The Person Was Never the Only Variable

Throughout this series, we have looked at several characteristics of food that can influence eating behavior. Foods combining fat and carbohydrate can carry greater reward value than would be predicted from either nutrient alone. Physical structure and texture influence how quickly a food can be eaten. Energy density determines how much energy can be contained within a given amount of food, and portion size affects how much people consume even when the food itself remains unchanged.

Researchers separate these variables because doing so makes them easier to study. Everyday eating does not separate them at all. A food can combine fat and refined carbohydrate while also containing substantial sodium. It can be energy dense, require very little oral processing and be served in a large package. It can be inexpensive, portable, shelf-stable and available almost everywhere. Advertising can repeatedly remind us that it exists, and a delivery service can reduce the distance between wanting it and obtaining it to a few taps on a phone.

None of these characteristics determines what any individual person will eat. Together, however, they create the conditions in which eating takes place. That distinction matters because our explanations for eating behavior have traditionally paid far more attention to the characteristics of the eater than to the characteristics of those conditions.

When someone eats more than they intended, we ask why they did not stop. When they repeatedly choose highly rewarding foods, we question their self-control. When attempts to change their eating are difficult to sustain, we assume motivation is the missing ingredient. The underlying model is remarkably consistent: the environment is treated as background, while the person becomes the explanation.

The research examined throughout this series makes that model increasingly difficult to defend. The person matters, but the person was never the only variable.

What happened when researchers changed the food environment?

Kevin Hall and colleagues provided an unusually clear demonstration in their 2019 inpatient randomized controlled trial at the National Institutes of Health. Twenty adults lived at the NIH Clinical Center for four weeks, allowing researchers to control the food available to them and accurately measure what they consumed. Participants spent two weeks receiving an ultra-processed diet and two weeks receiving an unprocessed diet, with the order randomly assigned. They were allowed to eat as much or as little of the provided food as they wanted.

The diets were designed so that the meals presented were matched for several nutritional characteristics, including calories, macronutrients, sugar, sodium and fiber. Yet participants consumed approximately 500 more calories per day during the ultra-processed condition. They gained weight during those two weeks and lost weight during the unprocessed condition.

The strength of this experiment is not that it proved ultra-processed food is uniquely harmful. It did not identify a single mechanism responsible for the difference, and ultra-processed describes a broad category containing foods with widely different nutritional and physical properties. Its importance lies somewhere else: researchers changed the food environment while studying the same individuals, and their spontaneous food intake changed substantially.

The participants did not need to become different kinds of people for the outcome to change. Their motivation, personality and moral character were not experimentally manipulated. The food was. That observation should be central to any serious discussion of eating behavior.

The category is less interesting than the mechanisms

Hall’s study created an important scientific problem. If participants ate substantially more during the ultra-processed condition, what characteristics of that condition produced the difference?

One clue was eating rate. Participants consumed the ultra-processed meals more quickly. As we saw earlier in this series, food structure affects oral processing, and oral processing affects how rapidly food can be consumed. Subsequent experimental work has strengthened the evidence that manipulating texture and eating rate can alter spontaneous energy intake.

Energy density provides another piece of the explanation. Foods that contain more calories per gram allow more energy to be consumed from the same physical amount of food. When high energy density is combined with rapid eating, the rate of energy consumption can increase dramatically.

Hyper-palatable nutrient combinations may contribute as well. Research by Alexandra DiFeliceantonio and colleagues suggests that foods combining substantial fat and carbohydrate can carry greater reward value than foods dominated by either nutrient alone. Tera Fazzino and colleagues have separately identified combinations of fat and sodium, fat and simple sugars, and carbohydrate and sodium that occur frequently in foods classified as hyper-palatable.

Portion size adds another layer. A large amount of a highly rewarding, energy-dense, rapidly consumed food presents a different eating environment from a smaller amount of a food requiring prolonged oral processing. Experimental research shows that increasing portions can increase consumption even without changing hunger or the food itself.

None of these mechanisms alone provides a complete explanation for the NIH findings or for overeating more generally. That is precisely the point. Eating behavior is produced by interacting systems, and a broad label such as ultra-processed can obscure as much as it reveals if we stop investigating once we have assigned the category.

The more useful scientific project is to determine which characteristics matter, under what circumstances they matter, and how they interact with the characteristics and experiences of the person eating.

