Category Archives: Theory/research

Choice behaviour: governed by self-control resources or something else…?

Several of my recent posts here have been about self-control, the mental resource that allows you to make choices that align with longer-term, more ideal goals rather than ones that prioritise shorter-term and more immediately rewarding options – like deciding to cook something healthy for dinner (which can be time-consuming but is in keeping with the longer-term goal of having a healthier lifestyle) rather than grabbing a burger and fries from some fast food place (which, although convenient and perhaps more hedonically rewarding for some people, isn’t necessarily a great choice from a longer-term perspective). I’ve described some of the many, many ways in which the self-control resource might be temporarily depleted, and I’ve outlined some strategies for how to improve your chances of still succeeding in making the more ideal choice even when your mental resources of self-control have been worn down.

I guess it might be surprising then if I were to tell you that maybe self-control isn’t a resource – maybe it isn’t a thing that gets used up and then replenished, depleted and then restored.

When trying to understand human behaviour from the perspectives of psychology and neuroscience, it’s useful to try to formulate models and concepts to test, as this gives researchers a framework with which to work and allows them to examine how this framework responds to different challenges and manipulations. The framework is usually devised initially as a way of explaining a bunch of existing research results that seem like they might all be the product of the same, or similar, processes in the brain. Unsurprisingly, though, when you’re trying to figure something out for the first time and you’re really not sure what’s going on, you might come up with a framework that seems to describe the existing results, but ultimately it might prove to not be the right explanation. As more research is done, new results might not seem as willing to fit into the framework as you expected. It’s not that the framework was horrible and terrible and obviously ill-conceived from the start – maybe it honestly did explain all the results at some point in the past. But as we develop a better understanding of the underlying processes, and as we refine our knowledge and do increasingly sensitive experiments to flesh the framework out, we might start to realise that it was kind of a coincidence that the original framework explained the results, and we need to come up with alternative ideas and theories that explain not only those older results, but the newer, awkward ones too.

It seems we might have reached that point for re-modelling the theoretical framework of self-control (and, as likely as not, we will reach that point again and again as our understanding progresses and the framework needs reworking and refinement). Inzlicht and colleagues put forward an opinion piece, ‘Why self-control seems (but may not be) limited’, in the journal Trends in Cognitive Sciences earlier this year, in which they proposed an alternative way of thinking about self-control: when you have done something mentally difficult and demanding, your brain wants to balance that out by driving you to do something more enjoyable and rewarding. So when you perform a task that takes mental effort and focus, you might then be motivated to switch to doing something less taxing and more inherently enjoyable. Inzlicht and colleagues propose that you aren’t using your self-control “resources” up when you make the decision to prioritise short-term, rewarding goals; your brain is just saying “enough of this difficult thing, let’s do something fun”. It’s about switching from one mode (being willing to do something challenging and difficult, things that you have to do) and switching to a different mode to balance that out (taking the easy and more rewarding option of doing the things you want to do).

Inzlicht and colleagues argue that this mode-switching between “have to” and “want to” would drive useful and adaptive behaviour, particularly from an evolutionary perspective. For example, it would have been evolutionarily beneficial for individuals to not become too easily engaged in focusing on tasks, but to be spontaneously driven to disengage from tasks and seek other opportunities, and thereby learn more about their environment by being motivated to explore it. However, I have to say that I have never been a fan of evolutionary explanations in psychology – there is far too much room for speculation and not enough basis for actual testable hypotheses.

While the evolutionary value of “have to” vs. “want to” modes remains a possibility but is quite impossible to test directly, Inzlicht et al. suggest a more intermediate account (rather than the grand overarching evolutionary account) of why these modes might be useful:

“Balancing exploitation versus exploration serves an adaptive function that leads to a general tendency to prefer balance between cognitive labor and cognitive leisure, or between mental work and mental rest. According to labor supply economics, the fact that time is limited leads people to prefer some optimal balance between externally rewarded labor and inherently rewarding leisure. Recent research suggests that cognitive control is intrinsically aversive, having inherent disutility.”

