Learned Overeating: Applying Principles of Pavlovian Conditioning to Explain and Treat Overeating

Learned Overeating: Applying Principles of Pavlovian Conditioning to Explain and Treat Overeating Purpose of Review This review provides an overview of recent findings relating to the role of Pavlovian conditioning in food cue reactivity, including its application to overeating and weight loss interventions. Recent Findings Both in the laboratory and in real life, cue-elicited appetitive reactivity (e.g., eating desires) can be easily learned, but (long-term) extinction is more difficult. New findings suggest impaired appetitive learning in obesity, which might be causally related to overeating. The clinical analogue of extinction—cue exposure therapy—effectively reduces cue-elicited cravings and overeating. While its working mechanisms are still unclear, some studies suggest that reducing overeating expec- tancies is important. Summary Pavlovian learning theory provides a still undervalued theoretical framework of how cravings and overeating can be learned and how they might be effectively tackled. Future studies should aim to elucidate inter-individual differences in Pavlovian conditioning, study ways to strengthen (long-term) extinction, and investigate the working mechanisms of cue exposure therapy. . . . . Keywords Pavlovian conditioning Food cue reactivity Overeating Obesity Food cue exposure therapy Introduction halting, and reversing the obesity “pandemic” has become a priority in public health. Overweight and obesity prevalence has reached dramatic pro- It has been proposed that the “obesogenic” environment is a portions. It is estimated that currently, around 2.1 billion indi- main contributor to weight gain and obesity, for it promotes a viduals worldwide are overweight (body mass index [BMI] > sedentary lifestyle and provides an abundance of easy-to-get 2 2 25 kg/m ), of which 600 million are obese (BMI > 30 kg/m ) high-calorie foods (e.g., [9, 10, 11� ]. Specifically, it is assumed [1, 2]. Given that obesity is associated with an increased risk that food-associated cues play a major role in overconsumption for diabetes, cardiovascular disease, psychological problems by evoking appetitive responses (such as food cravings), moti- and various mental illnesses [3–6], and societies face explod- vating one to (over)eat. Human Pavlovian learning in such “food ing healthcare costs due to obesity-associated morbidity [7, 8], cue reactivity” has however received little attention in both sci- entific studies and obesity interventions, even though experts have emphasized its importance in overeating and its contribu- tion to the difficulty in achieving long-term weight loss [12–15]. This article is part of the Topical Collection on Food Addiction This review provides an overview of recent findings on Karolien van den Akker and Ghislaine Schyns equally contributed to the Pavlovian learning in food cue reactivity and food intake, includ- manuscript. ing its application to weight loss interventions. * Karolien van den Akker karolien.vandenakker@maastrichtuniversity.nl * Ghislaine Schyns Food Cue Reactivity and Pavlovian ghislaine.schyns@maastrichtuniversity.nl Conditioning Faculty of Psychology and Neuroscience, Department of Clinical The obesogenic environment includes numerous food cues, Psychological Science, Maastricht University, P.O. Box 616, 6200, MD Maastricht, The Netherlands such as the smell of freshly baked bread when walking past 224 Curr Addict Rep (2018) 5:223–231 a bakery or the sight of delicious Italian ice cream when hiking Though these findings seem in line with a role for though the park. Exposure to food cues activates a central Pavlovian conditioning in food cue reactivity, a relevant ques- appetitive state [16], which consists of diverse responses: psy- tion is whether the findings are ecologically valid. The labo- chological (craving, urge, or desire to eat), physiological pre- ratory experiments are generally conducted under highly con- paratory (e.g., increased salivation and insulin release), and trolled conditions with very few distractors present and using neurocognitive (e.g., brain activation patterns, allocation of arbitrary stimuli as CSs. In contrast, in daily life, CSs concern attentional resources). The physiological preparatory re- non-arbitrary naturalistic stimuli, and the CS–US associations sponses (cephalic phase responses) aid in digestion, absorp- will, for example, be more spread over time and embedded tion, and metabolism [17, 18], and overall, stronger cue- within a more distracting environment. In a recent study, we elicited appetitive responding is thought to promote therefore examined whether appetitive conditioning also oc- (over)eating [11� ]. Indeed, increased levels of food cue reac- curs under real-life circumstances, using times of day as con- tivity (e.g., cue-elicited desires to eat) have been associated ditioned stimuli [42� ]. Participants were tested over a period of with overeating, unsuccessful dieting, a higher BMI, and eat- 15 days in their natural environment, using a smartphone ap- ing psychopathology [11� , 19–24]. plication for administering a questionnaire (eating expectan- Although food cue reactivity has a strong genetic component cies, eating desires) and for instructing them to consume choc- [25], Pavlovian