Real foods contain several variables at once

Consider the transformation of a potato into a potato chip. The starting ingredient is recognizable, but the finished food creates a very different eating experience. Water has been removed, fat and sodium have been added, energy density has increased, and the physical structure has been changed into something that fractures easily and can be eaten quickly. The finished product is commonly sold in packages containing considerably more than a person might consume if each portion had to be individually prepared.

There is no single moment in that transformation when the potato becomes a dangerous food. Nor does the example establish that eating potato chips necessarily leads to overeating. What it illustrates is how multiple appetitive properties can change simultaneously.

The same principle applies to many foods. Processing can increase convenience while changing texture. Added fat can increase both palatability and energy density. Refinement can reduce the physical work of eating. Packaging can increase the quantity immediately available. Flavor formulation can increase sensory appeal. Each characteristic has the potential to interact with the others.

This is where the idea of supernormal appetitive properties becomes useful, provided we use it carefully. Nutrition science does not have an established category called “supernormal foods,” and the concept should not be used to suggest that certain foods somehow bypass human agency or overwhelm the brain in the manner of a pharmacological drug. It can, however, describe a broader environmental phenomenon: modern technology allows us to concentrate, combine and optimize appetitive properties beyond the configurations commonly encountered in minimally processed foods.

The human attraction to those properties did not originate in a factory. The capacity to find energy, sweetness, fat, sodium and familiar sensory experiences rewarding is part of normal human biology. Technology simply gives us unprecedented ability to manipulate the intensity, combination, accessibility and delivery of those signals.

Availability changes behavior before eating begins

The physical characteristics of food are only part of the environment. A food cannot influence eating if it is unavailable, and modern food systems have dramatically reduced the effort required to obtain food.

Behavioral scientists sometimes describe the effort standing between a person and an action as friction. Even small amounts of friction can influence behavior because every additional step creates another opportunity for an action to be delayed, reconsidered or abandoned.

Food once required relatively substantial friction. Obtaining it could require growing, gathering, hunting, shopping, preparing, cooking and cleaning. Many of those demands still exist, particularly for people trying to prepare meals from basic ingredients. At the same time, other parts of the food environment have removed them almost completely. Ready-to-eat food can be purchased at gas stations, pharmacies, office kitchens, vending machines, grocery checkouts, drive-throughs and countless other locations. Delivery platforms can move prepared food from a restaurant to a person’s home with almost no physical effort from the consumer.

Convenience is not inherently harmful. For someone working long hours, living with a disability, caring for children, lacking cooking facilities or simply trying to feed themselves during a difficult week, convenience foods can make adequate nutrition possible. The point is not that friction is good. It is that friction is a behavioral variable.

Reducing the effort required to perform a behavior tends to make that behavior easier. Increasing the effort tends to make it harder. Food choices are not exempt from this basic feature of human behavior.

This also gives environmental thinking a constructive role. Preparing enough dinner for leftovers reduces the friction involved in eating the following day. Keeping satisfying snacks available reduces the effort required to respond to hunger before it becomes extreme. Washing or preparing foods in advance changes the number of steps between wanting them and eating them. Environmental design works in both directions.

The commercial environment has reasons to remove friction

Food availability is not random. Most of the food environment exists within markets, and markets create incentives. A company benefits when consumers choose its product over another product and return to purchase it again. That creates incentives to make food appealing, recognizable, affordable, convenient and easy to obtain. Product developers test formulations and textures. Companies conduct consumer research. Restaurants refine menus. Retailers study placement and purchasing behavior. Advertising increases product recognition and salience. Delivery systems reduce the effort between deciding to eat something and receiving it.

None of this requires a conspiracy, and understanding it does not require assigning malicious intent to everyone working in the food industry. The incentives are sufficient. A product that people strongly prefer has commercial value, and reducing barriers to purchasing that product can increase its value further.

The important consequence is that our food environment is not a neutral collection of everything humans could possibly eat. It has been shaped partly by commercial selection. Products that consumers enjoy and repeatedly purchase are more likely to survive, expand and be imitated. Products that fail to attract buyers tend to disappear.

When the characteristics that make a food commercially successful overlap with the characteristics human appetitive systems find rewarding, biological and economic incentives can reinforce one another. That does not make consumers helpless. It does mean that choice occurs inside a deliberately constructed choice environment.

