising, can that have any positive effect on habits? And if so, in what, in what conditions?

Um, the second idea is ah to look at the value of forming very specific, what are called if‑then plans, and I'll talk to you about what those are.

And then the third and final one is the idea of using kind of creative perspective‑taking as a way of getting outside of the well‑worn ways of doing things and creating space to do something new.

So let's tackle the first one of these which is sort of simple awareness and vigilance. So, earlier I gave you that example of the visual illusion, right.  And that was a case where consciousness was basically useless. Right? You couldn't, you couldn’t just by knowing that the tables were the same length, you couldn't force yourself to see them as the same length. Ok? Because that automatic visual system is so strong and so overpowering that it dominates what you see regardless of what you consciously know to be true.

Fortunately, the same is not true for habits. There is more scope to use mindfulness and to use what we call sort of vigilant monitoring or vigilant mindfulness to create a little bit more space to ah, to overcome old ways of doing things.

So, in summary, in these studies, what we've normally done is gone out into the field and ah, recruited a whole bunch of people who are trying to change behaviors of different sorts, and then just ah, measure the different strategies that people naturally use. Ok.

So the top ones when people are trying to change behavior, so maybe reduce snacking, or making themselves exercise more, or giving up or reducing alcohol consumption, or cigarettes. There are there are sort of three main strategies that people use. One is distraction, ok and so they try to do everything possible to do something other than the problem behavior.

The second one is stimulus control, so they avoid any exposure to the to the cue. So they go and hide the um, the ice cream, for example, or um make sure they don't bring it into the house in the first place.

And the third strategy is what we call "vigilant monitoring" or "vigilant mindfulness." And that’s basically the strategy where you, instead of trying to force the unwanted behavior out of mind, alright, or avoid exposure to the cues. Instead you actually direct as much mindfulness to it as you possibly can. And you sort of cultivate an orientation of of vigilance towards whatever you think is the cue that is triggering the unwanted response.

And, and all of those strategies work to some degree. Distraction doesn't work very much. Stimulus control does work, so making sure that the cookies are hidden at the very back of the cupboard works to some degree, but the most, the most effective strategy is this vigilant mindfulness.

Ok, but what we found in these studies is that it is really important that you're mindful about the right thing. Ok? And so the right thing is knowing what are the cues, and that could be the people, the places, the times, the decisions points that are triggering you to go down that well‑worn path, that you've decided you don't want to go down anymore. Ok?

And then there’s two other things that I think are really important to direct consciousness to. One is that we know, not just from our research but from ah 40 years of research in psychology, that humans have a very um a very powerful tendency to mistake things that are familiar with things that are good, or to infer, another way of thinking about that is, we have a very compelling tendency to assume that because we have done something before, that it must reflect our choices and our preferences and our values.

Ok?  So simply being aware of that we have this sort of this inherent bias to partly infer our preferences by looking backwards in the mirror and saying, "What did we do in the past?" and that may actually be a sort of cognitive bias rather than a valid insight, I think can be another thing that gives you a little bit of scope to, to do something different.

And then the third one is that we know that when people are acting out of habit, when they’re highly routinized in a domain, it also forces them to shut down searching for new information. Ok? And so that new information, information that counters what you already know, is much less likely to be processed because you don't direct any attention to it.

And I'll give you a little demonstration of this. So what I'm going to do is flash up some words on the screen, all at once, will be about 10, and I want you to see if you can remember as many of them as possible. OK? So I'm just going to put it up for about two seconds, so be ready.


David:  Ok, so who thinks the word "sandwich" was on the list? Maybe about a fifth. Who thinks the word "bread" was on the list? OK. So more seems about half thought that "bread" was on the list and maybe a third thought that "sandwich" was on the list.

"Sandwich" was on the list, "bread" was not on the list. Um ah, yeah, so.

So what's interesting...? I'll go back just again, so you're sure. That's the original list.

Audience:  [laughter]

David:  Just so I didn't cheat. I didn't cheat. Ok, so why on earth do we do that? That's very weird.  Right?  We do that, because obviously this set of terms is so familiar. Alright? It makes up a very well worn, comfortable, familiar sequence of events that we're all very accustomed to.

Ok? And so what some of you did, the majority of you did, was filled in what was actually a missing piece of information by seeing "bread" as having been on the list when it actually was not. Ok? And obviously, this is just a metaphor, but we do exactly the same thing every day in decision making when we are acting out of habit.

That is we see, we do two things actually, we see information as being there that is actually not to support the old way of doing things, and we also miss information that might point us to doing something else. So I think to sort of tie those that sort of idea about how does mindfulness, how does vigilance, how does consciousness help in this process, I think it's really these three things.

