To Abstain or Not Abstain: What to Do When Eating is Out of Control

Currently the number of overweight or obese adults in America exceeds 65 percent (CDC, 2016). Numerous health agencies and treatment specialists are increasingly acknowledging that for most Americans struggling with weight-related health issues, diet and exercise are not the solution (Corwin & Grigson, 2009; Gold, Graham, Cocores, & Nixon, 2009; Dagher, 2009). Recovery, instead, is more about changing behavior. For many, simply adhering to a regimen of less food and rigidly following an exercise routine produces failure, relapse, and in some instances makes the problem worse. In the most troubling cases, it can devolve into binge eating disorder (BED). The challenge begins with patients’ inability to control their eating during a binge-eating episode. In fact, “lack of control” over eating is a critical diagnostic criterion for BED (APA, 2013). But there is still great controversy in the field around whether or not binge eating should be labeled as an addiction. It is critical to acknowledge that the “addiction” diagnosis will drive treatment in a certain direction and we need to look at the efficacy of that treatment over a range of binge-eating patterns.

One key area where things grow murky is in the considerable disagreement over how the terms “food addiction” or “craving” should be defined. There is no real scientific consensus. A concise, all-inclusive, and clear scientific delineation of the concept of food addiction continues to prove elusive. In 1956, Randolph described food addiction as “One or more regularly consumed foods to which a person is highly sensitive and [which] produce a similar path and symptoms to other addictive processes.” Almost forty years later, the American Psychological Association (APA) described an addiction as a maladaptive pattern of substance use leading to clinically significant impairment or distress manifested by failure to fulfill major role obligations, physically hazardous in situations, related to legal problems, and continued despite negative consequences (APA, 1994). With the exception of caffeine, there is currently insufficient scientific evidence to label any common food, ingredient, micronutrient, standard food additive or combination of ingredients as addictive or within the category of Substance-Related and Addictive Disorders (APA, 2013). In other words, in the APA’s definition of addiction—“a maladaptive pattern of substance use”—we have no substance.

If we define binge eating as an addiction, then abstinence—the planned, disciplined way of eating—becomes a cornerstone of treatment and recovery. The premise is that “everything in moderation” will not work in a person with a biochemically addictive sensitivity. Avoiding foods altogether may seem like a release at first. One can solve food cravings, obsessions, overeating, and obesity by just saying “no.” Clinically speaking, abstinence is distinct and measureable, whereas moderation is open ended. The abstinence philosophy is often paired with Twelve Step programs, support, and the notion of spirituality. The strength gained through regular contacts with accountability partners, shared-goal meetings, and phone contacts with peers acts as a hedge against turning to eating or self-destructive consumption of drugs or alcohol.

There are multidisciplinary arguments based on genetic evidence, brain imaging, opioid or endorphin dependence, neurological similarities, cross addictions, serotonin malfunction, language in the DSM-V, and animal studies, as to whether or not food is addictive. But what really matters is how we design therapeutic guidelines that not only stop binge-eating behavior, but also demonstrate relapse prevention or long-term recovery. 

Communities formed around the “food addiction” model tend to celebrate weight loss and make it the goal. It’s an appealing notion because weight loss is easily measured. Some subscribe to the erroneous idea that food addicts “wear their recovery” because it is reflected in their size. If in fact we treat the disorder as an addiction, then treatment should follow the same fundamentals used in the treatment of alcohol and drug dependence—in other words, use complete abstinence as the measure of permanent recovery. Abstinence is needed in drug addiction because pre-existing neurological vulnerabilities combined with brain changes during the repetitive use of the drug have permanently changed the brain. The inability to stop using is said to be unmanageable and as such beyond people’s control. Food addicts, according to the addiction model, must accept their inability to eat certain foods the way normal eaters can (Foushi & Werdell, 1994). Using this model, the only hope to defeat compulsion is total omission of the specific foods or macronutrients.  

Addiction is a possible way to understand the impact of psychological factors on weight gain. It also provides implications for treatment and prevention (Brownell & Gold, 2012). But it is still up to us to determine the range over which the addiction model is useful or even applicable to food disorders. There are many areas in which the model breaks down.