Furthermore, Inzlicht et al. remind us of the mind’s tendency to be very easily influenced, and how this influence can be used to make the effects of “self-control depletion” miraculously disappear. For example, Job et al. (2010) found that a person’s beliefs about whether or not self-control is limited affect their actual ability to implement their self-control. If you don’t believe self-control is limited, you’re not as prone to lapses of judgement that would usually be attributed to depleted self-control. And if you think it’s limited but are simply told by researchers in a study that it’s not, then you start to make fewer of those lapses of judgement too. In short, your beliefs on the subject can influence your behaviour, which shouldn’t really be the case if self-control is a fixed resource.

So we are perhaps driven to engage in more enjoyable, rewarding pursuits not because we have used up our self-control resources and have experienced an inability to fight off a lapse in judgement, but simply because the feelings of fatigue, boredom and negative emotion that might come from doing something mentally demanding prompt us to seek out something more enjoyable and leisurely. This is an interesting alternative to the “resource” theory of self-control and seems to make sense at face value. However, a lot of research still has to be done before we can dig a hole for self-control resources in the graveyard of punctured theories. Inzlicht et al.’s paper was an opinion piece, and while their concept of mode-switching does explain a lot of the results they discussed, there is much to be tested within the rough framework they have proposed in order to methodically and comprehensively flesh it out.

For example, if our brain undergoes switches between “have to” and “want to” modes, then the research so far hasn’t really explored what prompts the switch back in the other direction – from the leisurely and enjoyable “want to” to the mentally demanding, task-focused “have to”. Is taking the easy, enjoyable option eventually so boring and dull that we want to be more challenged and we switch back to the “have to” mode? Self-control research has primarily focused on lapses of self-control, whereas switching from “want to” to “have to” mode would be a reassertion of self-control. Research simply hasn’t looked much at that to date.

And what sort of timecourse does the mode-switching follow? How much is it driven by internal factors (i.e. just natural fluctuation in the state of the brain) and how much is it driven by external factors (for example, you switch into “have to” when you suddenly realise you’ve got a deadline and you can’t procrastinate any longer). Is the timecourse of mode-switching predictable? How can we track it?

But perhaps the thing to keep in mind is this: despite the proposition of this hugely different framework, and despite the chance that it might ultimately prove to be a more accurate explanation of behaviour when someone chooses a short-term, inherently rewarding goal over a long-term, more ideal goal… it doesn’t really change any of the strategies for dealing with that behaviour. In my previous posts on self-control, I mentioned that evidence suggests strategies such as having a mental plan to enact when you feel tempted to go with a less-than-ideal choice. Regardless of the framework for understanding that choice-making behaviour, having a mental plan helped people prioritise the longer term, more ideal goals. Did it do that by helping to somehow replenish self-control resources? Or did it do that by somehow encouraging the brain to stay in the “have to” mode? There are a lot of hypotheses to generate and test before we can tell whether the mode-switching framework better explains behaviour than the self-control resource framework, and maybe even both could ultimately be wrong. Time will tell!

Mortality o’ersways their power: effects of mortality salience on choice

Self-esteem is a powerful driver of behaviour. When your self-esteem is diminished – which can result from myriad different causes, from failing to achieve a task you set for youself, to a harsh comment from someone, or any situation that makes you feel vulnerable or disempowered in some way – you are strongly motivated to rectify the situation and restore your self-esteem. From a psychological perspective, the main way to do this is by reasserting your identity and beliefs in some way, either by disparaging the identities and beliefs of people who don’t share the same ones as you, or by doing things that bolster yours.

What does that mean for consumer behaviour? Previous research has a lot to say about how people change their consumption choices when their self-esteem is diminished by mortality salience – that is, being reminded of their mortality. It might seem harsh to be reminding research participants of their mortality, but a lot of us probably have our mortality made salient multiple times per day. Reading a news article about a fatal car crash or watching a TV show about a murder investigation would be examples of things that would prime us to be more aware of our mortality, and perhaps respond in slightly different ways to things we subsequently encounter. It seems subtle, but that sort of priming can have a powerful effect.

