Cognitive Dissonance and Anorexia: Optimizing Differently

Part 5: Asking optimization questions to move beyond dissonance reduction.

Posted Nov 25, 2020

“Imagine a world full of brains, and far more memes than can possibly find homes. Which memes are more likely to find a safe home and get passed on again?” (Susan Blackmore, The Meme Machine)

Variants on optimization questions and answers

In the previous parts of this series (starting here), I’ve offered an account of how cognitive dissonance can help us think differently about anorexia. In this final part, I wrap up with some reflections on dissonance and optimization in the context of decision-making about recovery and the rest of life.

In the end, for me what cognitive dissonance theory most usefully teaches us is that there are defaults at play into which we may have little insight and yet which may shape a lot about what we do and don’t do, and that dissonance reduction is the structure quite a lot of them take. In the previous post, I suggesting that dissonance reduction is optimizing for tolerable, because to the extent that it’s a reduction and not elimination, it’s settling for an in-between.

This suggests a useful question to ask in the effort to understand any human behaviour (our own or someone else’s): “What is this optimizing for?” There are all kinds of optimization going on here and now, in the system that is you, in parallel (and at loggerheads) and at multiple levels of generality. Cognitive dissonance is part of some of them, and often where it exists it gets entangled with other cognitive biases, like the sunk cost fallacy (I’ve invested so many years in this relationship I can’t let it go now; I spent such a lot on this home gym equipment I couldn’t possibly get rid of it now). These biases may cluster together to shape our operational definitions of waste and loss, and guide insights and actions in one direction or another, not unrelated to what “your” explicit values and aspirations are, but with no necessary primacy for those.

To intervene in this complex system, you can put a probe into it: You can ask the optimization question of yourself, to understand what’s happening now and also to shape what you do next. If you ask, “What is this optimizing for?” where this is the current set of behaviours and attitudes you’re practising and manifesting (in thoughts, words, actions), doing so is in part a way of attuning yourself to dissonances in play. 

The “this” is importantly depersonalized here: You can look at your behaviours and attitudes as interesting little pieces of a big and complex system none of which is necessarily optimized for anything that “you” (think you) care about. The “behaviour” part is also relevant: Even if it’s the cognitive-emotional stuff that feels the most fraught (whether in the context of recovery, or something else, say a breakup), attending to the surrounding actions (whether exercise routines or phone use or anything else) can often be a helpful way to begin to generate clarity. And the “this” can refer to a single behaviour (What is my daily walk to the park optimizing for? What is this thing of keeping my phone switched on overnight optimizing for?). Or, zoomed out, it can also encompass the entire collection of things you’re doing at the moment: What is this life setup optimizing for? Finally, it’s a bit harder, but you can also try it with physical states (What is this bodyweight optimizing for?) or cognitive patterns (What is this protracted anticipation of my walks optimizing for? Or what is this posthoc dissection of all our conversations optimizing for?).

The answers we end up giving may often merit further interrogation (OK, leaving my phone on at night is optimizing for that warm-yet-empty feeling of knowing he might text me but probably won’t, and that is, in turn, optimizing for keeping this breakup uppermost in my mind night and day, which is optimizing for resisting the breakup and the change; or, the daily walk is optimizing for anxiety relief, but that is optimizing for sustaining the system in which I get anxiety relief only from walking, and increase anxiety by not walking, hence walking is optimizing for anxiety maintenance). You’ll have to decide where to stop, but continuing to push beyond where it feels comfortable can be revealing, since many of our most powerful objects of optimization (beauty, wealth, slimness, etc.) have optimization benefits that are themselves unclear or dubious (so if I were slimmer I’d get more dates with guys who wouldn’t have swiped right if I’d been plumper?). This is often because dissonance reduction is at work: e.g., if I’ve been aspiring to this all my life it must have value, so I’ll keep aspiring to it.