For example, applying the addiction model, “food addicted” individuals will need to pinpoint the exact food or chemical component of food to which they are addicted so as to omit it from their diet completely. The problem here is that while the addictive compounds in drugs of abuse can be succinctly identified, highly palatable foods contain multiple ingredients. In searching for an addictive culprit, researchers have been unable to determine which component is addictive (Volkow & Wise, 2005). Is it sugar, fat, sugar and fat combined, salt, refined starches, highly palatable foods, processed foods, gluten, chocolate, fast foods, trigger foods, excess volume or all the above (Hadigan, Kissileff, & Walsh, 1989; Guertin & Conger, 1999; Allison & Timmerman, 2007)? Furthermore, even nonpalatable foods can come to be desired and potentially overconsumed (Pelchat, 2009). 

Studies evaluating macronutrient composition of binge episodes, both in laboratory feeding studies and in recorded information from food diaries, do not support preferential consumption of carbohydrates during binge versus nonbinge (Yanovski et al., 1992; Yanovski & Sebring, 1994; Walsh, 1993; Elmore & de Castro, 1991). There is always the very real outcome if we abstain from one food to which we claim to be addicted (e.g., jelly beans and donuts), it will rapidly be replaced by another different but equally palatable selection (e.g., chocolate bars and brownies). The addiction model would call this “cross addiction.” 

Nevertheless, there is no way to get around the fact that we can abstain from alcohol, nicotine, stimulants, opioids, and marijuana, because these chemicals are not necessary for human survival. Food, however, is. And lest we forget it, the dopamine reward circuit in the brain is tasked with reinforcing food-seeking behavior. We are met with the conundrum of the addiction model. Not only can we not omit food the way we can omit mood-altering substances, science has so far been unable to single out an ingredient, substance or family of food ingredients to omit. 

If we single out sugar, then we will need to acknowledge that biochemically all carbohydrates—be they table sugar, high fructose corn syrup, refined sugars, complex carbohydrates, whole grains, fruits, root vegetables, and even milk sugar—are transformed into glucose once they enter the body. Since carbohydrates are essential for growth, maintenance, and repair, could humans possibly omit all carbohydrates from their diet and live to tell about it? Perhaps we could limit it to simple sugars present in cakes, candies, sodas, and other sweets. But does the brain differentiate between the sugars contained in refined carbohydrates as opposed to natural sources like whole grains, fruits, and vegetables, since both are broken down into the same glucose molecule when they pass by the brain? 

The same argument could be applied to fat and the specific types of fat one feels compelled to eliminate: fried, processed, saturated, trans, omega-6, and others. When people describe the foods they crave, the list usually includes a combination of sugar and fat. Fat gives food its appealing texture (mouthfeel) and transfers the sensory stimuli (taste and flavor) more effectively (Benton, Greenfield, & Morgan, 1998). Remove the fat from a cookie or muffin and you end up with a dry or hard, unappealing product no matter how much refined flour or sugar it contains. Milk chocolate isn’t about the sugar or caffeine, but the entire sensory experience that, in part, depends on the contribution of fat.

If glucose, or even excess glucose is the culprit, the brain’s reaction to glucose would necessitate it entering the brain through the blood-brain barrier and altering the neurochemistry in a fashion similar to drugs of abuse. The sugar would stimulate the neurological reward pathways, which would eventually lead to changes consistent with the tolerance, withdrawal, and cravings associated with dependence. Historically, everything from treatment centers to countless programs like Overeaters Anonymous worked from the belief that individuals were addicted to sugar and white flour. A successful and well-known eating disorder center treated their patients by having them abstain from all sources of sugar and white flour. This was enforced, including when any form of sugar was listed among the first three to five ingredients on a food label. A good example of food containing these hidden sugars would be salad dressing, which would be avoided. 

The treatment center claimed that the foundation of their success in treating obesity was making all patients adhere to this strict abstinence protocol. But part of the dietary intervention allowed for snacks such as rice cakes. Interestingly, rice cakes and glucose share the highest score of one hundred in the glycemic index—a measure of the speed at which a substance fuels the rise and fall of blood sugar. Since the program helped participants to appropriately control their intake of rice cakes, they did not binge. Therefore, it was more likely that the control features, and not the sugar abstention, were responsible for the center’s success. 