Mortality salience can also come from profound, obvious events – a fair amount of mortality salience research was done following the terror attacks of September 11, when people were reminded acutely and en masse of their mortality. Arndt and colleagues have discussed why consumption in the US jumped by 6% in the months following September 11, suggesting that people acted to enhance their self-esteem via materialism and consumption. The relationship between mortality salience and consumption behaviour is a bit more complicated, though – as you might guess, you’re not going to improve your self-esteem and reassert your values by doing something you don’t care about, so if you’re very non-materialistic then your response to mortality salience won’t be to indulge any materialistic urges. We would act according to our personal values and according to how we think those vaules are perceived by others.

Self-control, which I previously wrote about here and here, could also play a role. When self-esteem is challenged and you make efforts to reassert it, self-control resources could be depleted. That means that if your priority is to bolster your self-esteem by reasserting key parts of your identity, you might funnel your cognitive resources into the domains of your identity that matter most to you, which could leave hardly any cognitive resources for the domains that don’t matter. For example – if eating well is important to you but being financially prudent isn’t, a hit to your self-esteem might cause you to focus more intently on purchasing and eating good food, but maybe you spend way more money than you needed to on way more food than you need because saving money isn’t important to your identity and you self-control was diminished.

These ideas about mortality salience, choice, self-esteem, personal values and self-control were tested to some extent by Ferraro and colleagues in their paper “Let us eat and drink, for tomorrow we shall die: Effects of mortality salience and self-esteem on self-regulation”. The researchers investigated how mortality salience affected people’s choice-making, taking into account whether the participants might base their self-esteem on things like their body image or their virtuous behaviour (such as donating to charity or making a socially conscious and responsible decision). Participants had their mortality made salient by being asked by the researchers to do things like recall what they were doing and how they felt when they first heard about the September 11 attacks.

The results were in keeping with what you might expect: for example, increased mortality salience caused women to choose chocolate cake over fruit salad if they didn’t strongly base their self-esteem on body image (relative to a more neutral control condition, where mortality was not made as salient), whereas women who did base their self-esteem on body image tended to opt for the fruit salad rather than the chocolate cake (very interestingly, mortality salience had no effect on the chocolate cake vs. fruit salad choice in men, which demonstrates the importance of considering potentially relevant differences between particular groups). In another experiment, participants were given the opportunity to donate part of the money they earned from participating to a charity. Participants who had scored highly in terms of valuing virtue and charity and social consciousness in themselves donated a greater portion of their earnings in the high mortality salience condition (relative to a control condition), whereas those who scored lower and didn’t value that sort of virtue in themselves as much donated less when their mortality was made salient. In short, the evidence supports the notion that mortality salience affects behaviour and causes people to make choices in line with the personal values on which they base their self-esteem.

It is impossible to avoid being made aware of our own mortality in all sorts of ways in daily life, so mortality salience will inevitably affect our behaviour in some way, whether it be subtly or overtly. However, simply being aware of this and being mentally prepared might minimise the impact. For example, as I’ve written about previously, having a basic plan in the back of your mind can help in situations where you might be tempted to priortise short-term, less ideal goals when your self-control is depleted – now that we are aware that a hit to our self-esteem might suck up a lot of our cognitive resources and efforts in order to reassert our personal values and beliefs, having such a plan to guide behaviour and minimise the likelihood of poor choices might be more valuable than ever.

Depleted cognitive self-control: impact on ethical decisions

I have previously written about how self-control can be depleted, leading to impulsive actions that prioritise short-term, less-than-ideal gains. Some research suggested that this might be because people fail to put together a plan of action that would prevent the lapse of self-control occurring, and that self-control might be re-established if, when in a tempting situation involving a choice between two conflicting options, we take the time to think about our long-term goals and the specific details of how we want to achieve them.