Appetite also arrives with a history

Even this expanded account remains incomplete if it treats everyone as responding identically. The same food can produce very different experiences in different people and in the same person on different days.

Someone who has not eaten adequately for twelve hours encounters a food differently from someone who has recently eaten a satisfying meal. Chronic restriction can increase food preoccupation and alter responses to foods perceived as forbidden. Sleep deprivation can influence appetite and food choice. Stress can change eating behavior. Medications can alter hunger and satiation. Food insecurity can affect how someone responds when food becomes available. Cultural experience shapes which foods are familiar, meaningful, comforting or desirable.

Time and money influence the environment as well. A person working two jobs does not face the same food-preparation environment as someone with several free hours each evening. Someone without reliable transportation does not have the same practical food options as someone who can easily reach several grocery stores. A parent feeding children is making decisions within a different set of constraints from someone preparing food only for themselves.

These differences matter because environmental explanations can become just as simplistic as individual ones. Saying that food properties influence intake does not mean everyone exposed to the same food will respond in the same way. Behavior emerges from the interaction. The relevant unit is neither the person alone nor the environment alone. It is the person in the environment.

Why willpower explains so little

Willpower feels like an explanation because it assigns a name to the difference between intending to do something and actually doing it. The problem is that it often tells us very little about what produced that difference.

Suppose someone intends to eat regularly during the workday but repeatedly reaches late afternoon without having eaten. Saying that they lack discipline does not tell us whether meetings prevented breaks, food was unavailable, preparation required too much time, morning appetite was low, workplace norms discouraged eating or the person had learned to ignore hunger through years of dieting.

Likewise, describing someone who eats more than intended as lacking self-control does not tell us whether they began the meal extremely hungry, were eating a rapidly consumed and energy-dense food, were distracted, had been restricting that food for weeks, were served a very large portion or simply found the food extraordinarily enjoyable.

A useful explanation should create possibilities for intervention. “You need more willpower” usually creates only one: exert more effort. Understanding the variables contributing to behavior creates many. Someone can eat earlier so every evening meal does not begin with extreme hunger. They can make satisfying foods easier to access. They can examine whether restriction is intensifying preoccupation with particular foods. They can notice how different meal structures affect satisfaction. They can change portions, packaging or where food is stored if those variables matter to them. They can choose convenience deliberately when convenience makes adequate eating easier rather than treating it as nutritional failure.

None of these strategies guarantees a particular outcome. They simply recognize that changing behavior can involve changing the conditions producing it rather than endlessly trying to strengthen resistance to those conditions.

Environmental understanding can increase agency

There is a common assumption that acknowledging environmental influence diminishes personal agency. That conclusion follows only if agency is defined as being unaffected by context. Human beings have never had that kind of agency.

We make decisions using bodies with physiological needs, brains shaped by previous experience and environments containing opportunities and constraints. Recognizing those influences does not eliminate decision-making. It gives us a more accurate map of the terrain in which decisions occur.

In practical terms, environmental understanding can increase the number of available choices. If someone believes the only problem is that they cannot control themselves around a particular food, the solution is perpetual resistance. If they understand that their response changes depending on hunger, availability, restriction, portion presentation, eating rate and context, they can experiment with any of those variables.

Agency becomes less about proving that the environment cannot affect you and more about learning how to construct environments in which the behaviors you value require less continuous effort.

Understanding appetitive properties should not become another reason to fear food

There is an obvious danger in this discussion. Nutrition culture has a long history of taking interesting scientific findings and converting them into lists of foods people should fear.

Research showing that certain nutrient combinations have high reward value can become a warning never to combine fat and carbohydrate. Research on energy density can become a mandate to fill the stomach with the fewest possible calories. Research on eating rate can become another set of rules about chewing. Research on portion size can become pressure to serve increasingly small amounts. The concept of ultra-processing can become an enormous new category of forbidden foods.

That would reproduce the very problem this series is trying to address.

Human beings are supposed to find food rewarding. Pleasure is part of eating. Convenience can be valuable. Energy-dense foods can be useful. Birthday cake does not become a public-health emergency because it combines fat and carbohydrate, and potato chips do not become evidence of failed appetite regulation because they are easy to eat.

The purpose of understanding appetitive properties is not to remove pleasure from eating. It is to remove some of the moral judgment from our explanations of eating behavior. Knowledge becomes useful when it increases curiosity and flexibility. It becomes another diet when it produces fear and rigid rules.