Though it has been an outlier in much of the food addiction research, a critical component of the reward system for food intake is the sensory appeal that food creates. It is possible that those with overeating issues are responding to enhanced sensitivity in sensory brain regions. Imaging of the brain’s somatosensory area, which maps the sensitivity of the tongue, lips, and mouth, showed greater activation in the taste perception of obese individuals (Wang et al., 2002; Urasaki, Uematsu, Gordon, & Lesser, 1994).

Our understanding of the mechanisms behind cravings and perhaps food addiction is confounded by this need to consider sensory properties, specificity, and combinations (Gendall, Joyce, & Sullivan, 1997; Yanovski, 2003). Whereas drugs activate the reward system through direct pharmacological effects (Volkow et al., 2005), pleasurable food activates the brain involving at least two processes: fast sensory signals and slow ingestion processes such as increasing glucose in the brain. 

Another medically supervised hospital program illustrates the dual and durational nature of rewards activation via palatable food. Patients in the study were placed on diets without sugar or white flour. Eventually, as days turned into weeks, patients began to crave carbohydrates, and they strongly requested something sweet. Given a package of sugar much like the one you pour into a cup of coffee, in response to their requests, the patients were upset rather than pleased to receive the packet. After all, they got what they asked for, that being something sweet. What the patients shared was that they would have preferred something like a candy bar, cupcake, brownie, donut or ice cream. The bottom line is that refined sugar alone is not what they really wanted or craved. Their cravings included the need for something soft, chewy, moist, warm, creamy, smooth, and sweet. 

This added complexity allows for behavioral models to contribute to our understanding of food cravings. In fact, the foods these patients craved were often referred to as “mama foods.” They represented love, warmth, security, and comfort. 

But there were also patients who did not want something sweet, but instead requested salt. Given a container of salt, they were actually miffed. Instead of a salt container, they were hoping for chips, pizza, Mexican food or something that was icy, spicy, tingly, tangy, crunchy, crackly, and salty. Sometimes referred to as “foods designed to cheer,” this selection of food is what we seek to motivate ourselves or lift our spirits at parties, sporting events, and celebrations. 

Overall, two interesting principles were observed: 

1. The foods craved were multisensory and experienced through taste, smell, sight, and mouthfeel
2. The pleasure derived was not necessarily due to a particular chemical entering the brain, but rather to a sensory response

Moreover, foods with the greatest sensory response are those that quickly transform from a solid to a liquid to a gas. This rapid stimulation of the sensory circuitry is present in several of the most highly craved and purchased foods, such as potato chips, Oreo cookies, pizza, French fries, chocolate bars, and chocolate ice cream. 

The breakdown of cravings can have cultural overtones, such as a “need” for biscuits and sweet tea, buttermilk and rice, fast foods like burgers and fries, or even specific items down to a brand, like Ben and Jerry’s or Coca-Cola. The specificity can even occur within a food group, such as chocolate. They may exclusively crave either dark chocolate, which has psychoactive ingredients, or milk chocolate, which doesn’t. If the brain was designed to crave specific ingredients—sugar, refined grains, fat, salt—why are the desires so variable and influenced by past experiences or memories?

Research on cravings of specific foods in the past has been limited by the absence of a reliable and validated measure of food cravings. Ideally there should be an assessment tool to identify which individuals are most susceptible to hedonic eating and what trigger foods are involved. A validated measuring tool would contribute to better identification of patients for early intervention, and would help determine the nature of that intervention—which may or may not include abstinence. 

One such validated tool is the Yale Food Addiction Scale (YFAS), which was developed to recognize those with an abnormal desire for specific foods (Gearhardt, Corbin, & Brownell, 2009). The Scale correlates with greater activation in the brain regions, similar to that seen in the context of drug reinforcement—for example cues, reward, and salience—and reduced activity in the inhibitory regions. Scoring is based on the symptoms in the DSM-IV-TR for substance dependence. It is important to realize that with the YFAS, we have clearly crossed into the behavioral aspect of cravings. The YFAS focuses on the assessment of eating behavior and not on substance-based addiction. Other possible assessment tools include the Three Factor Eating Questionnaire (TFEQ; Stunkard & Messick, 1985), Food Preference Questionnaire (FPQ; Geiselman et al., 1998), Food Craving Inventory (FCI; White, Whisenhunt, Williamson, Greenway, & Netemeyer, 2002), and the Power of Food Scale (PFS; Lowe et al., 2009).

