You may have had success losing weight on a low calorie, low carb, low glycemic or low fat diet. While I have found low carb to work best, both from research and personal experience, I think any of these approaches can be effective, and some diets work better than others in individual cases. But in many cases a restrictive diet is not enough and you find yourself regaining the weight. One of the main reasons for this is that you’re fine so long as you avoid certain "forbidden" foods, but you find yourself unhappy with these restrictions over the long term. When you resort to "willpower" to seriously reduce food intake or restrict the types of food you eat, your body often fights back with unpleasant food cravings that are difficult to resist. Often these cravings result in loss of self-control and you give in to the temptation to eat, undermining the diet. Sometimes all it takes is just one excursion with a high carbohydrate or high fat snack, and you re-ignite those latent cravings. This doesn’t have to happen often for you to sabotage your hard won efforts and abandon the diet. What you need an effective approach to long term weight management that is psychologically sustainable.
The root problem is the cravings. If you’ve lost weight on a restrictive diet, the cravings are still there in latent form, and they can be extremely difficult to resist when activated. The usual advice is to "be disciplined" and "exercise your willpower". But this is the whole problem! How do you do that? I address precisely that question in this post on the Deconditioning Diet. As you’ll soon see, this diet is really a systematic technique for extinquishing your cravings and gaining control over your eating without relying on "willpower", medications, supplements or special foods. The Deconditioning Diet–and Hormetism, the philosophy behind this blog– can provide you the tools to lose weight and control appetite without unreasonable effort.
The Deconditioning Diet. The purpose of the Deconditioning Diet is not to lose weight directly but rather to permanently change how you respond to food. The diet "rewires" your brain and endocrine (hormone) circuits, using Pavlovian conditioning and controlled application of "dietary stress" to gradually eliminate cravings and build up your capacity to take control of what and when you eat. This is an important distinction from other diets, which achieve weight loss by asking you to restrict what and when you eat — forever. Many of these diets do work in the short term, by reducing net calories (Weight Watchers), reduced caloric density (low fat or high fiber diets), reducing insulin and suppressing appetite (Protein Power, Atkins and other low carb diets, Mediterranean diets, Paleo diets), or using flavor control to induce satiety (Shangri-La Diet, Flavor Point Diet and Sensa tastants).
All of these diets have some merit and are more or less effective for different individuals. I personally believe that the low carbohydrate and insulin reducing diets are the most effective, based on the science so-well summarized by Taubes, Eades and others (See Diet Links in panel at right). However even these insulin lowering diets don’t necessarily alter your underlying appetite or your food preferences! You need to stay "on the wagon" indefinitely, and add or limit certain foods, supplements, or timing of eating patterns. Once you go "off" the diet, perhaps when you are tempted or stressed, your "unmasked" dietary cravings will return and you will become vulnerable to weight regain. (So called "emotional eating" is sometimes blamed for the failure of diets. Perhaps so, but emotional eating only works by re-activated dormant cravings, especially those for certain "trigger foods"). By using deconditioning techniques to silence your cravings, you will gain the freedom to lose as much weight as you want and to eat or not eat at will. It is important to realize that this will require time and patience. It’s like learning to lift heavy weights or run a faster mile. Without training, you will only be able to get so far before you exceed your capacity, and it would be futile to push yourself beyond your limits. But you’ll see progress week by week.
If you are interested in the scientific basis for the Deconditioning Diet, continue reading from this point. But if you are impatient, or bored with the science, or are interested in the science perhaps but not right now–and you just want to try it out the diet, just skip to the bottom of this page, to the section called the Deconditioning Diet Plan
. . . . . . . . . . . . . .
THE SCIENCE
. . . . . . . . . . . . . .
The Deconditioning Diet retrains how your brain and physiology respond to food and food cues, in order to reduce or eliminate the intense cravings that would otherwise undermine any diet. You can do it without any pills, supplements, surgery or other artificial means. Let’s begin by considering the nature of cravings, and where they come from.


The source of cravings. Why do we get cravings in the first place? Where does hunger come from? Clearly, most of us have enough fat on our bodies to skip the next meal, or for that matter, many meals. Studies of starvation show that indeed we can last about a week if not allowed drink. And given water and electrolytes, people of average weight can survive more than a month. A recent world record for fasting was set by Agasi Vartanyan, a Russian man who lost 58 pounds during a 50-day fast, drinking only water. We are capable of tapping into our glycogen and fat stores for energy for several days, without losing significant muscle mass. And we can spare muscle mass even longer by exercising while fasting (see Fitness page). (Note: The conventional wisdom is that fasting is unhealthy, that it invariably leads to "starvation mode" and "rebound bingeing" and eating disorders such as anorexia nervosa and bulimia. I will address those concerns below). For now the key question is why most people can’t even skip one meal or snack without irresistible cravings and loss of control over their eating behavior.