A lapse of self-control can have undesirable consequences for the individual. Persistently valuing enjoyable but unhealthy foods rather than prioritising longer term dietary goals could result in all sorts of negative health outcomes; persistently valuing the thrill of an impulse purchase over a more considered, thoughtful approach to buying items can result in a lot of money wasted and a whole lot of space taken up by things you don’t want or need. But what about further consequences to impaired self-control? When your self-control is depleted, what sort of wider impact might that have?

When it comes to self-control and ethics, it looks like there might be a problem. Gino and colleagues (including Dan Ariely – a research psychologist and behavioural economist whose pop science books on decision-making you might have come across before) conducted a series of experiments to find out what effects depleted self-control would have on moral awareness and behaviour (“Unable to resist temptation: How self-control depletion promotes unethical behaviour” in the journal Organisational Behaviour and Human Decision Processes). Perhaps unsurprisingly, they found that depleted self-control had consequences for behaviour and choices in a range of ethical situations – short-term, unethical choices with a quick pay-off were prioritised over longer-term, more ideal goals. You can imagine the potential magnitude of the consequences of that in the real world.

In one experiment, Gino and colleagues examined whether self-control depletion led research participants to cheat on a problem-solving task (they could cheat by overstating their performance, which would get them more money at the end of the task, so there was a financial incentive to cheat). Self-control was depleted by showing one group of participants a soundless video of a woman being interviewed, with irrelevant words appearing as subtitles at the bottom of the screen – these participants were instructed to focus on the woman’s face and ignore the words, which requires plenty of self-control (you can do the Stroop test to see just how difficult it can be to ignore written words and how much control you need to exert to ignore them). Another group, the control group, was shown the same video but weren’t given any instructions – they could look wherever they wanted on the screen. The results? 34% of participants in the self-control depletion group cheated by overstating their performance, whereas only 13.7% of participants in the control group cheated – a significant difference.

Subsequent experiments revealed that self-control depletion actually made participants less aware of ethical situations and ethics-related concepts – once self-control was worn down, the topic of ethics was kind of de-prioritised and placed at the back of the mind.

Still, in that initial experiment, only 34% of people cheated after being depleted in self-control. What could be an influencing factor in who does or doesn’t cheat when given the opportunity? A person’s individual sense of morality and ethics probably plays a role, and the researchers investigated this by using a questionnaire to assess how central a sense of ethics/morality was to each participant – how strongly their self-identity hinged on this sense of being a moral and ethical person.

Again, the findings are perhaps unsurprising. Individuals with a strong moral identity (ones who considered it important for themselves to feel that they were caring, compassionate, fair, friendly, generous, helpful, hardworking, honest and kind) didn’t cheat that much more when their self-control was depleted compared to when their self-control wasn’t depleted. People with low moral identity cheated much more after depletion relative to when they weren’t depleted.

It’s interesting to consider how this might apply to other ethical situations we might encounter in daily life. When, as consumers, we have the option to purchase something of ethically dubious provenance (for example, clothing from high street retailers who are known to have a poor track record in terms of ethical production lines or sustainable use of environmental resources), are we more likely to choose the less ethical option if our self-control has been worn down? The experiments conducted by Gino et al. demonstrated that self-control can be worn down in different, seemingly irrelevant ways (such as having to ignore the presentation of words, or having to write an essay that doesn’t contain certain letters of the alphabet) and still have an impact on ethical choices. So after a long day at work and exerting self-control in order to stay focused and productive, we’re all probably pretty cognitively depleted, and that might mean impaired ethical awareness and increased favour towards unethical options that offer swift pay-off. Gino and colleagues also demonstrated, in a final experiment, that resisting the temptation to cheat depletes self-control as well – so trying to prioritise an ethical choice might wear you down to the point where you make an unethical one. And this is complicated by the influential factor of identity – all these things may only be an issue for you in proportion to how important a particular ethical position or issue is to you and your identity.