conditioning plays an important role as well. olate provided by the experimenter. At one specific time of Every time palatable food is consumed, the food can be asso- day (CS+), participants always received a signal to complete ciated with cues in the internal or external environment, and the questionnaire and the instruction to consume chocolate, these cues then promote cue reactivity upon their next encoun- whereas on another time of day (CS−), participants only com- ter. In Pavlovian terms, food entering the digestive system is an pleted the questionnaire. Results indicated considerably unconditioned stimulus (US), and once a cue (conditioned stim- heightened eating desires and eating expectancies to the ulus; CS) has become associated with the US, it can stimulate chocolate-associated time of day (vs. CS−) from the fifth appetitive responses, i.e., food cue reactivity (conditioned re- day forward [42� ]. Thus, consistent with the findings of labo- sponse: CR), which promotes food intake [13, 21, 26]. For ratory studies, this study shows that cue-elicited eating desires example, consider Carl who repeatedly eats crisps in the eve- can be learned with relative ease in the natural environment. ning when watching his favorite TV show: this context (CS; Associative learning processes might therefore play a critical watching his favorite TV show in the evening) may become role in the experience of many daily eating desires. associated with eating crisps (US) and if this is the case, just the Given the potentially important role of Pavlovian learning in CS, that is, just the look of Carl’s favorite TV show, will elicit a the onset and maintenance of obesity, several recent studies desire for the crisps and promote its intake. Theoretically, any have examined whether overweight individuals differ in appe- titive Pavlovian learning from normal-weight individuals. cue can become associated with palatable food intake, includ- ing the sight and smell of food, a hormonal or other internal Interestingly, the findings point towards impairments in acquir- physical state like hunger or satiety, specific rituals, cognitions, ing appetitive responses (US expectancies, US desires, CS lik- emotions, and so on [21, 27–29]. Several laboratory studies ing/preference) to food-associated stimuli in overweight and have examined this Pavlovian learning account of food cue obese individuals [34� , 43, 44� ]. In two studies, this was reactivity by repeatedly pairing an initially neutral stimulus reflected by reduced differential responding (i.e., smaller differ- (CS+) with tasty foods (such as a piece of chocolate or a sip ences in responding between the CS+ and CS−)[34� , 44� ]. of milkshake), and comparing the appetitive responses to this These findings are intriguing and raise the question if and stimulus with the appetitive responses to another stimulus (CS how such impairments are causally linked to overeating. It −) which is never followed by food intake. Stimuli that have might be that heightened CS− responding (which drove the been examined in these studies include trays [30, 31], boxes deficit reported in [44� ] reflects an impaired ability to inhibit [32], geometric figures [33, 34� ], objects [35], negative emo- appetitive responses to stimuli signaling non-reward in an ap- tions [36], and (virtual) environments/contexts [37–39]. A con- petitive context, which, in the natural environment, might plau- sistent finding is that after conditioning, the CS+ (vs. CS−) sibly increase reactivity and overeating to stimuli not associated evokes conditioned responding, as reflected by heightened psy- with food intake [45, 46]. The findings might also indicate a chological (e.g., explicit eating expectancies, eating desires, tendency towards overgeneralization in overweight and obese CS+ liking) (e.g., [30, 38, 39], psychophysiological (e.g., sali- individuals [34� , 44� ]: a reduced ability to discriminate between vation) [34� , 35, 40], and behavioral responding (e.g., approach predictive and non-predictive stimuli is a core feature of over- tendencies, place preferences, food intake/choice) (e.g., [30, 36, generalization, and heightened responding to the CS− could 37, 41]). Findings also suggest that appetitive conditioning usu- result from overgeneralization from the CS+ to the CS− [47]. ally occurs quickly; in most studies, only 3 to 6 CS–US pairings In the example of Carl, overgeneralization of appetitive were sufficient, while one study even reports evidence for suc- responding might imply that while he originally learns to asso- cessful acquisition after only one CS–US pairing [35]. ciate the eating of crisps to a particular context (e.g., sitting on Curr Addict Rep (2018) 5:223–231 225 the couch in the evening and watch his favorite TV show), he is expectancies, eating desires) reduce with non-reinforced CS also cue reactive to a wider variety of stimuli that are similar to presentations (e.g., [32, 51]). However, studies have also con- the original CS+ but are never actually paired with eating (e.g., sistently suggested that a complete extinction of eating desires other people’s living rooms, or watching TV in general). is difficult to achieve: towards the end of extinction, eating Overgeneralization thus can easily lead to a large number of desires to the CS+ often remain higher