There is much confusion about the biology of hunger. The best, most comprehensive analysis I’ve read about the science of hunger is the very last chapter, "Hunger and Satiety", in Good Calories, Bad Calories, by Gary Taubes. I strongly urge anyone reading this post to get hold of that book and read the last chapter first. (The rest of the book is an excellent critique of the hypothesis that dietary fat plays a central role in cardiovascular disease and obesity). In "Hunger and Satiety", Taubes pulls together an impressive array of research by the top physiological psychologists of the past century–Bernard, Cannon, Pavlov, Mayer, Kennedy, and especially Le Magnen and Stricker. His synthesis of this research lends overwhelming support to the hypothesis that hunger and satiety are governed at a fundamental level by how hormones regulate the availability of metabolic fuels like fat and glucose in the bloodstream. And while a number of different hormones (including glucagon and leptin) and digestive enzymes are involved, it is insulin–the master metabolic hormone–that most directly regulates hunger, as well as the equilibrium of fuel partitioning between fat deposition and fat mobilization from the adipose tissue.
Hunger is perceived by the brain. But the brain cannot directly measure the total amount of caloric energy in the body. Instead, it gauges the level of available calories in the bloodstream, specifically the level of sugars, fatty acids, and ketones in the blood. In normal individuals (people without diabetes or hypoglycemia) blood sugar is controlled within narrow limits — typically between 70 and 150 mg/dl, with an average normally around 90 mg/dl — by homeostatic regulation. A relatively constant level of blood sugars and free fatty acids (FFAs) or ketones in the blood ensures that the cells throughout the body are getting the food they need. At a fundamental level it is not our entire organism, but rather the individual cells in all our tissues that need to be fed! Starving the cells in the tissues can lead to impaired cell function and even cell death. That is why we need a nearly constant level of metabolic fuels in the bloodstream. Even if we have plenty of fat and glycogen "locked away" in the pantry of our fat cells, if these metabolic fuels cannot get released into the bloodstream, our brain will never know that. The irony is that you can be obese and literally "starving" at the same time, if you can’t unlock those energy stores.


Blood sugar is hormonally regulated. If blood sugar temporarily goes up after a meal, the pancreas secretes the storage hormone insulin, which enters the bloodstream and acts to shunt the excess glucose into glycogen and adipose stores. If a person keeps eating, more insulin is secreted (along with other regulatory hormones such as leptin). A similar process occurs with blood fats. If their levels are raised, and insulin is present, lipoprotein lipase works to release these lipids from circulating triglycerides and lipoproteins in the bloodstream, and transport them to be stored in adipose tissue. On the flip side, if glucose dips, the pancreas secretes the complementary "releasing" hormones, including glucagon, norepinephrine which release glucose from glycogen stores in the liver and muscles, and adipose triglyceride lipase (ATL) which breaks down stored triglycerides from fat cells to release free fatty acids (FFA) and glycerol back into the bloodstream; and to generate sugar from glycogen stores in the liver and muscles, or from muscle protein via gluconeogenesis (literally "new glucose formation"). A simplified view of the action of insulin and glucagon to control blood sugars is depicted in the attached figure. A more complete picture would include the other hormones and enzymes involved in transport of sugars and lipids.
If the blood sugars dip quickly, and if glucagon, the stress hormones and ATL lipase are sluggish in responding to release stored energy, the brain perceives hunger. This is a signal that calories are needed. The brain detects and integrates the overall amount and rate of change in level of available metabolic fuels–glucose, fats, amino acids and ketones. Some researchers believe that the rate of change in glucose and fat concentrations may be more important than their absolute values in how the brain calculates hunger. Levels of insulin, leptin and other hormones also modulate the receptors in the brain and other tissues that determine availability. All we need to understand is that the craving is a signal that the release hormones and enzymes are not being being activated to deliver sugar and fat to the bloodstream as fast as the brain would prefer. The brain doesn’t know or care whether these fuels come from the last meal or the body’s fat and glycogen. However, if these bloodstream calories cannot readily come from storage, they must come from food, and fast. The brain’s cravings are an urgent signal to eat!
What causes blood sugars to dip? One factor is any activity such as exercise that suddenly increases the demand for energy. But normally, exercise increases the level of epinephrine and norepinephrine (adrenaline) that lead to glucose release from glycogen in the tissues. In a well adapted individual, the glucose is made available readily. In sedentary or obese individuals, it make take some adaptation to up-regulate the necessary enzymes and hormones for this so exercise can induce hunger or light-headedness if sufficient glucose or fatty acids are not at hand.
Insulin, the gatekeeper. The most common cause of a dip in blood sugar is an increase in the storage hormone — insulin, without a concurrent replenishment of blood sugar or fats. Insulin is very efficient at removing glucose from the bloodstream and pushing it into the tissues, into muscle and liver glycogen first and–when these stores are full–into adipose tissue (fat), by splitting the glucose into glycerol, which then "fixes" free fatty acids into triglycerides (fats) within the fat cells. So long as insulin is elevated, this process cannot be reversed–the insulin prevents fats and sugars from being released from the fat cells and glycogen stores. So when insulin rises quickly, with no food coming in, blood sugar dips, and we get cravings.
