What Do We Know About Making Better Food Choices?

Make a healthy diet less onerous with these few easy tweaks

Photo by Lucian Alexe on Unsplash

---

I'd like you to picture two scenarios.

Scenario 1: It's Sunday and you're planning your week. Nutrition is important, so you imagine eating kale and broccoli and stuff like that. You see yourself feeling energized, strong, and your body is a temple.

Scenario 2: It’s Monday and you just finished a difficult assignment, 5 minutes before the deadline. Your brain feels like mashed potatoes. What was it again that you wanted to eat? Crisps and a burger? Sounds vaguely familiar... let's get that! Sometime in the middle of the bag of crisps, you realize that this isn't kale, nor broccoli. Fuck.

If you haven't experienced this, good for you. For the other mortals out there, I've two questions that I wanted to address with this piece:

Why do we eat food we know is bad for us and how can we make better¹ food choices?

Mental Organizer

I'll start with reasons why food consumption isn't rational, which will involve dipping our toes in individual, social, and structural factors that govern our food choices. The common thread running through the discussion is the misalignment between our ideal and our actual selves I mentioned above. I will then go on to discuss what we know about improving people’s food choices, and how you can apply this knowledge to construct self-nudges: structuring your environment so that the hypothetical choices of your ideal Sunday self are more in line with the cognitively flushed-out Monday (or you know, any other weekday) self that actually makes the choice.

Factors Determining Food Choice

Values

What do you value? Most people value taste, price, convenience, novelty, and health². We like what tastes good, is cheap and easy to access, has some variation in it, and - not so importantly - is healthy. Values act like guidelines for our choices. The fun thing with values is that you might value kale when I ask you (explicit value), but would actually never consider eating it (implicit values).

So, one possible way the misalignment between the selves is through value mismatch.

Habits

Habits often form when we perform repetitive behaviors in stable contexts, and achieve some sort of reward by doing so³. Over time, the reward becomes less and less important, and the context - especially the cues in it - takes over. Thus, we often eat at the same time (breakfast, lunch, dinner), in the same places (kitchen usually triggers food considerations even when we're not hungry), or during certain mental states (think on Bridget Jones gorging on ice cream). Habits were even found to be more predictive of food choices than intentions and sensitivity to food temptations⁴. From this, we can infer that most of our behavior regarding food (what we buy, order, eat, or even think of), is habitual.

The misalignment between the selves might, in this instance, occur through the discrepancy between what you intend to do and what your habits are. The psychologists being psychologists of course have a truly endearing term for that: behavior-intention gap⁵.

Emotion

The Bridget Jones example also touches upon emotions as a motivator for our behavior. Another example could make this even clearer: think about the craving you experience when you're hungry and go shopping. The basket quickly balloons with stuff that you'd probably not buy had you gone shopping sated. But affect can also be subtler than that: you can have a slight inkling toward certain food or aversion to another. This might not even register in your conscious mind. After all, who goes around questioning why they like something or not? We just do.

Thus, another way our selves might end up misaligned is through emotion: our cool Sunday selves envisage we'll be a paragon of virtue because thoughts are cheap. But when your Monday self is standing in front of kale or burger, the likelihood of choosing the former is quite low.

Heuristics

Let us stay in the supermarket. Next up: heuristics. How do they come into play when we make our decisions? A general idea is that we like to save ourselves the mental effort of going through labored and thorough decisions. Instead, we take (often warranted) shortcuts:

• Buy the brand you know.

• Always put some bananas in the basket.

• Get the cheaper option.

And so on.

While most of the time the heuristics are helpful as they save us the mental effort and consistently deliver on our implicit goals and motivations, they can also become stale and not relevant.

Yet another way our selves might end up out of whack is through using rules of thumb that aren't working for us anymore.

Social influence

Social influence is not a negligible part of our decision-making as well. Take, for example, social norms: What are most of the people eating in your environment? What do they think is normal? All these impact your decision on some level by making certain foods normal and prevalent in the environment.

Let's think on the scenario from the introduction again: you stand in front of the kale dish and the burger, only this time you see that one colleague after another chooses the burger. Well, of course, you won't go for kale now. You don't want to be the weird hipster dude that stands out. On the table, the conversation also revolves around meat and how your colleagues like it, especially Todd. Todd boasts he eats meat every day because it's healthy and awesome and manly. Nobody likes Todd.

Social influence can misalign the selves too. We are social creatures and conformity and ostracism are influential forces to us.

Physical environment

Lastly, let's look at the physical environment. In the supermarket, most of the stuff that they want you to buy is on the eye level, or at the end of the aisle. The opposite is the case for items that have lesser margins, for example. These are placed either a bit too high or at your feet. Price tags that shave off a few percent are also enticing.