than to the CS− (e.g., reactivity-evoking stimuli, promoting frequent cravings and [30, 31, 51]—even when experimentally eliminating US ex- overeating [34� ]. Further, reduced learning about the predictive pectancies through explicit instructions after acquisition [52]. value of a CS promotes the formation of context–US associa- This difficulty to completely extinguish eating desires might tions [48]. This can lead to contextual, rather than strong cue- partly explain why it is so difficult to stick to one’sdiet. Back specific, reactivity (e.g., to Carl’s living room). to the example of Carl: watching his crisps-associated favorite Another study suggesting altered appetitive learning in TV show in the evening while no longer eating crisps will take obese individuals found an opposite pattern: successful acqui- some time for desires for crisps to reduce, and as long as the sition of a swallowing response to a CS+ vs. CS− in over- cued desires are still present, not eating the crisps will be weight but (unexpectedly) not in normal-weight individuals difficult. If extinction of appetitive responses (such as eating was found. One may argue that findings reporting reduced desires) is slow, this could diminish one’s chances to diet learning could have been plausibly caused by overweight/ successfully [53]. In addition, it has been argued that partial obese individuals using implicit or explicit strategies that can (in contrast to continuous) reinforcement of food cues might lead to reduced learning about CS–US contingencies, like be frequently practiced by a significant portion of unsuccess- being reluctant to report appetitive desires [49], having ful dieters who try to abstain from eating in response to certain mindsets that suppress appetitive responding [50], or suffering food cues but regularly fail [32]. Partial reinforcement is well from a greater cognitive load due to thoughts about eating, known to result in slowed extinction of appetitive responses dieting, and weight [44� ]. Thus, moreresearchisneededto (the partial reinforcement extinction effect)—possibly leading examine differences in the acquisition of appetitive responses to a greater difficulty to extinguish appetitive responding and in overweight/obese versus normal-weight individuals, taking hence, more risk of lapses during one’s dieting attempts. into account these potential explanations, while preferably Why are cue-elicited desires seemingly so difficult to ex- assessing sensitive physiological indices of appetitive learning tinguish? It has been proposed that the difficulty to extinguish in addition to self-report measures. eating desires could indicate the presence of separate response To summarize, food cue reactivity has been shown to be systems: possibly, US expectancies stem from a system that related to overeating and weight gain and can partly be learned serves to prepare an organism for the incoming US, whereas through Pavlovian learning principles. Laboratory studies have eating desires might stem from another (hedonic/evaluative) system that is based on the mere activation of the US repre- shown that relatively few CS–US pairings are sufficient to form CS–US associations, and that—consequently—encounters of sentation in memory and that is less sensitive to extinction [30, CSs enhance food cue reactivity. While this model has also 52, 54–56]. The implication is that to reduce acquired eating been established in real-life circumstances, it remains of interest desires more effectively, it may be fruitful to investigate tech- to study whether overweight individuals might be different niques that specifically target this hedonic/evaluative response from normal-weight individuals in their appetitive conditioning system (see e.g., [54]). capacity. Initial evidence suggests that overweight individuals One important characteristic of extinction is that it does not show impairments in the acquisition of appetitive responses, reflect mere “unlearning” of the CS–US relationship; it largely forming an interesting direction for future investigations. reflects new learning of a second—contextually controlled—in- hibitory association (CS–noUS) that competes with the original CS–US association [57]. So, building upon the example of Carl: Extinction of Appetitive Responding practicing extinction of not eating crisps while watching the TV show leads to a novel association: namely that the TV show is no Given that food cue reactivity is partly learned, it may be longer a predictor of crisps intake (CS–noUS), while the original reduced through extinction, which refers to the decline in association between the TV show and crisps intake also remains (appetitive) responding with repeated CS–noUS pairings after intact (CS–US). Consistent with this account, a large amount of acquisition [13]. Extinction of appetitive responses in real life (mostly animal and some human) data show that after extinction can be achieved during a diet: when CSs, such as a chocolate- procedures, seemingly extinguished appetitive responses can re- associated time of day, or watching one’s crisp-associated fa- turn under certain conditions—as demonstrated by phenomena vorite TV show in the evening, are deliberately no longer such as rapid reacquisition (the rapid return of responding with followed by food intake (noUS), food cue reactivity (e.g., new CS–US pairings after extinction, e.g., when eating crisps eating desires) decreases over time. Several laboratory studies while watching the TV show once after a diet), reinstatement have indeed shown that conditioned responses (such as eating (the return of responding when the US is provided after 226 Curr Addict Rep (2018) 5:223–231 extinction – e.g., eating crisps again after a diet), renewal (the are thus exposed to intake-predicting cues (CSs) such as see- return of responding after extinction with a change in context, ing, smelling, and tasting the food, while for example being in e.g., extinction of evening cravings during a holiday and an overeating-associated context, and/or while experiencing a returning home), and spontaneous recovery (the spontaneous specific emotion, though food intake is not permitted (noUS) return of responding when time has elapsed after extinction, and food is thrown away at the end of the session. In line with i.e., after a period of successful dieting) [13, 52, 56]. Extinction a learning-based interpretation of food cue reactivity and over- is dependent on the context for expression, and therefore, a eating, the (small-scale) studies that have been conducted on change in context may promote a return of conditioned responses CET suggest it leads to substantial reductions in cue-elicited [57, 58]. Various types of (internal and external) stimuli can food cravings and binge eating in bulimic patients [63–68]. In provide such contexts [59, 60� ]. For example, one might learn overweight and obese samples, CET has also found to signif- that recent CS–US pairings are part of the acquisition “context,” icantly reduce cue-elicited cravings, binge eating, eating in the and recent CS–alone presentations are part of the extinction con- absence of hunger, and improved (maintenance of) weight text. When providing renewed CS–US pairings, this returns an loss [69–75]. Studies examining its long-term efficacy are individual to the acquisition context, resulting in a rapid return of however sparse with mixed findings [63, 70, 75], suggesting responding [58]. For example, Carl previously repeatedly con- that exposure protocols may require optimization. A number sumed crisps in the evening may have (largely) extinguished his of techniques derived from learning theory may be suitable for or her evening-crisp-cravings by refraining from eating crisps for this (for recent overviews, see [14]). Here, we focus on tech- awhile(CS–alone presentations, extinction). However, after niques that have received (recent) empirical investigation. months of strictly sticking to the crisps-less diet, Carl has some friends over and consumes crisps again in the crisp-associated Foods to Be Included in Cue Exposure Sessions Aconsistent context (reinforcing the original CS–US association). This return finding in CET studies is that after one to seven exposure to this original acquisition context may elicit a renewed craving sessions to a certain high-calorie food, intake of that specific for crisps in the evenings over the next days and risks the lapse food during an ad libitum intake test is decreased, compared turning into a full-blown relapse. The ease of conditioned re- with ad libitum intake after a control lifestyle intervention [72, sponses to re-emerge after extinction might explain the observa- 73, 75]. Several prior studies have additionally examined tion that although a considerable proportion of dieters are able to whether exposure effects generalize to (often personalized) achieve initial weight loss, only few are able to also successfully high-calorie food items not explicitly included during expo- maintain their weight loss—most dieters regain the lost weight sure sessions. These studies found that, generally, exposure (or even more) (e.g., [61, 62]). effects do not transfer to non-exposed foods [72, 73, 75]. Thus, insights into mechanisms of extinction can explain why The implication for CET is straightforward: to achieve clini- cally relevant effects on overeating and weight loss, the ther- diets are often unsuccessful: first, dieters likely experience per- sistent eating desires even after having successfully refrained apist should include all relevant food cues in the exposure from intake for a while due to not reaching (full) extinction sessions, i.e., those that are identified by the subject as cues (which may be exacerbated by partial reinforcement extinction for overeating. effects). Such prolonged eating desires could be (too) hard to resist over a longer period of time—promoting lapses in the diet Occasional Reinforced Extinction It has been argued that oc- that, in their turn, again strengthen the CS–US association. casionally reinforcing the CS–US relationship during expo- Second, when dieters finally achieve (partial) extinction of food sure therapy may lead to a reduced risk of relapse of binge cue reactivity, responding can easily return, thereby promoting and overeating [76]. Occasional reinforcements might reduce (re)lapse. Therefore, the long-term effectiveness of dieting efforts the risk of relapse is by allowing a reinforced trial (CS–US and treatments might be considerably improved by focusing on pairing) to become associated with extinction (CS–alone pre- strengthening extinction learning and diminishing the magnitude