The abstinence model lends itself to black and white thinking, is geared to success or failure, and locks one into absolute behavior assessments with zero tolerance. Cross the line and there is no turning back. Avoid triggers, scrutinize one’s diet meticulously, and police it for life. One fringe benefit of this approach, proponents maintain, is the enhanced self-esteem gleaned from surmounting difficult challenges. Proponents further contend that this self-esteem has a halo effect in other areas of patients’ lives. This clearly defined measure of success appeals to those who struggle with ambiguity and prefer this dichotomous style as a familiar way of solving problems. 

However, the demand to eliminate trigger foods is conversely a good recipe for low self-esteem by defining symptoms of failure that kick in immediately when the trigger food is sampled. This approach is identified with a strong sense of self-control. Additionally, successful abstainers find they need a high level of structure and support to maintain stable recovery. If they can interrupt the chain of an impulsive response, it will initially and temporarily validate the “all or nothing” approach. Many overeaters do not respond to this approach, and even for the successful this model contains many seeds for disaster.

There are secondary benefits to describing the eating behavior as a disease or addiction. This may reduce societal stigma towards people of size (Gearhardt, 2011). Perhaps with this diagnosis bingers and overeaters can stop blaming themselves and accept compassion and seek help. The route of public blame for lack of willpower or character defects only feeds low self-esteem and deepens the problem. But a deleterious aspect of the addiction disease approach is that it suggests those who struggle with overeating will never be able to achieve control. The concept of “unmanageable” that is so helpful in the Twelve Step context provides no hope, making food addiction a cruel diagnosis. The public perception that calling food addiction a disease provides someone with “low willpower” and a self-delusional excuse to continue eating, plays out in the overeater’s psyche as well. And its presence will be reflected in embarrassment at the addiction diagnosis. And then there are the practical problems that the purely neurological explanation for binge eating—“Look, I can’t help it! My brain is hard-wired to binge!”—can cause in efforts to understand the role of environment and emotions in the disorder (Herrin & Matsumoto, 2011).

The trap is set, however, for that moment when one deviates from the triumphant and continuous days of abstinence. At this point, the person believes they have failed miserably. They can potentially spiral down from there, as the self-esteem derived from abstinence is now supplanted by disillusionment and self-reproach, to the belief that recovery is unachievable. In this way, abstinence actually increases the chances of relapse. And when the relapse does occur, it can be more devastating and prolonged. Simply addressing the problem by avoiding tempting foods will not prepare one to develop the skills necessary to deal with the real problems underlying cravings, and abstinence does not establish resilience. This population needs help developing the ability to bounce back after a lapse so they can handle mishaps with equanimity. This is a more realistic way to maintain a healthy path of recovery.   

The abstinence approach is one of passive yielding at a time when someone with the typical pattern of chronic dieting and unsuccessful attempts at weight loss needs to shift to active coping. The abstinence approach doesn’t change dysfunctional thinking patterns, but only makes the thinker dependent on rigid, externally derived standards to determine self-worth. Addiction and abstinence offer limited opportunities to learn new skills that can be applied independently to sustain continued improvement. Outcomes should be focused not just on abstinence, but also on quality of life, interpersonal relations, healthy food choices, heightened activity levels, vocational performance, emotional stability, improved self-worth, increased self-efficacy, and reducing insomnia.

The fact is, once overeating evolves into a binge, it seems impossible to overcome by sheer willpower, even if people are aware and know they need to take action and stop. While this suggests the “unmanageable” part of an addiction definition, there is a critical distinction. It is this “point of no return” in arriving at binge eating through a process, and not the chemical agency of the food substance itself, that is unmanageable. A parallel occurs in the moment of recovering from asphyxiation. Science says a person should first breathe out to rid the lungs of accumulated CO2, but no matter how aware one is of this fact, gasping for air is everyone’s response. This is a perfect example of “unmanageable” behavior. It is easy to see in this context that willpower is not an option. The reality is that when you binge and feel distress afterwards, you are out of control. But control is manageable, as we shall see, and it can be changed by intervening before the point of no return through practicing various tools provided by the therapist which include behavioral modification, CBT, and mindfulness. 