The critical things are having a process where you understand what might be the points in this process, where we're being cued to do, to just repeat the same old way of doing things. The second is to be aware that there will be a tendency to assume that because something was done in the past, that it's the way things should be done in the future and is the optimal way of doing things. And then the third one is that we are probably going to ignore information that is there, and maybe see information as being there that actually is not.

Ok, so the the second idea is, um in terms of overcoming ah habits, there is some very good research now on something called "implementation intentions," you may have heard about these before. Um and this is essentially a memory phenomenon. Um and ah what was, what is being discovered by ah researchers in cognitive science is that we do a much better job carrying out future plans if we don't just set an intention like, "I'm going to eat more healthily," but we set an intention that is connected to a specific environment or cue. Ok?

So for example, if a person is trying to lose weight, it is going to be much more effective for them to say, "Whenever I go to a restaurant and open a menu, I'm only going to look at salad options," then for them to walk into the restaurant and say, "I'm going to eat healthily now." Ok?

So there’s something very powerful about tying the intention to an environmental cue, and this is an area of like active ah research at the moment. There is some data showing that this works for habits too, right and for the very reason that, especially if you can find the right cue that’s triggering you to do something, triggering you to act in the old well‑worn way, if you can kind of re‑appropriate that cue and link it with an intention to do something else that is a really effective way to go.

And so just so that you know, implementation intentions have been shown to be effective for a whole bunch of things, but here is just an example.

So certainly for making yourself exercise, doing things like recycling, and there is also very good data showing that ah implementation intentions work for helping people in controlled gender and race stereotyping,ok. So it’s a little bit, so just to sort of tell you practically what that would involve.

Um it sounds a little heavy handed, but basically in these studies on gender, gender and stereotype control, ah people would...I know you've heard about these studies before, where ah people are looking at vitae's of applicants for jobs, right, and they will sort of randomly change the name to make it female or male.

It is actually the same vitae, but they just randomly rotate it. And it's a way of working out you know whether someone has a gender bias because they keep giving the job to the male, even though the genders were randomly assigned.

Um so in these studies, those sort of biases can be reduced significantly by just having a process where people have formalized and signed onto the idea of ah ignoring race as a cue in the decision process.

Ok so the final idea that I wanted to ah tell you about is this sort of idea of creative perspective taking. Um and to sort of bring this point alive to you, I'm going to do one little final exercise with you. And so I'm going to like randomly assign ah people in the room, and I want you if you were born between January and June, to imagine how you might solve a problem that I'm about to give you. Ok?

If you were born between July and December, I don't want you to imagine how you would solve it. I want you to imagine how the person sitting next to you might solve it. Ok? Is that clear?

I'm going to describe, I’m going to describe this problem, and from January to June, you need to imagine yourself in this situation and how you would solve the problem. And if you're July to December, you're not thinking about yourself. You're thinking about the person next to you and how they might solve it. Ok? So, you ready?

Here's, here’s the problem. Ok? So a person is in jail. They have a piece of rope, but the piece of rope is half the length required to climb out the window and down to safety. The person cuts the piece of rope in half, and climbs to safety. How did they do it?

Take ten, I'll I’ll repeat it. So this person is trapped in jail. It's either you or the person sitting next to you, trapped in jail, and has a piece of rope. The piece of rope is half the length required to climb out the window into safety. So what the person does is cuts the piece of rope in half and climbs to safety. How did they do it?

OK. I'll tell you in a second what the, what the answer is, but I'll just give you 10 seconds in which you're thinking about yourself in this situation or the person next to you.


David:  Alright does someone want to...? Sir, what do you think the answer is?

Male Audience Member:  He split it. So a rope has two parts, then he ties them together.

David:  Right. That's right. So cutting it in half, means cutting it lengthways, not cutting it across. Ok? Cuts it lengthways, ties the two pieces together and climbs to safety. That's the, that’s the answer. So I want to ask now of the people who are imagining themselves, so if you were born January to June, can you tell me how many...Put your hand up if got that solution.

OK. I'm going to do a highly scientific poll here. Ok, and if you were thinking about someone else doing it, put your hand up if you solved it. Ok I can't tell at all.

Audience:  [laughter]

David:  But I can't tell you, it looks like about half. But what I can tell you is what the data shows, and the data shows that people are much better at doing this kind of insight problem if they imagine someone else grappling with the problem than if they imagine grappling with it themselves. And why would this be the case? Well, the argument is that um stepping outside of the self, literally engaging with someone else's consciousness, projecting into someone else, is a way of escaping, ah escaping the familiar well worn, uncreative ways of solving problems. Ok?

And so it’s amazing to think that something as simple as, instead of you know if you're confronting a problem and trying to act in a novel way to come up with some new solution, to think that one way you can do that is as simple as not thinking about, "How am I going to solve this problem?" but thinking about how the person next to you might solve it.