sentations), leading to a slower rate of reacquisition with of returns of appetitive responses. renewed CS–US pairings (e.g., when one experiences a lapse in one’sdiet) [76]. In CET, one way to implement this tech- nique is by encouraging individuals to occasionally consume Cue Exposure Therapy: Effects, Working small amounts of food during cue exposure exercises, in order Mechanisms, and its Optimization to learn that eating a small amount of food is no longer a cue for overeating [13]. In support, animal and human appetitive Cue exposure therapy (CET) is the direct clinical analogue of conditioning [51, 76], as well as fear conditioning studies [77, experimental extinction. In CET, individuals who suffer from 78], has reported evidence for occasional reinforcements dur- overeating or binge eating are repeatedly and long-lasting ex- ing extinction to reduce relapse of responding (i.e., rapid re- posed to palatable food cues (CSs), while refraining from acquisition). In studies on CET, occasional reinforced extinc- tion has not yet been studied in isolation, but some studies intake (i.e., CS–noUS). During CET sessions, participants Curr Addict Rep (2018) 5:223–231 227 have incorporated this technique in exposure sessions by let- such as high heart rate and sweating, which maximize the expec- ting participants take small bites of high-calorie food previ- tancy of the US to take place—the heart attack. The larger the ously associated with overeating [74� , 75]. These studies pro- mismatch between perceived expectancy of the heart attack to vided indirect support for the effectiveness of this technique occur and the actual absence of such a heart attack (noUS), the by showing that specific overeating expectancies—specified stronger expectancies are violated and hence the better strength- for taking one bite of the food—significantly reduced after ening of the CS–noUS association. Some experimental studies CET vs. control. Nevertheless, examining the pure benefits have indeed provided evidence for the benefits of focusing ex- of adding occasional reinforcements to CET in (sub)clinical posure sessions on expectancy violation, rather than habituation: samples awaits investigation. panic disorder patients who received two sessions of exposure therapy aimed at the violation of danger expectancies reported Working Mechanisms of CET Gaining insight into the possible fewer panic attacks, less anxiety, and less avoidance of fearful working mechanisms of CET could greatly help to optimize cue situations after therapy, compared to patients who received two exposure protocols. It has been proposed that the reduction in sessions of habituation-focused exposure ( [84]; see also [85]). cue-elicited eating desires could be a primary driver underlying It is likely that violation of CS–US expectancies also play a CET’s effectiveness, which is why most prior CET studies have significant role in CET. For example, a dieter may verbalize focused on achieving a reduction (or “habituation”)of eating the expectancy “If I feel exhausted and chocolate is available desires during exposure sessions. That is, exposure sessions are (CS), then I will lose control and eat the entire chocolate box usually terminated when cue-elicited cravings (e.g., [79], arousal (US).” To maximize violation of overeating expectancies dur- [71], or anxiety levels (e.g., [65])) have reached a pre-determined ing CET, a patient is exposed to the box of chocolate when threshold (e.g., 10 on a 100-point scale). Likewise, a large feeling exhausted, while exposure is continued until “losing amount of fear studies aimed to achieve habituation of fear within control” is no longer expected (for example by asking how exposure sessions (within-session habituation or WSH) and be- much time it will take to lose control and continuing exposure tween exposure sessions (between-session habituation or BSH). slightly beyond this point). To answer the question whether It has for long been argued that treatment success was dependent expectancy violation or habituation determines the effective- on successful WSH and BSH [80, 81]. However, research point- ness of CET, it was studied whether the reduction of CS–US ed out that indices of habituation are not consistently related to expectancies (used as index of the violation of CS–US expec- better treatment outcome in anxiety disorders [82]. Instead of tancies) or WSH and BSH of cue-elicited eating desires were focusing on habituation of fear, it has been proposed that expo- associated with the intake of food after exposure [72, 73, 74� , sure sessions should be designed in a way that maximizes the 75]. Data of four experimental studies, similar in design, were violation of CS–US expectancies, since this should effectively aggregated to overcome power issues (see Table 1). It was strengthen the inhibitory CS–noUS association [83]. Specifically, found that greater expectancy violation, but not habituation, exposure sessions should aim at maximizing the expectancy of was modestly but significantly associated with reduced intake the individual’s catastrophic event to occur. For example, expo- of exposed food after exposure. Thus, focusing on the reduc- sure sessions for panic disorder should carefully introduce CSs, tion of eating desires during CET might not be necessary for a Table 1 Correlations between