The environments food decisions take place aren't sterile, as much as your Sunday self would like to think so. They are filled with nudges and smells and sights and predetermined choices and flashy displays. Which is how you'll find yourself reaching for the burger and not the kale. Nobody got ever enticed by the smell of kale (is there even any? And yes, I do eat kale).

In short, there are several factors that play a role in shaping our diets:

• individual (values, beliefs, habits, emotions, heuristics);

• social (social norms, identity, peer influence);

• and material (physical environment and choice architecture, pricing, media, and marketing).

So our food choices aren't rational. But is that an issue? After all, if you can get what you want and are happy with it, there's no problem, right?

If it were that easy, you wouldn't experience so many mismatched occurrences between your ideal Sunday self and your Monday, burger-loving self. And it turns out you're not alone, people want to eat healthier, but don’t⁶. They also want to eat more sustainably, but don’t⁷.

Increasingly, people also come to recognize the environmental and moral costs of some of the food that they consume and want to change⁸. Yet change is hard, precisely because simply knowing that we ought to do something doesn't make it so. There are barriers.

So, the mid-takeaway is this: many people want to eat differently (in this case - healthier, more sustainably, and ethically) but often, they don't. This is what the rest of the piece is going to be about: how to bridge the gap between what we want and how we actually behave. How? Knowing what we know about the factors that influence our food decisions, we can reverse engineer some solutions.

Effectiveness of various food choice interventions

But first, let us establish the effectiveness of what I'll discuss in the recommendations. Next, I'll outline the findings that have emerged in the past few years, and discuss their pros and cons. You can skip this part if you’re not interested in the data behind the recommendations.

Take a meta-analysis of Cadario and Chandon (2019) that looked at the effects of nudging in affecting people's food choices⁹. The overall effect was modest, Cohen's d = 0.23. This means that the difference between the no nudge and nudge scenario differs by 0.23 standard deviations. Since we're talking about food, this meant approx. −124 kcal/day.

But the possibility doesn't end there. They also investigated 3 types of nudging interventions. Either they focused on cognition (generally, how people think, perceive, and pay attention to food), affect (how people feel about certain food), or behavior (how people actually act, despite their thoughts or emotions). They found that the effects of the latter - behavior-based interventions - could go as high as d = .39, which amounts to −209 kcal per day. Why would the behavior-based interventions find the largest effects? One of the reasons might be that they prompt the behavior itself, not a precursor - such as cognition or affect.

The results are broadly in line with another meta-analysis by Broers and colleagues (2017), which found effect sizes ranging from d = 0.28 to d = .39¹⁰.

From that, we might surmise that nudging food choices seem to be moderately successful.

What might be the downside of the findings? For one, we don't know the scope of the unpublished studies without significant results. Studies that find positive results tend to be published more. I can imagine that in a field that overhyped such as nudging, there might be a couple of studies lying around in a file drawer, contributing to a publication bias (e.g. here). Publication bias likely inflates the actual effect sizes found in the meta-analyses, so the actual effect is likely smaller.

Second, most of the included studies are field studies, where it is exceedingly difficult to control for confounders. Third, while most of the studies last for weeks or months, we still can't say how the interventions will age. It might be that the effect would disappear had the study been going for longer.

Some authors also argue that if a nudge is removed, its effect also disappears¹¹. This makes sense since most of the nudges aren't geared toward capacity building - knowing why you do something - but rather toward direct behavior change without necessarily understanding the reason why.

With those caveats in mind, shaving off a few points in the effect sizes would probably be the most prudent way to interpret the results.

All in all, I'd still say we are dealing in modest effect sizes, and that - considering the low cost of implementing the interventions - nudging food choices is still worthwhile.

How to structure your environment for the best self-alignment

But let's now turn to some concrete examples you can realign the selves. I'll order these according to 3 contexts in which food choices are likely to take place: Supermarket, cafeteria/restaurant, home. Of course, some of the strategies fit in more than one category, so if you see that something from the supermarket could fit you at home, go ahead!

Cafeteria/restaurant

Precommitment

Most of us have access to, and regularly visit, some sort of cafeteria. The food there tends to be on the cheaper side (compared to a restaurant), and the taste varies. But, what we lose on the taste we gain in convenience.

Most of the cafeterias offer some sort of program where you can preorder food. Why would you do this? By now, I hope you understand the difference between your ideal self and your actual self. So, ideal selves are usually dealing in ideal conditions: not hungry and not nudged this way or that by the environment or colleagues or whatever. Sadly, those selves usually don't buy our food. Our actual selves (the selves that do buy the food) are hungry and impulsive and cognitively spent: they are not in the best frame of mind to make long-term choices.

Preordering food can tilt the scales¹².

What if there isn't any program for preordering the food? You can create a commitment device for yourself. Look at the weekly menu at home and make choices as to what you'd like to eat. Write down your choices or jot them on your phone. If you still don't trust yourself, text your choices to your colleagues for extra encouragement (and who knows, maybe they also want to eat better, so you might end up in a virtuous cycle).