It’s difficult to predict how generalisable the results of the Gino et al. study are to other ethical situations – cheating for monetary gain is one quite specific scenario, but there are plenty of other contexts and quandaries of an ethical or moral nature, and the stakes might be sufficiently different to change the effects of self-control depletion on moral/ethical choices (consider how complicated it might be to examine the effects of self-control depletion in a situation with ambiguous moral outcomes, such as the classic moral dilemma, the trolley problem). The main implication of this study’s findings, however, is that depleted self-control does seem to lead to prioritising short-term, unethical options that have quicker pay-off (such as receiving more money at the end of the experiment) over long-term, ethical options with less clearly defined pay-off (such as maintaining an ethical and fair approach in life and whatever satisfaction and benefits might be derived from that). Additionally considering the self-control depletion research I previously discussed, which found that lapses of self-control seem to be due to poor planning of responses to resolve the conflict between choices, a few recommendations might be made:

(1) Avoid having to make choices with an ethical component following an activity that might have depleted your self-control, e.g. if you aim to be an ethical consumer, try not to go shopping after a long day at work when you are likely to be cognitively depleted.

(2) If having to make that choice when you’re depleted is unavoidable (e.g. you have to purchase a birthday present for a friend, and the friend’s birthday is tomorrow, and you really want to give your friend the present tomorrow, meaning you have to buy something tonight directly after work, even though it has been a really tough day and you are probably very much cognitively depleted!), have a mental action plan of how you would achieve your long-term goal of being a more ethical consumer and consider those details when you are tempted to purchase an item that doesn’t fit well with your more ideal goals.

(3) Purposefully direct your attention to alternatives as part of your mental action plan. If you dwell too much on the less ideal item you’re thinking about purchasing, the effort to resist purchasing it could be enough to wear down your self-control until you cave in and buy that not-so-brilliant option anyway.

Controlling self-control: how self-control might be restored after you’ve used it up

Self-control – the ability to control our impulses and actions – is a critical factor every time we make a decision between conflicting goals. As consumers, we employ self-control in order to make informed, conscious choices rather than impulsive, spur-of-the-moment, often-regretted impulse purchases. We use self-control in our attempts to achieve higher goals, like when we endeavour to buy things that are closer to our ideal standards (e.g. ethically or sustainably produced items, or food that is healthy) rather than whatever provides the quickest and most impulsively gratifying pay-off (e.g. items that aren’t ethically produced or sustainable but are cheap or convenient, or food that is not healthy but is convenient and delicious). Self-control also seems like a dynamic and variable thing – we all know people we would consider to have particularly good or particularly poor self-control, but we also know that our own level of self-control can vary, maybe depending on what mood we’re in, how tired we are, who we’re with, where we are, whether the item we want to buy is full-price or on sale, or dozens of other factors.

So how does self-control work? It has been the focus of a huge amount of very interesting research, and over the years, a lot of work has been done to try to characterise how it changes in response to different contexts and circumstances. If I said “self-control is like a muscle – you need to use it to develop it,” that makes some kind of intuitive sense – it does seem like maybe the more you exercise your self-control, the better at it you might get. But what if I said “self-control is a finite resource – if you use it too much, you deplete it and don’t have enough”? That also makes some sort of intuitive sense – exerting that self-control takes effort (since it might involve suppressing emotions or strictly focusing your attention, which can be very mentally demanding) and effort can wear you down after a while. The research to date actually suggests that the latter analogy is the more accurate one, with self-control becoming diminished over time as it is exerted.

However, maybe we can still have some control over our self-control, as it were. Self-control might be a resource that gets depleted with use, but is there perhaps some way to mitigate the depletion? Is there a way to replenish the self-control more quickly than normal, or to prevent it from being run down as much?