z-scores of kcal intake of foods included (exposed) and not included (non-exposed) in CET and within-session habituation (WSH) and between-session habituation (BSH) of eating desires, and expectancy violation among participants of four studies [72, 73, 74� , 75] WSH of eating desires BSH of eating desires Expectancy violation n =140 n =94 n =150 z intake exposed food − 0.05 − 0.08 − 0.18* z intake non-exposed food 0.02 − 0.06 − 0.12 In all studies [72, 73, 74� , 75], WSH was calculated subtracting the end eating desire from the session’s peak eating desire (measured on a 100 mm Visual Analogue Scale [VAS]), and, in case of multiple exposure sessions, the average was calculated. BSH was calculated for three studies (which included 2 or more exposure sessions) by subtracting peak desires in the last sessions from peak desires in the first session. Expectancy violation was calculated by subtracting post-intervention expectancies from pre-intervention expectancies (measured on 100 mm VAS); in studies where expectancies about eating specific foods were asked, scores were averaged. As expectancies were assessed using a 5-point scale in [73], scores were transformed (1 = 0; 2 = 25; 3 = 50; 4 = 75; 5 = 100) *p < .05; sample sizes of each correlation vary due to measurement differences between studies To compare the intake during the taste test in all studies, z-scores were calculated. Some differences between studies are worth mentioning: general exposed foods were included in the taste tests described in [72, 73], while foods in [74� , 75]were personally selected.In [75], one general exposed food item was additionally included during the taste test, but no general non-exposed food item was included as comparison and therefore not included in the present investigation. Pure kcals were converted into z-scores for the data of [72, 74� , 75], while the intake in [73] was converted into a percentage of daily required energy intake for children, and then converted into z-scores 228 Curr Addict Rep (2018) 5:223–231 food cue exposure session to be effective. Instead, the reduc- initially neutral stimuli—both in the laboratory and in real tion of CS–US expectancies might be an important treatment life—can easily elicit conditioned appetitive responding after a target, like in anxiety disorders. few stimulus-food pairings. Extinction of some appetitive re- However, to critically evaluate whether expectancy violation sponses (e.g., desires) seems more difficult to achieve, and even and/or habituation are causally related to treatment success, it is after successful extinction, appetitive responding remains prone necessary to manipulate them. In a recent study, two exposure to relapse—perhaps explaining the general difficulty to achieve conditions were compared to a no-treatment control condition successful long-term weight loss. Cue exposure therapy, the clin- [74� ]. Participants in the exposure conditions either received ex- ical equivalent of extinction, seems effective in lowering cue- posure explicitly aimed at the habituation of eating desires (ex- elicited cravings, overeating, and binge eating, but still little is posure habituation condition) or exposure explicitly aimed at the known about the most effective exposure strategies and its work- violation of CS–US expectancies (exposure expectancy violation ing mechanisms. Reductions in overeating expectancies seem to condition). In all conditions, food intake was measured, while be important. Future research may aim to elucidate inter- expectancy violation and habituation was also measured in both individual differences in Pavlovian learning (e.g., as a function exposure conditions. Results indicated that although expectancy of weight status) and their consequences for eating behavior, to violation was stronger in the exposure expectancy violation vs. study how successful long-term extinction of appetitive exposure habituation condition during exposure sessions, the re- responding is best achieved, as well as to examine CET’s work- duction from pre- to post-therapy was equally strong in both ing mechanisms. conditions. Moreover, consumption during an ad libitum intake Acknowledgements This work is supported by the Netherlands test was similarly reduced across the two exposure conditions Organization for Scientific Research (NWO): Vici Grant 453.10.006, relative to the control condition. These results might question awarded to Anita Jansen. the importance of explicitly manipulating the violation of expec- tancies during exposure (see also [52]). They indicate that al- Compliance with Ethical Standards though greater reductions of CS–US expectancies are related to better outcome, it might not be essential to explicitly tackle these Conflict of Interest The authors declare that they have no conflict of CS–US expectancies during cue exposure to reduce them suc- interest. cessfully, since focusing on the reduction of eating desires led to Human and Animal Rights and Informed Consent All reported studies/ similarly reduced expectancies and food intake after the interven- experiments with human or animal subjects performed by the authors tion. Interestingly, similar findings have also been documented in have been previously published and complied with all applicable ethical the anxiety domain. For example, patients who completed a