Ok, so I'm close to the end of my time. Um I want to just finish by summarizing this, and saying that I hope I have convinced you in this talk that habit is one of the the main factors that explains this all to common misalignment that human beings have between their attitudes, intentions and values on the one hand, and the behaviors they enact on the other, and that we have reason to be optimistic about our scope to intervene in that process.

And so some of the things we can do are get in and mess with the the context, the people, the places, the action sequence, the decision points,

Mess with that process in a small way to create a window of opportunity to do something else.

We also saw that mindful vigilance can help. This isn't as bad as those visual illusions. Being conscious can help, ah especially if you’re focusing on the right cues and you’re aware of the other biases that habits involve, like making you feel like the familiar thing is the best thing and making you ignore information.

And then finally, we can make progress by forming these very specific "if, then" plans and by sort of creatively departing the self and projecting into other people as a way of thinking outside the box and doing something new.

So with that, I thank you very much, and I'm very happy to open it up to, to questions.

Audience:  [applause] Female Audience Member : I read a study in um, science sometime last year, and it, I wonder if this applies. It had to do with when you go into a room that is different, that you sort of reset. And this would be a case where, you know, you forget what you went into the room for.

David:  Yes. Right.

Female Audience Member:  And so, since I read it and I'm not rereading it, I may not have all the details right, but would that have anything to do with...? It would be a habit in not having a familiar...? But then when you go back into the other room, you can figure out why you were going into the other room.

David:  Yes. That is exactly right. That is a very interesting question. So that research is about, it’s called "event boundaries." And the basic idea is that people, you know, if we think of ourselves in computational terms, people timestamp um timestamp behaviors and link them to specific places. Ok? So you are much more likely to create an event boundary when you move from one room into another, which is why you can, you know, have known why you were going into the garage before you went to the garage.

You cross over that event boundary, and suddenly you are in a new space and you've got no idea what you were, you were doing. Alright? So it’s a, it's a similar phenomenon, and it makes, I think it reinforces that point that um, space and memory and space and action are very tightly bound up, and we don't give the, we don’t give the environment the credit that it deserves as helping us sometimes and hindering us at other times in doing what we want to do.

Female Audience Member:  So um, if you're in a setting where you are trying to make a change to habitual behavior, I wonder if it's helpful, like suppose you also have a context change going on, you know, something's happening at the workplace. Is it helpful to present people with data that explains that context change is a good time to do habitual change?

David:  Uh huh, that's a really good question, um eh, but a complex one.  So, the the basic challenge is, you remember I said earlier on in the talk, I said that our sort of intentional self is a rule‑based, verbalizable, declarative memory is the word we use for it, um system.  And that's not the same system that our habits are stored in. Right?  Which is why I can't if I was a brilliant concert pianist, I couldn't give you five rules that would immediately make you able to to play a Chopin piano concerto. Right?  Um so to to a lesser extent, the same problem arises here, and that is that can you informationally um ah convince people to to change their habits? And I think you can do multiple things. First of all, as I said throughout, you've got to get people motivationally in the right place so they're interested in changing in the first place. And so that data you're talking about could be part of that case. Right?

So, I see those things as sort of interventions to make sure the intentional self is on board first.  Right?  And then people also understand why you're doing this weird thing of making them, you know, deal with additional change while other changes are already taking place.

So that could be part of building the case for the intentional self, but those interventions by themselves are not going to disrupt the habitual self.  Ok?  So that wouldn't be enough in of itself. In addition, you'd need to be doing things to alter the environment, um mess up the normal decision making process in some not too irritating way, but enough to create a little space for people to do something different.

So that, that's the way I would tackle that.

Female Audience Member:  So, I'm a department head and I'm thinking of this very practically. So if I always have my faculty meetings in the same room, and people sit in the same seats and they have the same bad behaviors. Can I, can I move to a different room and break up some of that? Should I tell them why I'm doing it, or just do it and see what happens?

Audience:  [laughter]

David:  Well that, that's great. Um well as an experimentalist, I think you should do it without, without telling them, and just observe the results. No, but um absolutely, I think that again coming back to this issue though of if, there’s um context changes is is great, but if the underlying ill will is so overwhelming, context change itself won't do anything for that. So, that's why we’re really talking about a parallel, like a push on two fronts in parallel. So an intervention that, you know, addressed all of those tensions, and and then paired that with environmental change would be the optimal thing to do, because you’ve, you've made intervention with the intentional self and an intervention with the habitual self, which really gives people the best opportunity to press restart.

Female Audience Member : I guess I'm also ah thinking about the practical side. And I wondered if you had some advice for us about what we do now, in terms of fixing our situation. Women in IT and attitudes and things? It sounds like people would have to really want to change to change.