Eventually, within hours, insulin levels will come back down, and the release factors (glucagon and the stress hormones) will rise, allowing ready access to stored sugar and fat. If the insulin secretion is modest, blood sugar will dip only slightly, and will readily be restored. But if the insulin secretion is heavy, blood sugar will drop faster, and you might experience intense cravings, possibly hypoglycemic "hunger headaches", light-headedness, irritability and all kinds of misery while waiting for insulin to normalize. And unless you are pre-adapted to quickly up-regulate expression of the release factors, which take some time for the body to produce, the response may be sluggish and weak. In the extreme case, if blood sugar continues to drop, the hypoglycemia could put you into shock or even a coma, similar to what diabetics experience when they misjudge the dose of insulin they are injecting and it overshoots.
What makes insulin levels rise and stay elevated? If insulin levels are maintained at a very low level, we will not get hungry. So what makes insulin levels rise? There are a number of factors, several related to the last meal eaten, but others related to the individual eater:
  1. Meal composition. Insulin responds most directly to the carbohydrates in a meal. Primarily this means simple glucose and starches (which themselves readily break down into glucose). But insulin will also respond to protein, although at a lower level than carbohydrates, although the insulin will be partially offset in effect by secretion of glucagon, which responds specifically to amino acids in the bloodstream. Insulin will not respond at all to fat, so long as no carbohydrate or protein are present. (Pure fats and oils will not raise insulin). Nor will it be elevated by non-caloric components of the diet, such as fiber, minerals or water.
  2. Meal size. Whatever the composition, the larger the meal, the more insulin is secreted. Even though protein is less insulinogenic than carbohydrate and is partially offset by glucagon, you can still get a large insulin rise if you eat enough protein. Whether or not carbohydrate is present in the meal, the elevated insulin will tend to prevent glucose from coming out of storage, and fat burning will likewise be suppressed.
  3. Meal digestibility. The rate at which insulin rises is controlled by the rate at which the energy macronutrients (carbohydrates, proteins and fats) in the meal break down. So-called "low glycemic" or slow-release carbohydrates, found in high fiber fruits and whole grains, release glucose more slowly can produce a slower insulin response, with a lower peak level of insulin. Fats can also slow the release of proteins and carbohydrates, although the insulin induced by the carbohydrate will cause the fat to be readily stored, since insulin "locks" fatty acids in adipose cells by combining three of them with glycerol from each glucose molecule.
  4. Time since last meal. Insulin levels do not come down right away, but take several hours. (In this respect, simple sugars are better than complex carbohydrates. While complex carbohydrates result in a lower peak insulin level, the "slow release" can keep insulin levels elevated longer). Insulin levels reach their lowest levels during sleep, which is often the best part of the day for weight loss, unless we eat a large meal late in the day or–worse–a midnight snack.
  5. Appetizers and snacks. A small amount of a delicious food can actually make us hungrier. This may seem paradoxical if you assume that we have an appetite of fixed size, and that eating the appetizer would use up some of the "space" in our stomachs otherwise reserved for the main course. But once you understand the insulin mechanism, it becomes clear that a small amount of food can cause a quick spike in insulin and a dip in blood sugar in the short term, thereby increasing the appetite. If you were to wait an hour after a small appetizer, you might lose some of that hunger, but not all. The appetizer effect is especially pronounced when the appetizer has carbohydrates. Rodin et al. found people consuming 50 grams of glucose 2.25 hours before lunch ate 200 calories more than those consuming the equivalent amount of fructose. (Fructose causes no insulin response). In other work, she also found that the obese secrete far higher levels of insulin in response to the smell or sight of food than do lean individuals.
  6. Body fat. Overweight or obese individuals get a double dose of unfairness. Their extra body fat (especially abdominal fat) leads to health problems like cardiovascular disease, diabetes, circulatory problems and inflammatory conditions. They also exhibit higher baseline insulin levels, and the insulin levels typically respond to meals with a larger rise and slower return to baseline. (See figure below). This comes about because their more abundant body fat becomes insulin resistant, requiring more insulin to do the same job of storing sugar. In some ways, insulin resistance is an adaptive response — if you are getting more fuel than you need, try to slow down the rate of storage by making the price of admission to the store room higher. Unfortunately, if eating continues and blood sugar levels rise beyond what the body can tolerate, the sugar must go somewhere. So the pancreas responds by putting out even higher levels of insulin–paying the higher price of admission. If this goes on too long, the pancreas can burn out, unable to keep up with demand, and the result is full blown diabetes. But well before this stage, "pre-diabetes" with its characteristic chronically elevated insulin levels (hyperinsulinemia) and insulin resistance, is health problem in itself.The relevance of this to the present discussion is that the hyperinsulinemia from being overweight makes one especially vulnerable to cravings. These cravings can even shortly after eating and between meals, since elevated or quickly increasing insulin levels are constantly robbing the bloodstream of available calories and preventing glycogen breakdown or gluconeogenesis (breakdown of stored protein into glucose) to replenish it. By contrast, lean individuals have a much lower basal insulin level, and a more muted response to eating meals. as shown in the figure below (excerpted from the Fast-5 Diet by Dr. Bert W. Herring), whereas overweight individuals have elevated insulin levels most of the time. So thin people have an easier job of resisting temptation. Eating literally leads to more eating and staying thin makes it easier to stay that way.