standards. purely behavioral treatment for animal phobias profited as much Open Access This article is distributed under the terms of the Creative from the treatment as patients who received a cognitive- Commons Attribution 4.0 International License (http:// behavioral treatment, in which dysfunctional cognitions were creativecommons.org/licenses/by/4.0/), which permits unrestricted use, challenged by behavioral tests [86, 87]. Moreover, the (similar) distribution, and reproduction in any medium, provided you give appro- priate credit to the original author(s) and the source, provide a link to the change in maladaptive cognitions in both treatment conditions Creative Commons license, and indicate if changes were made. was found to mediate the improvement [87], which generally fits with the findings in CET: changing overeating expectancies is important, while explicitly challenging these expectancies does References not seem to be crucial. To recapitulate, studies indicate that CET is a promising intervention to target cue-elicited cravings, overeating, binge Papers of particular interest, published recently, have been eating, and weight, although long-term effects are not yet highlighted as: studied. CET may be further optimized by incorporating in- � Of importance sights from learning theory. The working mechanism of CET is still unknown, though initial evidence suggests that the vi- 1. NCD Risk Factor Collaboration (NCD-RisC). Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled olation of US expectancies is necessary. It is clear that more analysis of 1698 population-based measurement studies with 19·2 work in this area is needed. million participants. Lancet. 2016;387(10026):1377–96. https:// doi.org/10.1016/S0140-6736(16)30054-X. 2. Ng M, Fleming T, Robinson M, Thomson B, Graetz N, Margono C, et al. Global, regional, and national prevalence of overweight and obe- Conclusion sity in children and adults during 1980-2013: a systematic analysis for the global burden of disease study 2013. Lancet. 2014;384(9945):766– Pavlovian learning theory provides a theoretical framework of 81. https://doi.org/10.1016/S0140-6736(14)60460-8. how cravings and overeating can be learned and how they might 3. GuhDP, ZhangW, BansbackN, AmarsiZ, BirminghamCL, Anis AH. The incidence of co-morbidities related to obesity and be effectively tackled. Experimental studies indeed show that Curr Addict Rep (2018) 5:223–231 229 overweight: a systematic review and meta-analysis. BMC Public 23. Staiger P, Dawe S, McCarthy R. Responsivity to food cues in bu- limic women and controls. Appetite. 2000;35(1):27–33. https://doi. Health. 2009;9(88) https://doi.org/10.1186/1471-2458-9-88. org/10.1006/appe.2000.0327. 4. Luppino FS, de Wit LM, Bouvy PF, Stijnen T, Cuijpers P, Penninx BWJH, et al. Overweight, obesity, and depression: a systematic 24. van den Akker K, Stewart K, Antoniou EE, Palmberg A, Jansen A. review and meta-analysis of longitudinal studies. Arch Gen Food cue reactivity, obesity, and impulsivity: are they associated? Psychiatry. 2010;67(3):220–9. https://doi.org/10.1001/ Curr Addict Reports. 2014;1(4):301–8. https://doi.org/10.1007/ archgenpsychiatry.2010.2. s40429-014-0038-3. 5. Pulgarón ER. Childhood obesity: a review of increased risk for 25. Carnell S, Haworth CMA, Plomin R, Wardle J. Genetic influence physical and psychological comorbidities. Clin Ther. 2013;35(1): on appetite in children. Int J Obes. 2008;32(10):1468–73. https:// A18–32. https://doi.org/10.1016/j.clinthera.2012.12.014. doi.org/10.1038/ijo.2008.127. 6. Scott KM, Bruffaerts R, Simon GE, Alonso J, Angermeyer M, De 26. Pavlov IP. Conditioned reflexes: an investigation of the physiolog- Girolamo G, et al. Obesity and mental disorders in the general ical activity of the cerebral cortex. London: Oxford University population: results from the world mental health surveys. Int J Press; 1927. Obes. 2008;32(1):192–200. https://doi.org/10.1038/sj.ijo.0803701. 27. Davidson TL, Kanoski SE, Tracy AL, Walls EK, Clegg D, Benoit 7. Cawley J, Meyerhoefer C. The medical care costs of obesity: an SC. The interoceptive cue properties of ghrelin generalize to cues instrumental variables approach. J Health Econ. 2012;31(1):219– produced by food deprivation. Peptides. 2005;26(9):1602–10. 30. https://doi.org/10.1016/j.jhealeco.2011.10.003. https://doi.org/10.1016/j.peptides.2005.02.014. 8. Leung HT, Reeks LM, Westbrook RF. Two ways to deepen extinc- 28. Vohs KD, Wang Y, Gino F, Norton MI. Rituals enhance consump- tion and the difference between them. J Exp Psychol Anim Behav tion. Psychol Sci. 2013;24(9):1714–21. https://doi.org/10.1177/ Process. 2012;38(4):394–406. https://doi.org/10.1037/a0030201. 0956797613478949. 9. Goran MI, Weinsier RL. Role of environmental vs. metabolic fac- 29. Wardle J. Conditioning processes and cue exposure in the modifi- tors in the etiology of obesity: time to focus on the environment. cation of excessive eating. Addict Behav. 1990;15(4):387–93. Obes Res. 2000;8(5):407–9. https://doi.org/10.1038/oby.2000.50. https://doi.org/10.1016/0306-4603(90)90047-2. 10. Swinburn B, Egger G, Raza F. Dissecting obesogenic environ- 30. Van Gucht D, Vansteenwegen D, Van den Bergh O, Beckers T. ments: the development and application of a framework for identi- Conditioned craving cues elicit an automatic approach tendency. fying and prioritizing environmental interventions for obesity. Prev Behav Res Ther. 2008;46(10):1160–9. https://doi.org/10.1016/j. Med. 1999;29:563–70. https://doi.org/10.1006/pmed.1999.0585. appet.2007.05.003. 