David:  Well, yeah that's true. Although the, I mean you guys know this this area much better than I do, I but I take seriously those those Gallup data.  Um I note they probably overstate the case. Ah, I know too of other very good scholarly work looking at the the leadership model and the the gender neutral drift that has taken place in the leadership model over time. Right?  So, it started to be the case that people's stereotype of a leader and people's stereotype of a male, basically overlapped almost 100%. Alright?  And ah, and so stereotypically female traits were not really represented in the leadership stereotype at all. And there are great studies showing at how the leadership stereotype has expanded and um now is much more gender neutral.  Alright?

Some studies even show that it has more a female stereotypic trait than male. So to me that's ah, affirming in that it suggests that the intentional side of the equation has made made significant progress, not that it's where it needs to be, but that there has been um progress there, and there's lots to feel good about.

The question is now, how do we structurally roll that out and make sure that it gets reflected in in outcomes? And I think that's where um some of this work and, you know, ah, a whole range of other more structural interventions and policy type interventions are going to be necessary, because I think that this shows pretty clearly that hearts and minds is one thing, but that by itself is not enough.

So drawing on these multiple fields, including this work, to say, "How can we go beyond just changing hearts and minds"?

Female Audience Member:  Um, learning is good, we tell our students, but there are many habits, bad habits that interfere with their learning, such as, ah texting or being on Facebook while doing your homework. Um, have you investigated student learning, through the lenses of uh your framework?

David:  That's a great question. We, I have not. Others have done that, and there are some very interesting study, actually some very interesting work gender ah differences in um sort of attentional control and multitasking. You may have seen this data, but ah there's some some good cognitive neuroscience data showing that women are better at juggling multiple tasks, um and showing that at the level of sort of underlying memory systems that...um yeah, so that's an area that's under active investigation. I haven't been involved in that work, but I think it's really interesting.

Female Audience Member:  I was just wondering how you would take your frame of looking at things and apply it to, for instance, the civil rights movement? You know, prior to the voting, um there was some people who maybe thought it was changing the hearts and minds.  And how does someone like you, from your perspective, um approach issues of social change? Um.

David:  That's a great question. It's a horribly difficult question, but thank you for asking it. [laughter]

David:  Um, I think, to be honest, what we, we don't think about it so much in terms of social justice. We think about it ‑‑ although it's a very legitimate way to ask the question ‑‑ um we think about it more as policy and um what kinds of policy interventions. We, our work in this area has made us ah very convinced that top down policy interventions, wherever possible, are really ah, are really the best way to go. So what I mean by that is things like um you know, with ah making cigarette packs out of view, um and things like you know, banning um the huge soda containers.

Um because, just, and the reason we have reached that view is that, looking at the data, we just know that this habitual dimension of life controls so much of what we do that it really does mean that um people by themselves are often not able to change. And so if you can make policy interventions that really make it much harder so the people have to sort of effortfully opt into doing the wrong thing that is, ah that is, you know a much more powerful way to go.

There’s this wonderful data looking at um organ donation data across European countries, and you might have seen it. And, you know there’s this weird, weird effect like some countries in Europe have organ donation rates that are close to 100 percent and others have the averages of about 12 percent. And the top is like 27 percent, and that was after a kind of massive public information campaign to get people to sign on to to donating their organs.

And so, why is there this huge gap where some countries have 100 and some of them have like 10? And the reason is simply that on the licensed registration forms, those countries have opted out. Um, the countries that have 100 percent have an opt out, and the countries that have 10 percent have opt in.

And so ah, to me that’s just, you you would think that something as as fundamental as, "Am I going to donate my organs after I die?" would be something that people would make an active effort for conscious choice about, but they they don't. So these little structural nudges whether it's something like that on a form or whether it's a policy intervention, ah, you know, is just much more effective in moving people in the right direction.

Jane:  Sorry.

David:  OK. One on your left.

Male Audience Member:  I was curious to know if your research is gender equal or does it have a difference, because then we could apply some of this to educational settings.

David:  That's a great question. So, just let me clarify so that I understand what you mean,  so do you mean, do we ever find, do we include both males and females in the research and then do we find any gender dif, gender differences?

Male Audience Member:  If you found gender differences, if females are more of this type or that type?

David:  Yup, that's a great question. We have not, we we always test for gender effects and we basically never find them.  We have, um and certainly they're never systematic. Um, so, and more generally we, it's very difficult to find um any sort of demographic variables that determine whether you're a more habitual oriented person or intentional person. We pursued that idea for a long time and ah and found basically no evidence for it. And there is a good reason for that, which is that, that memory system that we have, that implicit learning memory system, it's so old in evolutionary terms, that the idea that it would, you know, for example it's uncorrelated with IQ.

Your ability to learn information in a habitual way is totally uncorrelated with IQ. So, there’s um, it's really such a sort of, it's almost like your lungs, the idea that there would be big demographic differences across something as fundamental as that. So we haven't found them.

Jane:  Everybody please join me in thanking. [applause]

David:  Thank you...

Transcription by CastingWords