Our answers may often include an acknowledgment that, OK, yes, currently in this area the things I’m doing or thinking are optimizing for convenience or ease, say, rather than something you might like to think you value more. Or maybe sometimes the opposite: the status signaling inherent in conspicuous un-ease (look at me, I’m balancing the unbalanceable). The what-I-say-versus-what-I-do contrast is often revealing here, and the gap can be terribly narrow yet make all the difference (the size of the leavings on a plate, maybe, or the few extra paces that didn’t really need taking). 

If we then go on to ask “what do I want this to optimize for?”, we're creating an opportunity to change the default. This is a more flexible way of asking “how can I make my life better?” or “what would make me happier?” because better is so vague and happiness might not quite capture your thing, and “this” can be at any level of specificity you like. Asking a question like this is one way to counter the myopia of eating disorders: the habit they have of making you optimize, endlessly, for only a few very boring things. Asking yourself what all/some of this is optimizing for, and what it could, is thus a nicely focused way of zooming out and seeing both what is and what could be.

This version of the question is interesting to ask at intervals. You get to observe the changes in the answer: from a life phase in which you want to optimize for expansion to one in which you want to optimize for calm, for endings, or for new beginnings. And then you get to work out what actions might allow you to do that, and test something out, and repeat, and learn.

Variants on freedom

 ABC-Clio. Reproduced with kind permission of M. Shermer.
"Imagine a world full of brains, and far more memes than can possibly find homes."
Source: Illustration by Pat Linse. Originally published in The Skeptic (US), 1997, 5(2), 43-49. Reprinted in 2002 in S. J. Blackmore, “Memes as good science”. In M. Shermer (Ed.), The Skeptic Encyclopedia of Pseudoscience (652-663). Santa Barbara, CA: ABC-Clio. Reproduced with kind permission of M. Shermer.

Defaults are powerful things, and they often involve dissonance reduction. Like the shortcuts our bodies so eagerly take when sitting or standing or walking or lifting, the defaults aren’t optimized for anything more than the system’s current path of least resistance. Like the genes and memes that are optimized for selfish replication regardless of the benefits or drawbacks to us lumbering organisms, their hosts, the default desire for dissonance reduction is just a self-replicating mechanism: When you satisfy it, you feel better (for a bit), so you do it again. Positive feedback kicks in, and the renewed dopamine hit or handy distraction from difficulty just helped that little habit loop become a little bit more habitual. 

Identifying that this is what’s going on may just help you select something that works better in the medium and long term, and possibly even the short term. After all, trying to write this difficult sentence without flicking to email as soon as I don’t know how to carry on can feel immediately better as well as helping me finish this post sooner, align my working practices with my image of an ideal working day and with my professional ambitions, etc. And in the same way, eating this whole cookie now can in fact give me energy and mental clarity for right now this afternoon, as well as aligning my choice and my day with the reasons why I want to recover—even if I know that in between there’ll be some of the predictable dissonance where the pro-recovery action meets the still-habitual anxiety and guilt. You try something, and it changes something, and now you know something you didn’t before, and now a slightly shifted reality is possible for you.

In the end, reduction is unwinnable. Only once the burden of partially resolved dissonance is lifted do you get the chance to find out how much easier life is when it’s gone, when there’s resolution in place of reduction, avoidance, or concealment.

Of course, there’s no absolute freedom in taking action against dissonance or any other evolved cognitive mechanism. If you decide to give it a go (ask the question, give an answer, act on the answer) because of reading this, it’ll just be because this little set of Psychology Today optimization memes won out for a bit against some other intersecting and competing threads of cultural/cognitive/biological evolution. The processes kept playing themselves out. But in a possibly happier direction.

And of course, the rest of your life won’t be a dissonance-free nirvana, but at least you’ll have given yourself a robust piece of evidence that unless behaviour changes, nothing much ever does. 


Blackmore, S. J. (1999). The meme machine. Oxford: Oxford University Press. Google Books preview here.