  7. Appetite cues. Many people may be surprised to realize that one of the most significant factors influencing insulin secretion is psychology–the aroma, sight, flavor and even thought of food! While most of the insulin secreted in response to a meal occurs after ingestion (the post-prandial phase), anticipation of a meal and the many cues that signal "food is coming" will induce a substantial secretion of insulin even before any food enters the mouth or the digestive tract (the pre-prandial or cephalic phase). This goes back to Pavlov’s physiological psychology of digestion, discussed in detail on the Psychology page of this site. Pavlov identified enzymes such as amylase among the digestive fluids that he isolated from salivating dogs, but he did not at the time realize that insulin and other hormones (including the hunger signaling hormone ghrelin) were part of those secretions. The secretion of insulin prior to eating can lead to a rapid drop in blood sugar and fatty acids, with a corresponding an increase in appetite. As Taubes says (GCBC p. 443): "the thought of eating makes us hungry, because the insulin secreted in response depletes the bloodstream of the fuel that the peripheral tissues and organs need to survive."Pre-prandial insulin secretion serves a useful purpose: to encourage eating of food when it is available, and to prime digestion to adapt to the quantity and even type of food being ingested. In her paper "Physiological Effects of Flavour Perception", Karen Teff indicates that pre-prandial insulin is secreted rapidly, within two minutes of stimulation by aroma, flavor or other food cues and peaks after about 4 minutes. By contrast, post-prandial insulin secretion does not start until 15 minutes after eating, and peaks at about 30-45 min. In lean individuals pre-prandial insulin secretion may represent only about 3% of post-prandial insulin release. But it has been noted that in obese humans and rats, pre-prandial insulin is significantly higher. Furthermore, pre-prandial insulin by "priming" appetite, leads to a significant increase in eating behavior, and thereby can be indirectly credited for causing much of the post-prandial insulin.

Neural circuits. Teff’s paper describes in some detail the nerve pathways that connect flavor and aroma to insulin secretion by way of the brain. Flavors and aromas are detected by sensory receptors in the nose, throat and mouth. These receptors activate neural fibers into the brain center for taste, specifically the nucleus of the tractus solitarus. Here, the strength of the signal can be modulated based on experience. The tractus solitarus then relays a signal to the vagus nerve, a major nerve in the parasympathetic nervous system, which in turn branches extensively into the primary digestive organs: the stomach, intestines, pancreas and liver. It is the vagus nerve which directly controls pre-prandial secretion of saliva, gastric acid, pancreatic enzymes and hormones. What this all means is that olfactory signals control pre-meal secretion of insulin, but that the strength of the signal can be modified by conditioning.
Teff doesn’t mention it, but I believe it is likely that the tractus soliatus–the brain center than modulates the vagus nerve–responds to more than just olfactory cues. It is likely that this site in the brain can be conditioned to respond to or ignore other environmental cues, such as appearance, texture and even apparently arbitrary conditioned stimuli like the flashing lights and musical tones used in Pavlov’s experiments. This brain center then integrates these responses automatically and passes the integrated signal on to the vagus nerve. Other brain "reward" centers, such as the hypothalmus, nucleus accumbens, and dorsal stratium are also involved in conditioned responses to food cues. One study by Rothemund et al in Germany showed that that the dorsal striatum was activated in obese women–but not lean controls–merely by their looking at photographs of high calorie foods.
Food cues and conditioning. Pavlov’s dogs naturally responded to the sight and smell of food (so-called unconditioned stimuli) but became conditioned to respond to perceptual cues that were not inherently or initially associated with food (conditioned stimuli). These conditioned stimuli included bells, flashing lights, and even specific musical passages. The dogs could learn to salivate and become hungry at specific times of day or after fixed intervals. Remarkably, both the conditioned and unconditioned stimuli could be readily unlearned. It is not surprising that dogs could be deconditioned from responded to buzzers and lights, but what is most surprising is that they could learn to no longer salivate when presented with food.
Like Pavlov’s dogs, we have become conditioned to a whole host of cues in our environment. These cues may be so-called "unconditioned stimuli"–cues that are naturally associated with food, such as the appearance and aroma of the food, or they may become associated through experience, such as the room we eat in, the social environment, the pattern of mealtime within our day, the aroma of coffee in the morning, driving up to our home after work, the words "dinner is served". Some of these cues have an especially strong and direct effect – odors and flavors in particular. The vagus nerve is particularly responsive to odors and the aromatic component of flavors. (Pinch your nose while eating and you’ll notice that most of flavor is smell, except for the limited range of sweet, salty and bitter perceptions that the tongue can detect).
The food cues we respond to are biologically arbitrary and culturally learned. What is appetizing in Japan or Ethiopia may not be appetizing in the United States. Seth Roberts recounts the story of a Gaku Homma, Japanese cookbook author whose initially impression of Coke was that it tasted "like medicine". If you think that food odors are naturally appetizing, ask yourself: Why do we respond to food odors differently than other evocative and pleasant odors – flowers and plants, soil, sea, even pleasant or sensual human scents? Smelling a rose does not make you hungry. Could we be conditioned to salivate and secrete insulin in response to the smell of a rose if we always sniffed a rose before gulping down a sweet drink? Rats that did not like the taste of saccharine, grew to like it when they were intravenously feed glucose.