To determine the answer to that question, it’s necessary to characterise self-control in a bit more detail. One theory proposes that self-control is composed of two stages: the recognition of the need for control, and the actual implementation of that control. When you give in to the temptation and go with an option that you know isn’t ideal, the cause of that lapse of self-control might have happened at either (or both) of those stages – you might not have recognised that the situation called for self-control and the option you chose wasn’t ideal in terms of your longer-term, more constructive goals, or you might have realised that but simply failed to do anything about it. To figure out if it’s possible to improve self-control, we need to figure out at which stage the lapse is occurring.

In neuroscientific terms, the two stages of self-control can be separated into two widely researched cognitive processes: the identification of a conflict between potential responses – the conflict stage – and the recruitment of the cognitive control needed to implement a resolution the conflict – the implementation stage. Different parts of the frontal cortex of the brain are responsible for these processes, which means that we could use brain imaging techniques to see if those different parts of the brain are activated during the different stages of enacting self-control. That is exactly what Hedgcock and colleagues did, as reported in a paper in the Journal of Consumer Psychology, getting their participants to perform a choice task while in an fMRI scanner. They found that when participants had their self-control depleted, activity in brain areas related to the implementation of cognitive control was diminished, compared to when participants had’t had their self-control depleted. However, depleting participants’ self-control had no effect on the activity in brain areas related to identifying conflict. This suggests that depleted self-control might be due to a depleted ability to summon up the cognitive control to do anything about the situation. So when you’re about to buy a chocolate bar instead of a banana as a snack, at some level your brain is probably doing a perfectly good job of going “this isn’t really in line with our long-term goals of being healthy, is it?” but you are then failing to summon up the capacity or motivation to suppress the desire for the chocolate bar, or to direct your attention to the banana and its health benefits, or to think closely about why it would be closer to your long-term goals to opt for the banana. You know it’s not the ideal choice, but you make it anyway.

Can we do anything about that? Hedgcock and colleagues tried to figure that out, introducing interventions that were designed to target the different stages of the self-control process. Participants initially provided some information about how strong or weak their preferences were for assorted healthy and unhealthy food items. After performing a cognitively demanding task in order to deplete their self-control, participants were randomly allocated to an intervention:

• In the conflict intervention, participants were told “… be mindful of the conflict between immediate desires and future health consequences of each option. Please write an example of this conflict below.” This was designed to target the conflict stage of self-control.
• Participants in the implementation intervention were instructed to “… be mindful of the behaviors that you will need to do in order to reach your health goals. Please write an example of these behaviors below.” This was designed to target the control implementation stage of self-control.
• Then of course there was a control group that received no intervention and were given no instructions (although in the interests of a more comparable control conditions, I think the researchers could have considered giving the control group a completely irrelevant, neutral task to write a brief response to).

Participants then indicated their preferences for the assorted healthy and unhealthy food items again. This allowed the researchers to assess whether any of the interventions had remediated the self-control depletion – less self-control would mean less healthy food preferences relative to the preferences indicated at the start of the experiment. If the conflict intervention or the implementation intervention worked at all, the preference towards unhealthy foods wouldn’t be as great as in the control condition.

The results: the implementation intervention significantly skewed participants’ preferences towards healthier food items relative to both the conflict intervention and the control condition, whereas preferences after the conflict intervention were not significantly different from the control condition. It seems that thinking about the implementation of efforts to achieve the more ideal goal might restore self-control to such an extent that achieving the more ideal goal actually becomes more likely.

Much more research needs to be done before we gain a thorough understanding of how self-control works, how it is depleted, and how it might be replenished, but the results of this study suggest that, on occasions when self-control needs to be put into action and choice needs to be made between conflicting options, it might pay to think about the behaviours that would need to be implemented to achieve the more ideal goal. Of course, it’s likely to be rather different in the real world compared to within the confines of an experiment – you probably won’t have researchers clearly and firmly instructing you to think about those behaviours carefully and write about them. To do that on your own would probably require (surprise surprise) extra self-control. But it might be of value to have plans and strategies for some of those awkward decision-making occasions that you can think likely to occur in your life, from not so brilliant dietary choices to impulse purchases at the shops, and anything where the short-term and long-term pay-offs are obstinately competing against each other to woo you.