From self-judgment to investigation

Perhaps the most important consequence of this research is explanatory. When an eating experience does not go as someone intended, the default question is often, “What is wrong with me?” That question immediately narrows the investigation. The presumed cause is already located inside the person, and whatever happened becomes evidence about their character, motivation or capacity for control.

A more useful question is, “What contributed to what happened?”

That question allows the person to remain part of the explanation without requiring them to be the entire explanation. Hunger can be examined alongside food structure. Satisfaction can be considered alongside portion size. Personal history can be considered alongside availability. Intentions can be considered alongside fatigue, stress, time, money and convenience.

Sometimes the answer may be remarkably simple: the food was delicious and the person chose to keep eating it. Pleasure is also a legitimate explanation for human behavior. Other times, a pattern becomes visible. Perhaps episodes of feeling out of control consistently occur after long periods without adequate food. Perhaps a food becomes far more compelling after it has been forbidden. Perhaps certain meals are consumed so quickly that satisfaction is difficult to notice. Perhaps a work environment repeatedly prevents eating until hunger becomes urgent.

Those patterns give us something self-criticism rarely does: information that can be used.

Your appetite is doing its job in the world we built

The concept that began this series was the supernormal stimulus: the observation from ethology that biological systems can sometimes respond especially strongly to exaggerated versions of signals they evolved to notice. Human nutrition research does not provide evidence for simply declaring modern foods “supernormal stimuli,” and doing so would flatten a complicated scientific literature into a catchy label.

What the evidence does show is more interesting. The reward value of food can change when nutrients are combined. Eating rate can change when physical structure changes. Energy intake can change when energy density changes. Consumption can change when portion size changes. Intake can change substantially when researchers alter an entire dietary environment.

Modern technology allows us to manipulate all of these characteristics with extraordinary precision, while modern commercial systems make the resulting foods inexpensive, recognizable, convenient and widely available. Human appetitive biology exists inside that environment. It responds to it because responding to environments is what biological systems do.

This does not mean every eating decision is predetermined by a package, a portion or a food formulation. It means that when we try to understand why eating is sometimes difficult to regulate, the scientifically relevant question extends beyond the character of the eater.

For someone who has spent years interpreting hunger, desire or overeating as evidence of personal failure, that distinction can be consequential. The goal is not to convince them that nothing is within their control. It is to show them how much more there is to work with once they stop treating themselves as the only variable.

Instead of asking why they cannot simply overcome the environment, they can begin learning how they respond to it. They can decide which parts are useful, which make caring for themselves easier, and which they might want to change.

The question is no longer “Why can’t I control myself?” It becomes “What is happening here, and what can I change?” The first question demands a verdict about the person. The second gives them somewhere to go.

References

DiFeliceantonio AG, Coppin G, Rigoux L, et al. Supra-additive effects of combining fat and carbohydrate on food reward. Cell Metabolism. 2018;28(1):33–44.e3.

Fazzino TL, Rohde K, Sullivan DK. Hyper-palatable foods: development of a quantitative definition and application to the US food system database. Obesity. 2019;27(11):1761–1768.

Forde CG, Mars M, de Graaf K. Ultra-processing or oral processing? A role for energy density and eating rate in moderating energy intake from processed foods. Current Developments in Nutrition. 2020;4(3):nzaa019.

Hall KD, Ayuketah A, Brychta R, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism. 2019;30(1):67–77.e3.

Hollands GJ, Shemilt I, Marteau TM, et al. Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. Cochrane Database of Systematic Reviews. 2015;(9):CD011045.

Kral TVE, Roe LS, Rolls BJ. Combined effects of energy density and portion size on energy intake in women. American Journal of Clinical Nutrition. 2004;79(6):962–968.

McCrickerd K, Forde CG. Sensory influences on food intake control: moving beyond palatability. Obesity Reviews. 2016;17(1):18–29.

Rolls BJ. The relationship between dietary energy density and energy intake. Physiology & Behavior. 2009;97(5):609–615.

Rolls BJ, Morris EL, Roe LS. Portion size of food affects energy intake in normal-weight and overweight men and women. American Journal of Clinical Nutrition. 2002;76(6):1207–1213.

Related reading

Educational information only. This article is not a diagnosis or individualized medical or nutrition treatment. New, severe, persistent, or concerning symptoms should be evaluated by an appropriate healthcare professional.