Keys to control of appetite and cravings. We can boil down the above research findings to three key points:
  1. Insulin control. A sudden rise in insulin level immediately causes hunger, so to control appetite we must keep insulin at a low and steady level.
  2. Insulin response to food cues. Insulin increases prior to and during eating, as a response to unconditioned and conditioned food cues such as taste, odor, texture, appearance and peripheral environmental stimuli.
  3. Conditioning of insulin response. Pre-prandial insulin secretion can be deconditioned from existing food cues or conditioned to new cues.
Let’s take up these points in order, with regard to the current array of popular diet and exercise protocols:
1. Insulin control: A sudden rise in insulin level immediately causes hunger, so to control appetite we must keep insulin at a low and steady level. The importance of insulin control is well known for its importance in weight loss, and forms the basis of low carbohydrate diets. Many advocates of these diets, including the Atkins, Zone, and Protein Power diets, as well as various Paleo diets recognize the role that a generally low insulin level plays in weight loss and health. Other protocols, including low glycemic diets, overall calorie restriction, intermittent fasting, and exercise have also been studied and advocated as methods of lowering insulin as a means of weight loss and improved health.
However, even among these advocates, insulin reduction is not fully appreciated as the basis for hunger signals and cravings. And while overall reduction in insulin levels is stressed, what is usually overlooked is the role that immediate, short-term fluctuations in insulin play in the psychology of appetite.
2. Insulin response to food cues: Insulin increases prior to and during eating, as a response to unconditioned and conditioned food cues such as taste, odor, texture, appearance and peripheral environmental stimuli. The stimulus-response relationship between food cues and insulin secretion is much less recognized, and mentioned barely if at all by the advocates of diet and exercise.
There are three recent dietary approaches that explicitly recognize the relationship between flavor and appetite: Seth Roberts’s Shangri-La Diet, and two approaches based on the concept of sensory-specific satiety: David Katz’s Flavor Point Diet and Alan Hirsch’s Sensa weight loss program. These are discussed in some detail in my blog postings on Flavor Control Diets. In different ways, these flavor control diets work to suppress the insulin response to flavors, either by using non-flavored food (thereby suppressing pre-prandial insulin) or by introducing a saturating level of a single flavor (thereby limiting the extent of pre-prandial insulin compared to multiple or complex flavors, which saturate at much higher levels). The resulting lower insulin levels do suppress appetite for a while. But as soon as flavor cues are reintroduced in combination with each other or with added insulinogenic calories, the appetite suppression effect goes away. In these flavor control diets, you have not unlearned any flavor cues, you just avoid their effect by sidestepping them based on food choices.
3. Conditioning of insulin response. Pre-prandial insulin secretion can be deconditioned from existing food cues or conditioned to new cues.This third point is almost totally ignored. In all the above diets, the stimulus-response relationship between food cues and appetite are taken as a given. Appetite is suppressed by avoiding or limiting cues like flavor, but not altering or extinguishing them. So the flavor control diets are not Pavlovian diets, because they don’t change conditioning. This is curious, because the research cited by Roberts and Hirsch actually support Pavlov’s early findings that food cues can be conditioned or deconditioned. (More about that in my blog posting on Flavor Control Diets). And that fact that food preferences and their intensity are not inborn, but can change significantly across cultures or within the life of a single individual.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
THE DECONDITIONING DIET PLAN
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Plan. The most important of the five principles of Hormetism is gradualism. Dramatic changes can be made in how we respond to food, but it is best to do this slowly, in small steps. Therefore, I have broken the diet plan down into three phases. You could skip ahead to the final phase if you want fast results and are highly motivated, but I think the chance of success is best if you follow the recommended order.
Phase 1. General insulin reduction phase. (One week). If you have been on a low carb diet, a high fiber diet, or a very low calorie or intermittent fasting diet, you can skip this phase. If you generally don’t have a strong appetite, but get cravings only for a very limited number of "trigger foods", you can skip this phase. But if you have a strong appetite all day long, snack throughout the day, and cannot miss a meal without going bonkers, then I recommend you start here with Phase 1. The goal of this phase is to reduce your basal insulin and normalize your blood sugar stability as much as possible and as soon as possible. This phase will generally take about a week. Follow these three guidelines in Phase 1:
  • Follow a low carb diet. I’m not going to review the details here, since there is so much information on this available elsewhere. Basically, just avoid or minimize consumption of foods and drinks with sugars and starches. The best plans are Atkins, Protein Power, Paleo or any plan in which carbohydrates are limited to no more than 100 grams per day. You can be a bit looser than that if you want, and go as high as 200 grams of carbs per day. Any higher and your insulin will stay high. The Zone and South Beach diets allow too many carbohydrates. Trendy "whole grains", fruits and other "low glycemic" carbs are still insulinogenic. Anything more than 200 grams per day could spike your insulin levels, make weight loss harder, and maintain your ravenous appetite. using normal diet and exercise. You can have an occasional treat–an alcoholic beverage, or a dessert–as long as you keep under your limit of 200 grams carbs per day. (Read labels!). You can also follow any other diet plan (e.g. Shangri-La or Flavorpoint) or exercise program that is compatible with low carb.