An alternative way of creating some sort of precommitment is to form implementation intentions around foods you want to eat or avoid. Implementation intentions take the form of if-then plans that consist of two elements:

1. a cue that triggers some (unwanted) behavior and (e.g. seeing a juicy burger on the counter, makes you want to order it);

2. a pre-planned response to the cue (e.g. ordering an alternative instead)

The complete implementation intention could thus look like this:

“If I see a juicy burger being prepared, then I will not order it and choose a healthy alternative instead.”

The more specific the implementation is, the better. Now I know how this sounds - ridiculous. Why would anyone do something just because of this tiny little mental litany? Well, it turns out that it truly works. A comprehensive meta-analysis by Adriaanse and colleagues¹³:

Implementation intentions are an effective tool for promoting the inclusion of healthy food items in one's diet (d = .51), but results for diminishing unhealthy eating patterns are less strong (d = .29).

Don't eat the first peanut

Another strategy that you can employ should you find yourself in restaurant settings or on a night out, is to never eat the first peanut¹⁴. What do I mean by that? Let's say you arrive at your friend's place and the food is being prepared. So your friend, being a thoughtful host, puts some snacks in front of you. You'll say: well, I'm not here for the peanuts, but a handful won't harm me. A few minutes later, the bowl is gone and you're full, heavy, and disgusted by yourself. The issue? You ate the first peanut, and the avalanche is now rolling down the hill with no sign of stopping. Sigh.

So the rule is, never to eat the first peanut. It's easier not to engage in the behavior at all than to control yourself once you start.

Supermarket

Defaulting is one of the poster children for nudging. Simply put, it means exactly what it says: instead of actively choosing, you default to a predetermined option or set of options. These options are usually prominently displayed and recommended. Think of annoying cookie settings when visiting a webpage. The default option - to get tracked the shit out of you - will be the one that is easiest to click on. It will have a vibrant, come-hither color, take up more space, and in some instances feel like you cannot even choose anything else.

You can use the same principle when you go grocery shopping: set default options for the foods you buy. Psychologists have a fancy name for it: stimulus control¹⁵. If you have something that you know reliably triggers a behavior, it stands the reason to remove the temptation altogether. If you know you can't be trusted with a bag of gummi bears, don't buy it in the first place.

The same can be applied for the foods that you want to eat more of: have a set of default items that you always buy that are preferably healthy, ethical, and environmentally friendly (or whatever values you aim for).

Defaults work because they play on our innate cognitive frugality (read: laziness). Making choices is difficult. Ergo, defaults reduce the friction and make a certain behavior more likely¹⁶.

Home

Introduce friction

So defaults didn't work and you find yourself with a bag of chips in the cupboard. That's fine. I also often find surprises in my basket after I've paid.

The next step in the arsenal is to induce friction and make the bag of chips somehow less accessible.

A personal anecdote: my girlfriend is in sole possession of my Steam password, so whenever I want to play, I have to ask her. I also instructed her to question my reasons and give in only when she feels I've earned it. It's not a perfect system, but it works. Another example is locking apps that you overuse. Messenger apps and crypto apps are my scourge, so I locked them with a (long) password. While not perfect, the little friction has saved me a few habitual - and unnecessary - check-ups already.

In a similar way, you can put the chips somewhere where you don't see them and preferably, somewhere out of reach. Installing a lock on the cupboard doesn't sound too farfetched either. The more difficult the stimulus is to reach, the better.

Resizing

The most effective nudging intervention, according to the meta-analysis I mentioned above, is size enhancements or decrements¹⁷. Since there have been many studies done on this subject already, we’ve quite a good understanding of the conditions that make resizing more or less successful in guiding our food choices. According to Holden and colleagues (2016)¹⁸, self-serving the food on a smaller plate reduces consumption quite a bit (d = 0.70). It also seems that the portion needs to be proportionally decreased to the plate size: if you put the same portion on a smaller plate, there are no differences between smaller and bigger plates (d = 0.03). Lastly, awareness plays a role. If you are aware that I’m manipulating the plate size, the effects are only half as large as if you’re unaware (d = 0.76 → d = 0.31).

So, the implications seem to be: a) either mindfully decrease the portion and serve it yourself or; b) let someone else do it without you knowing.

Prevalence

Prevalence simply means accounting for how many instances of a certain option (let's stick with chips as an example) there are in the available set of options. This means, for example, going from 1 out of 3 to 2 out of 3 options for a certain type of dish in a menu. So, if you’re trying to make yourself eat better, improving the ratio of good to bad food is something to consider.