From a treatment perspective the addiction model for drugs can offer useful strategies. For example, addiction treatment employs a solution known as peer group support. These are not just acquaintances, but individuals with similar compulsive tendencies and mutual understanding. They share the same difficulties in trying to stop their destructive behavior. Because the make-up of these groups offers empathetic listeners, people to confide in comfortably and with whom to share deepest, darkest, and most embarrassing experiences and secrets, forgiveness is authentic and advice is believable. While acquaintances tell you what you want to hear, fellow compulsive eaters tell you what you need to hear. So group support is commendable, but what if the common thread is grounded in abstinence from an offending item, which in this case is food and not a drug?

At this point, it should be obvious that the black and white aspect of abstinence is impractical, on the one hand, but the sense of community support for something akin to abstinence, on the other hand, is quite useful in moments of crisis when slips, lapses, relapses, and even collapses occur. This brings us back to the cloudy issue of defining food abstinence and the resulting binge. 

Perhaps abstinence could be absence of the behavior known as “bingeing.” According to the DSM-5, a binge is defined as consuming an abnormally large amount of food in a short period of time compared to what others might eat over the same amount of time under the same or similar circumstances (APA, 2013). What constitutes a binge may mean exceeding a dietary boundary by either calories or amount, even if that amount is within “normal” limits. Or perhaps abstinence is defined as abstinence from a trigger food such as brownies or a food category such as sugar or fast food? The murkiness in our current standard of definitions becomes evident in real-world uncertainty as the individual asks, “Should I confess if I inadvertently ate a trigger-like food, but it did not initiate the loss of control?” or “I had what I thought was a trigger food, but the ensuing binge consisted of foods I did not necessarily like.”

One of the most unfortunate consequences of relying on draconian abstinence regimens plays out in the binge eater who opts for gastric bypass. In practice, these surgeries are meant only as, and are most effective when, a tool to allow one the ability to practice long-term maintenance skills. Many who qualify for surgery do so because the pattern of eating is now threatening health or life. But with scale-based weight-loss thinking and “control” relinquished to a surgical mechanism, there is little incentive to develop other tools such as slow eating, slow chewing, chewing small portions, planning, and meal preparation. Even if the practices are presented as postsurgical prescriptions, why should people be motivated to practice such tedious tasks when the weight on the scale is dropping rapidly without them changing their behavior? Likewise, if they lose weight, cease bingeing, and limit their eating through solely applying abstinence, why would they explore the underlying causation for their binge eating, acquire coping skills, adopt healthy food choices, introduce joyful movement or address their insomnia?

A nonabstinence approach around moderation would provide a more gradual reduction in the frequency of inappropriate behaviors, such as bingeing and the loss of control. This invokes replacing the trigger food with a normalized pattern of ingestion, which is another way of saying “all things in moderation.” The process may seem painstakingly slow since moderation is a monumental task, but it’s an attainable skill with great benefits. Learning moderation in meal-taking allows that the individual can achieve control—not the control of saying no, but a more balanced, if not complex control, one that gives the person the chance to establish a healthy relationship with food. 

Because the problem may have evolved as a result of repetitive inappropriate behavioral responses to stressful events or emotional upheaval, the solution may involve rewiring neural pathways and establishing alternative brain circuits. Here’s where group support can be influential. When unpredictable episodes arise, and they are sure to, support that focuses on healing can minimize guilt and shame. A forgiving atmosphere relieved of the ticking abstinence clock, can short circuit the overeater’s loop of binge, guilt, and binge. Whether we talk of neuroplasticity, resilience or neuroregeneration, the supportive community presents a pattern interrupt more conducive to long-term behavioral extinction. Add counseling to minimize the negative, withdrawal-type consequences of resisting the craving (e.g., anxiety), and outcomes can be much more permanent than those resulting from one-dimensional food elimination. 

Rather than reduce choices in an all-or-nothing absolutism, solutions need to include experimentation on how to appropriately deal with lapses and relapse. Otherwise we are not addressing the underlying problems—boredom, interpersonal stressors, anxiety, depression, poor self-esteem—that set off binges. We need to look at various self-monitoring skills like mindfulness practice to increase our ability to listen to our thinking, reduce overreaction, cultivate the sense of satisfaction, and work towards moderation. Support groups can be called upon to subvert a crisis and undermine the common practice of eating alone to avoid embarrassment.

To date there is no comparative data exploring the advantages and disadvantages of the two approaches: abstinence or moderation. Both good recovery and not-so-good recovery have been observed in both approaches, so it is imperative that science and research look at this. It’s unlikely that any single form of treatment will prove superior for all patients, because the disorder is too heterogeneous. Eventually it may be necessary to match patients with the strategies that fit their individual characteristics and circumstances. 