  • Schedule your meals and avoid snacking. Contrary to much of the advice you’ll hear, you should eat only at defined mealtimes and no more than three times a day. All meals must be at least 3 hours apart, preferably longer. This idea of meal scheduling comes from the Pavlovian concept of stimulus control–putting your behavior and reactions "on cue", so that you can control them. (See discussion on the Psychologypage). You can choose your mealtimes and you can eat as much as you like at each meal, but try to keep the actual eating at meals to less than 45 minutes. Schedule the meals ahead of time — decide the previous day when you will eat, don’t eat based on hunger. You can modify the schedule during the day, as long as you decide at least 30 minutes before eating when you are going to eat. Do not eat after dinner, and never within two hours of bedtime. This may seem rigid, but it is important. Allowing long breaks without food between meals (and especially at night) is absolutely critical to establishing a low basal insulin level. The idea of frequent small meals or snacks "at will" throughout the day has been promoted as a way to keep keep hunger at bay. That approach is wrong. While it may avoid insulin spikes on the high end from large meals, it never allows your insulin levels to hit bottom. Small snacks continues to "prime" your appetite by reinforcing food cues as stimuli for pre-prandial insulin secretions. Even small caloric treats can sabotage you. One candy, one nut, one mint—you are back to Square One! Save your nuts to eat as an appetizer before a meal, and your candy to eat as a dessert with your meal. You can have calorie free beverages (coffee, tea, or herb tea) between meals with no added sugar, but you can drink as much water as you would like.

  • If you cannot avoid snacking, give it up in stages. If you must snack, make sure the snacks are mostly fat, and have no more than 10 grams carbs or 5 grams protein. Then, transition to giving up snacks. Start with a small defined "no snack" window that you can live with. The most important no-snack window is the one is after dinner. Give up all eating during the two hours before you go to sleep. This will allow your insulin to "hit bottom" during your sleep. Once you have that in hand, give up snacking within two hours before lunch and dinner, but allow snacking otherwise. Then go to no snacks during the week, but allow cheating on weekends. Then eliminate snacks entirely. This gradual approach may work if you can’t do it all at once. You can also use shaping to reduce snacking behavior (See Psychology page). If you have cravings, wait until 15-30 minutes after they subside to a low rumble before eating. Never reinforce a raging hunger craving by eating when it is most intense, or you will reinforce it!
  • Prepare "emergency" snacks. You may find the above recommendations easy to follow most of the time. Usually you can convince yourself to postpone a snack time craving until the next meal, since you can eat as much as you want at mealtimes. But every so often you may get a horrendous craving for something caloric between meals or after dinner. You need to plan for this and always have available high caloric, non-insulinogenic snacks. It is important that these snacks have no or little glucose or starch, and even protein should be minimized. A high fat snack is best. Here are some of my favorites: (a) a sugarless cream soda: in a tall glass, pour two tablespoons heavy whipping cream over ice cubes. Fill the glass with Calistoga or other sparkling water. You can even add a little flavoring like vanilla or orange extract if you want (and vary this to keep it interesting), but nothing with sugar or other carbs; (b) herb tea or decaf coffee with heavy cream; the cream makes this luxurious and filling. Don’t worry about the saturated fat, because you will "burn" it, and not store it, if there are no carbs; (c) EVCO (extra virgin coconut oil: this contains medium-chain triglycerides (MCTs) which are readily metabolized to give instant energy and appetite suppression; can be taken straight, combined with heavy cream, or "floated" in tea or coffee; (d) macademia nuts (a truly high fat nut); (e) non caloric drinks sweetened with max 1 tsp of pure non-insulogenic sugar: preferably erythritol or xylitol; don’t add sucrose, HFCS, or other glucose containing sugars; (f) ultra low glycemic fruits: a small serving blueberries, strawberries, apples, oranges or grapes — the three least insulinogenic fruits. No bananas, watermelon, figs or other sugary fruits! You can pour cream or erythritol-sweetened cream over the fruit.
  • Start a craving log. Play detective and start writing down your craving episodes to figure out what your cues are. Low carb diets will generally reduce cravings, but probably won’t eliminate them. It is important to note all the cravings you have during Phase I. Write down the following information in a pad or electronic device, that you always have with you. Note the time and date, the nature of the craving (general hunger or for a specific food, and what may have set off the craving. Seeing or smelling food? A stressful conversation? Tired after work? Or just the time of day? If you don’t know, speculate. (Note: I have recently being experimenting with a blood glucose monitor as a way to better understand how hunger and cravings influence blood sugar and available energy. You might consider this if you are inclined to be a self-experimenter).
Phase 2. Cue extinction and counter-conditioning. Continue with the low carb diet. If you are starting here, start a craving log as described in the last bullet under Phase 1. Now you are going to start to deal with your cravings directly! Use the following guidelines during Phase 2:
  • Never eat when you are hungry. This is the goal of Phase 2. Eating when hungry reinforces your appetite. You are going to learn to eat only when you are not hungry. Often this just means waiting until the hunger subsides, then eating.