You can, of course, apply this at home as well. By making certain (healthy) foods or dishes prevalent in your home environment, you'll be more likely to eat them. Here, I could also mention the mere-exposure effect¹⁹, where the prevalence of something in your day-to-day life makes you develop a positive affinity to it. Meaning: keep the broccoli around long enough and you'll likely come to like it. This of course ties with the supermarket strategies I mentioned above.

General applicability

Habit discontinuity

I broached the concept of habits several times before. In the section about setting your own defaults, I discussed how shopping habits have likely already formed. Thus, the path that you take in the supermarket is often the same. The products you pick up tend to be also. Similarly, at home, you've developed cooking routines of how to combine the items you buy into dishes you cook. You've also probably developed snacking routines and routines I haven't even thought of.

These routines - or habitual behaviors if you will - are bound to a specific context. Ergo, it stands the reason that when the context changes, the associated routines are also disrupted²⁰.

When is habit discontinuity relevant?

Moving

It probably doesn't happen often, but you are likely to move during certain periods of your life. Moving to your first flat, moving in together with your partner, for work, internship, etc.

Each time you move, you're surrounded by a new environment. This is a tremendous opportunity to ditch the (bad) habits you don't want to do anymore and pick up new ones. In food choice, this could mean:

• setting up new and better defaults in the supermarket and/or;

• looking up new recipes and learning new dishes.

Of course, all these are possible without changing a location. The idea behind the habit discontinuity is that the change is easier when you move because you are no longer exposed to the old cues that trigger unwanted behaviors,

Go to a different supermarket

This might not always be practical, but you can choose a different supermarket to do your grocery shopping in. This creates the situation where you are surrounded by new cues and can presumably build new shopping habits. For this to work, it's likely best if you create the shopping list as you go. You probably know instinctively which staples you buy anyway, but it's important not to get bogged down by checklists at this stage.

Takeaway

To sum up, we suck at making rational food choices. There's a host of individual, societal, and environmental factors that influence us, causing misalignment between the ideal and the actual selves. This wouldn't be a problem had we been happy with the choices we make, which often isn't the case: we want to eat healthy, ethical, and environmentally-minded diets, but we don't. To bridge the gap between what we want and what we actually do, I proposed applying self-nudging interventions, such as precommitment, friction, and habit discontinuity, which have been found to be modestly effective in improving food choices.

Deciding what to eat doesn’t have to be an onerous exercise. You can mitigate the inherent pull toward yum but unhealthy foods through decent environmental design, a few firm rules, and a bit of patience. The rest is up to you!

1. defined as choices that your ideal self would like to take.

2. Why Americans Eat What They Do: Taste, Nutrition, Cost, Convenience, and Weight Control Concerns as Influences on Food Consumption - ScienceDirect

3. Psychology of Habit | Annual Review of Psychology (annualreviews.org)

4. The power of habits: Unhealthy snacking behaviour is primarily predicted by habit strength - Verhoeven - 2012 - British Journal of Health Psychology - Wiley Online Library

5. The Intention–Behavior Gap - Sheeran - 2016 - Social and Personality Psychology Compass - Wiley Online Library

6. 75 Percent of Americans Say They Eat Healthy — Despite Evidence To The Contrary : The Salt : NPR

7. The Elusive Green Consumer (hbr.org)

8. Consumer Attitudes Towards Environmental Concerns of Meat Consumption: A Systematic Review (nih.gov)

9. Which Healthy Eating Nudges Work Best? A Meta-Analysis of Field Experiments | Marketing Science (informs.org)

10. A systematic review and meta-analysis of the effectiveness of nudging to increase fruit and vegetable choice - PubMed (nih.gov)

11. Nudging and Boosting: Steering or Empowering Good Decisions - Ralph Hertwig, Till Grüne-Yanoff, 2017 (sagepub.com)

12. Advance Ordering for Healthier Eating? Field Experiments on the Relationship between Time Delay and Meal Content | Request PDF (researchgate.net)

13. Do implementation intentions help to eat a healthy diet? A systematic review and meta-analysis of the empirical evidence - ScienceDirect

14. Self-nudging and the citizen choice architect | Behavioural Public Policy | Cambridge Core

15. Frontiers | In Support of a Distinction between Voluntary and Stimulus-Driven Control: A Review of the Literature on Proportion Congruent Effects | Psychology (frontiersin.org)

16. This touches on habits as well. Remember, if you repeatedly perform a certain behavior in a stable context, chances are you'll build habits over time. In fact, it's likely you've already built some shopping habits. If they don't serve you anymore, read on, I'll discuss a strategy that might help.

17. see #8

18. Whether Smaller Plates Reduce Consumption Depends on Who’s Serving and Who’s Looking: A Meta-Analysis | Journal of the Association for Consumer Research: Vol 1, No 1 (uchicago.edu)

19. Mere Exposure Effect - Biases & Heuristics | The Decision Lab

20. Context change and travel mode choice: Combining the habit discontinuity and self-activation hypotheses - ScienceDirect