While abstinence is not effective as a long-term strategy, there could be advantages to advocating a window of time or “trial period,” in which peoples’ temporarily eliminated foods are identified to be triggers. Somewhat analogous to the detox period that drug addicts submit to, this period could help the individual dial down the overwhelming impulses to eat highly palatable foods, fortify appropriate decision-making skills, and develop a strong support group of accountability partners. After this transitional period, and as patients gain more self-awareness and a collection of coping skills, exposure to the trigger food can be reintroduced in a controlled setting with group or therapy support until clients gain trust that the unmanageable component of the craving is not permanent. Throughout this recovery process it’s important to continue addressing the underlying causes, measuring the mastery of skills on peoples’ specific checklists. 

Trained therapeutic resources certified in eating disorders should be part of the recovery, using any of the available tools—cognitive behavioral therapy (CBT), acceptance and commitment therapy (ACT), exposure response therapy (ERT) or dialectical behavioral therapy (DBT)—to subvert dysfunctional cognitive styles. 

Then a wide range of nuanced stabilizing skills can be developed, and there are many available. For example, emotional eaters need to identify and learn to communicate feelings, be assertive, reframe societal notions of beauty, make peace with family, and resolve abuse issues to change their eating behaviors. Patients should be reinforced for factual reporting rather than for giving information favorable to the treatment team or therapeutic community. Patients should be encouraged to analyze and modify their dysfunctional thinking styles. Success will be measured in increments and not absolutes; for instance, a half-hour binge is not the same as a binge lasting three hours, and a three-hundred-calorie binge is an improvement over a three-thousand-calorie eating episode. Giving yourself permission to eat cheesecake in the company of others is not the same as sneaking a quart of ice cream into the basement and eating until it is empty even though you are full. The suggestion that failure to achieve full symptom remission is a lack of motivation or a sign of emotional instability should be omitted from the discussion completely. Using a motion monitoring device, activity is increased throughout the day or undertaken because you enjoy it. Joyful movement leads to program longevity and is far more valuable than putting in a begrudging hour at the gym. Patients must accept that this food compulsivity is a chronic disorder that may be accompanied by co-occurring disorders. Support groups, supplemental therapy, and accountability will be a part of long-term recovery without cultivating exclusive reliance on the external support structure. Place a premium on reduced dependence on external support, setting safe boundaries, and refining self-management skills.

If we accept that binge eating is an addiction, is the cure abstinence? And if so, is the pain associated with that avoidance greater than the disease? Will permanently giving up the foods you love elevate you to a better state of mind, make you lose weight, improve your health, and maximize your quality of life? Or will you continue to experience health problems only to be deprived and constantly substituting temporary and often equally detrimental alternatives? A history of varying outcomes suggests that the answer may not be the same for everyone. It is a difficult pill for some scientists to swallow, posing a challenge as to how to advise treatment. Nevertheless, there is a need to clearly separate the underlying causes of overeating in order to appropriately assign the right form of treatment and prevention (Brownell & Gold, 2012).

It is the assertion of this article that human behavior, with regards to appetite and pleasure, covers a continuum—a spectrum much like the one we use to assess autism. It is not either/or. Appetite relates to the motivation to eat and runs the spectrum from the anorexic design, with circuitry that prompts aversion to food, to the other end of the spectrum, the classic and definable BED with the compulsion to eat literally anything to satisfy the urge. We are not looking at a specific disease, but a continuum with variations in circuitry, nuclei, neurotransmitters, transporters, and receptors. Since there are peer-reviewed papers to support both models, we can expect and should welcome variations in treatment approaches between a binge eater and those individuals who have occasional binge-eating episodes. 

Allison, S., & Timmerman, G. M. (2007). Anatomy of a binge: Food environment and characteristics of nonpurge binge episode. Eating Behaviors, 8(1), 31–8.

American Psychiatric Association (APA). (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Psychiatric Association (APA). (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author. 

Benton, D., Greenfield, K., & Morgan, M. (1998). The development of the attitudes to chocolate questionnaire. Personality and Individual Differences, 24(4), 513–20. 