  • Devise a counter-conditioning plan. Before you start the next step, you will need to do some planning. Make a list of two or three activities you like to do that do NOT involve eating or being near food, that you can do immediately after your your cue exposure session, and that you can sustain for at least 15 minutes. Be ready to do one of these activities after your next craving, and vary the activities. They could include going for a short walk, making a phone call, taking a bath. The more active the better –watching TV or reading a book may not get your mind off food.
  • Expose yourself to cue signals without eating. Read over your craving log to note the specific food cues, including time of day, that triggered your cravings. Now your want to "simulate" the cravings to stimulate your appetite, but without giving in. You can do this two ways — proactively and when the cravings happen between meals. (Don’t worry about cravings right before meals, just enjoy eating!). Whether or not you are having the actual craving, go to the refrigerator or do whatever you would normally do in that situation. Visit the kitchen, the pantry, open the refrigerator. Go look close up at the specific foods, smell them. If you are in your own home, you can touch and handle the food. Inhale the food fragrance in 5 to 10 deep breaths, enjoying the pleasure of each inhalation. You can also do this with coffee, alcohol, anything you like. One great way do to cue exposure is when you are accompanying others to eat (when you don’t) or preparing a meal for others. Enjoy the food prep, just don’t eat the food while you are cooking. If you away from home, walk though a market, bakery, or coffee shop with aromas that you like, foods with a prominent signal like doughnuts, cinnamon buns, coffee or cocoa. Or with visual cues like ice cream, chocolate or candy. You can pick one or two specific foods at a time and work on those. You can also apply this to alcohol, either using odor or appearance of drinks as cues. Remember cue is the full context, including the room and walking into the room at that time of day. Don’t be afraid of exposing yourself to the food and experiencing cravings, look forward to it!.
  • Vary the food cue exposure. Do the cue exposure between meals at different times of day, in different contexts. Include a variety of foods. Don’t avoid your strongest trigger foods. If you are worried you might give in to them, then try other less craved foods first. To keep the cravings in check, I recommend saturating your sense of smell by alternating between smelling the trigger foods and other fragrant food smells that don’t give you cravings. Smelling a variety of spices (curry, oregano, rosemary, ginger) works especially well. The idea here is that multiple fragrances will fully saturate your sense of smell and dampen the craving response.
  • Implement your counter-conditioning plan. Have your planned activities ready and in your head. If it helps, implement the plan right after your cue exposure session. You may find that you don’t need to.
  • Eat later. Keep your eating confined to defined meal times, if you can manage to, and never eat immediately after doing food cue exposure session. Always insert at least 30 minutes (or a 15 minute counter-conditioning session) between the cue exposure and the meal. Eating must come at least 15 minutes after the cue exposure session. If you can wait 30 minutes or an hour that’s better, but 15 minutes will work. That’s because pre-prandial insulin secretion is brief, and ends within about 15 minutes. Most importantly–do not eat the very food you were craving! Eat something else, preferably in a different food class. If you craved a sweet, eat something not sweet, like meat or cheese. If you craved something spicy, eat something bland. You can eat a whole meal if it is time for that. You can return to the craved food on a day that you aren’t craving it.
  • Beware of extinction bursts and resurgence. You might have some initial success in suppressing your food cravings. Good for you, but remain vigilant! Strong cravings have a way of popping up again. Like a crying baby, a craving will fight to get your attention. It may be a hours later, or days later…even weeks after you thought you had dampened the craving. It may be that the food cues are slightly different and the craving arises in slightly different context. Perhaps at a different time of day. Or when you are stressed. Treat each craving as an opportunity to demonstrate your extinction and counter-conditioning techniques. And remember, the worst time to give in to a craving is during an extinction burst or resurgence, because this will strengthen it just like hitting the jackpot strengthens a gambling habit. Resurgence is the best time possible to execute your counter-conditioning plan! At the very least, "shape" intense cravings to become weaker by waiting until they have substantially died down before eating anything.
Phase 3. Meal skipping. Don’t move to Phase 3 until you are having some success with suppressing cravings in Phase 2. This could take a few weeks, but it may be shorter. Your hormone regulation is rebalancing, and this adaptation can take a few weeks. You may still experience headaches and cravings, and the cues may be different. Keep your craving log going at least until you are confident that you understand your main food cues and temporal cues. In Phase 3 your are going to increase the intervals between meals. The goal here is not to lose weight (although you will) or to do a spiritual penance of some sort. Rather, the goal is to further dampen your insulin levels and further increase your independence from food cravings. You will gain the freedom to eat when you want and not be a slave of the clock. At the same time, we are not advocating that you acquire an eating disorder! You still love food and need food, its just that you don’t need it so often and you want to be able to lose weight.
But first, a word about "starvation mode" and eating disorders. There is a lot of scare-mongering out there regarding something called "starvation mode". The claim is that you must snack every few hours and never skip a meal, because your metabolism will slow down signficantly and weight loss will halt. And if you fast for more than a day, your body will somehow enter "starvation mode" and start consuming muscle protein. This is a myth. It has been adequately debunked by a number of studies of fasting. Not only does reducing your caloric intake not cause your metabolism to slow down, it also does not result in a loss of your hard-earned muscle. There is one imperative rule that goes along with this statement: you have to be involved in some sort of resistance exercise, such as lifting weights. While long term caloric restriction on its own can cause you to lose muscle mass (such is the case with hospital patients who are on a low calorie diet and confined to bed rest), the combination of caloric restriction with resistance exercises has been proven to be very effective at preserving your muscle mass. Research published by the American College of Nutrition showed that men and women undertaking a 12 week long, very low calorie diet consisting of only 800 calories and only 80 grams of protein per day, were able to maintain their muscle mass as long as they were exercising with weights three times per week. Additional supporting information can be found on the excellent sites on intermittent fasting and exercise at Log My Loss, Fitness Blackbook, Leangains, and Eat-Stop-Eat.