Brownell, K. D., & Gold, M. S. (2012). Food and addiction: Scientific, social, legal, and legislative implications. In K. D. Brownell and M. S. Gold (Eds.), Food and addiction: A comprehensive handbook (pp. 439–46). London: Oxford University Press. 

Center for Disease Control and Prevention (CDC). (2016). National center for health statistics: Obesity and overweight. Retrieved from http://www.cdc.gov/nchs/fastats/obesity-overweight.htm 

Corwin, R. L., & Grigson, P. S. (2009). Symposium overview: Food addiction: Fact or fiction? The Journal of Nutrition, 139(3), 617–9.

Dagher, A. (2009). The neurobiology of appetite: Hunger as addiction. International Journal of Obesity, 33(Suppl. 2), S30–3.

Elmore, D. K., & de Castro, J. M. (1991). Meal patterns of normal untreated bulimia nervosa and recovered bulimic women. Physiology & Behavior, 49(1), 99–105.

Gearhardt, A. N., Corbin, W. L., & Brownell, K. D. (2009). Preliminary validation of the Yale Food Addiction Scale. Appetite, 52(2), 430–6.

Gearhardt, A. N. (2011). Can food be addictive? Public health and policy implications. Addiction, 106(7), 1208–12.

Geiselman, P. J., Anderson, A. M., Dowdy, M. L., West, D. B., Redmann, S. M., & Smith, S. R. (1998). Reliability and validity of a macronutrient self-selection paradigm and a food preference questionnaire. Physiology & Behavior, 63(5), 919–28.

Gendall, K. A., Joyce, P. R., & Sullivan, P. F. (1997). Impact of definition on prevalence of food cravings in a random sample of young women. Appetite, 28(1), 63–72.

Gold, M. S., Graham, N. A., Cocores, J. A., & Nixon, S. J. (2009). Food addiction? Journal of Addiction Medicine, 3(1), 42–5.

Hadigan, C. M., Kissileff, H. R., & Walsh, B. T. (1989). Patterns of food selection during meals in women with bulimia. The American Journal of Clinical Nutrition, 50(4), 759–66.

Herrin, M., & Matsumoto, N. (2011). If food addiction is real, how do we treat eating disorders? Psychology Today. Retrieved from https://www.psychologytoday.com/blog/eating-disorders-news/201104/if-food-addiction-is-real-how-do-we-treat-eating-disorders

Pelchat, M. L. (2009). Food addiction in humans. The Journal of Nutrition, 139(3), 620–2.

Randolph, T. G. (1956a). The descriptive features of food addiction: Addictive eating and drinking. Quarterly Journal of Studies on Alcohol, 17(2), 198–224.

Randolph, T. G. (1956b). The descriptive features of food addiction: Addictive eating and drinking. New Haven, CT: Yale University.

Stunkard, A. J., & Messick, S. (1985). The three-factor eating questionnaire to measure dietary restraint, disinhibition, and hunger. Journal of Psychosomatic Research, 29(1), 71–83.

Volkow, N. D., Wang, G. J., Fowler, J. S., Wong, C., Ding, Y. S., Hitzemann, R., . . . Kalivas, P. (2005). Activation of orbital and medial prefrontal cortex by methylphenidate in cocaine-addicted subjects but not in controls: Relevance to addiction. The Journal of Neuroscience, 25(15), 3932–9.

Walsh, B. T. (1993). Binge eating in bulimia nervosa. In C. G. Fairburn and G. T. Wilson (Eds.), Binge eating: Nature, assessment, and treatment (pp. 27–49). New York, NY: Guilford Press.

White, M. A., Whisenhunt, B. L., Williamson, D. A., Greenway, F. L., & Netemeyer, R. G. (2002). Development and validation of the Food-Craving Inventory. Obesity Research, 10(2), 107–14. 

Yanovski, S. Z, Leet, M., Yanovski, J. A., Flood, M., Gold, P. W., Kissilef, H. R., & Walsh, B. T. (1992). Food selection and intake of obese women with binge eating disorder. The American Journal of Clinical Nutrition, 56(6), 975–80.

Yanovski, S. Z., & Sebring, N. G. (1994). Recorded food intake of obese women with binge eating disorder before and after weight loss. The International Journal of Eating Disorders, 15(2), 135–50.

Yanovski, S. Z. (2003). Sugar and fat: Cravings and aversions. The Journal of Nutrition, 133(3), S835–7.