Some of you may also be concerned that fasting will lead to eating disorders such as anorexia or bulimia. The literature is fairly clear here, that these conditions are caused by specific psychiatric or metabolic disorders. Eating disorders may result in fasting, but fasting does not cause eating disorders. Nevertheless, to be very clear, I do not recommend intermittent or extended fasting for anyone with a metabolic or hormonal disorder such as diabetes or hypoglycemia, or for anyone who has or suspects that they have an eating disorder.
Here are the eating guidelines for Phase 3:
  • Continue your low carb eating and snack avoidance. But if you start to get carb cravings, be vigilant and implement cue extinction and counter-conditioning. By now, you should not need to resort to emergency snacks very often, but have them ready at hand in case.
  • Continue your craving log.
  • Begin to skip meals and/or vary meal times. Pick some strategic meals during the week that you can miss. The easiest might be breakfast, but lunch is fine. If you find that that is too hard, an easy alternative is to vary the mealtimes to gradually increase the time between meals. If your three meals were at 8 a.m., noon and 6 p.m., then switch to eating breakfast at 7 a.m. or dinner at 7 p.m. Eventually, skip breakfast and eat an early lunch. During this period, experiment with just changing the scheduled meals times by 30-60 minutes in either direction to add variabilty. But set the schedule at least 30 minutes ahead of actually eat. You can work around social commitments. You want to do two things: increase intervals without food and disrupt patterns. This will really help to get rid of cravings that are based on time of day cues. Eventually, try to skip the hardest meal of the day for you–probably dinner. Note your reactions in your craving log. Your endpoint should be no more than two meals a day. If you can handle one meal a day, great, but consider two meals a victory. A fuller discussion of the most
  • Build up to doing longer fasts at least once a week. This is a technique known as "intermittent fasting" (IF) that is very helpful in deconditioning appetite as well as improving insulin sensitivity and overall metabolic health. Don’t do this if you are diabetic, hypoglycemic or have a metabolic disorder, without a doctor’s oversight. For healthy individuals, this is quite safe. There are good many websites on intermittent fasting and different schools of thought regarding the ideal length and frequency of the fast — which can range from 12 hours every day, to 24 hours once or twice a week. Check out especially fast-5 and Eat Stop Eat, if you are interested. These mini-fasts are actually much more difficult in the anticipation than in the doing. Many people find they feel wonderful on their fasting days–clear headed, full of energy, focused. I and many others even find that exercising and intense workouts on fasting days bring great results. If you’ve built up slowly by adapting to a low insulin regimen, you may find the same. Do not eat right after a workout and you will get additional benefits by prolonging the drop in insulin You may experience what is known as "benign dietary ketosis" during these short fasts. This is a completely healthy phenomenon, not to be confused with the dangerous ketoacidosis and blood acifidication that diabetics get from insufficient insulin. During ketosis, you will find your blood sugar is typically very even and you have bountiful energy while you are living off your own fat! For a fuller discussion of how to get started with IF, see my post on Learning to fast. Bon apetit!
Success. I hope you will be as successful as I have been in suppressing your appetite, eliminating cravings, and losing weight. By eating less frequently and less reactively, you will avoid the short term insulin spikes that create cravings in the first place. So once started, the Deconditioning Diet is self-reinforcing. This leads to long-term stable blood sugar levels and a nice even energy. You may find, as I have, that your mood improves, that you become less irritable. These are all consequences of continuous access to easy energy. The door to your "fat pantry" will stay open longer and more consistently, because insulin spikes are not there to close the door. Further, by eating less and less urgently, your levels of glucagon, the hormone involved in gluconeogenesis and lipolysis, will increase, and stored glycogen and fats will be more easily broken down to raise blood sugars and fatty acids, increasing your sense of satiety throughout the day.
Be careful not to overdo it. Know your healthy goal weight and don’t overshoot. You still want to enjoy eating, just not in a compulsive way. The Deconditioning Diet is not a path to eating disorders, because it is based on the principle of positive reinforcement, not aversive conditioning. Food is not the enemy; it is and should remain a pleasure. You’ll still have a healthy appetite and eating food will be just as pleasurable as before, but not compulsive. By eating at pre-scheduled times, you will be in control of your appetite, rather than your appetite being in control of you. By alternating between eating and fasting, you will have adapted your body to be able to readily switch between storage mode and fat burning mode. (People who eat or snack every few hours can’t do this very well). You may be eating less, and less frequently, but you can still enjoy food to the maximum. And you’ll have the additional benefit of a constant and even level of energy, something that few people have.
Please leave a comment. You can also check out (and maybe start) a discussion on the Diet Forum linked to this blog.