11.� Boswell RG, Kober H. Food cue reactivity and craving predict 31. Papachristou H, Nederkoorn C, Havermans R, Bongers P, Beunen eating and weight gain: a meta-analytic review. Obes Rev. S, Jansen A. Higher levels of trait impulsiveness and a less effective 2016;17:159–77. https://doi.org/10.1111/obr.12354. Meta- response inhibition are linked to more intense cue-elicited craving analysis showing the importance of cue-elicited cravings for for alcohol in alcohol-dependent patients. Psychopharmacology. eating behaviour and weight gain. 2013;228(4):641–9. https://doi.org/10.1007/s00213-013-3063-3. 12. Boggiano MM, Dorsey JR, Thomas JM, Murdaugh DL. The 32. van den Akker K, Havermans RC, Bouton ME, Jansen A. How partial Pavlovian power of palatable food: lessons for weight-loss adher- reinforcement of food cues affects the extinction and reacquisition of ence from a new rodent model of cue-induced overeating. Int J appetitive responses. A new model for dieting success? Appetite. Obes. 2009;33(6):693–701. https://doi.org/10.1038/ijo.2009.57. 2014;81:242–52. https://doi.org/10.1016/j.appet.2014.06.024. 13. Bouton ME. Learning and the persistence of appetite: extinction 33. Andreatta M, Pauli P. Appetitive vs. aversive conditioning in and the motivation to eat and overeat. Physiol Behav. humans. Front Behav Neurosci. 2015;9:1–8. https://doi.org/10. 2011;103(1):51–8. https://doi.org/10.1016/j.physbeh.2010.11.025. 3389/fnbeh.2015.00128. 14. Jansen A, Schyns G, Bongers P, van den Akker K. From lab to clinic: 34.� van den Akker K, Schyns G, Jansen A. Altered appetitive condi- extinction of cued cravings to reduce overeating. Physiol Behav. tioning in overweight and obese women. Behav Res Ther. 2017;99: 2016;162:174–80. https://doi.org/10.1016/j.physbeh.2016.03.018. 78–88. https://doi.org/10.1016/j.brat.2017.09.006. Study 15. Woods SC. The eating paradox: how we tolerate food. Psychol Rev. suggesting appetitive conditioning impairments in overweight 1991;98(4):488–505. https://doi.org/10.1037/0033-295X.98.4.488. vs. normal-weight individuals. 16. Bouton ME. Learning and behavior: a contemporary synthesis. 35. Blechert J, Testa G, Georgii C, Klimesch W, Wilhelm FH. The Sunderland: Sinauer Associates, Inc.; 2007. Pavlovian craver: neural and experiential correlates of single trial 17. Power ML, Schulkin J. Anticipatory physiological regulation in naturalistic food conditioning in humans. Physiol Behav. 2016;158: feeding biology: cephalic phase responses. Appetite. 2008;50(2– 18–25. https://doi.org/10.1016/j.physbeh.2016.02.028. 3):194–206. https://doi.org/10.1016/j.appet.2007.10.006. 36. Bongers P, Jansen A. Emotional eating and Pavlovian learning: 18. Woods SC, Ramsay DS. Pavlovian influences over food and drug evidence for conditioned appetitive responding to negative emo- intake. Behav Brain Res. 2000;110(1–2):175–82. https://doi.org/ tional states. Cogn Emot. 2017;31(2):284–97. https://doi.org/10. 10.1016/S0166-4328(99)00194-1. 1016/j.appet.2015.02.018. 19. Ferriday D, Brunstrom JM. “Ijustcan”thelp myself’: effects of 37. Birch LL, McPhee L, Sullivan S, Johnson S. Conditioned meal food-cue exposure in overweight and lean individuals. Int J Obes. initiation in young children. Appetite. 1989;13(2):105–13. https:// 2011;35(1):142–9. https://doi.org/10.1038/ijo.2010.117. doi.org/10.1016/0195-6663(89)90108-6. 20. Ferriday D, Brunstrom JM. How does food-cue exposure lead to 38. Astur RS, Carew AW, Deaton BE. Conditioned place preferences in larger meal sizes? Br J Nutr. 2008;100(6):1325–32. https://doi.org/ humans using virtual reality. Behav Brain Res. 2014;267:173–7. 10.1017/S0007114508978296. https://doi.org/10.1016/j.bbr.2014.03.018. 21. Jansen A. A learning model of binge eating: cue reactivity and cue 39. van den Akker K, Jansen A, Frentz F, Havermans RC. Impulsivity exposure. Behav Res Ther. 1998;36(3):257–72. https://doi.org/10. makes more susceptible to overeating after contextual appetitive 1016/S0005-7967(98)00055-2. conditioning. Appetite. 2013;70:73–80. https://doi.org/10.1016/j. appet.2013.06.092. 22. Jansen A, Theunissen N, Slechten K, Nederkoorn C, Boon B, Mulkens S, et al. Overweight children overeat after exposure to 40. Meyer MD, Risbrough VB, Liang J, Boutelle KN. Pavlovian condi- food cues. Eat Behav. 2003;4(2):197–209. https://doi.org/10. tioning to hedonic food cues in overweight and lean individuals. 1016/S1471-0153(03)00011-4. Appetite. 2015;87:56–61. https://doi.org/10.1016/j.appet.2014.12.002. 230 Curr Addict Rep (2018) 5:223–231 41. Bongers P, van den Akker K, Havermans R, Jansen A. Emotional to eat chocolate. Behav Res Ther. 2008;46(3):375–91. https://doi. org/10.1016/j.brat.2008.01.003. eating and Pavlovian learning: does negative mood facilitate appe- titive conditioning? Appetite. 2015;89:226–36. https://doi.org/10. 57. Bouton ME. Context, ambiguity , and unlearning : sources of relapse 1016/j.appet.2015.02.018. after behavioral extinction four mechanisms of relapse. 2002;52: 976– 986. doi:https://doi.org/10.1016/S0006-3223(02)01546-9. 42.� Akker K, Van Den HRC, Jansen A. Appetitive conditioning to specific times of day. Appetite. 2017;116:232–8. https://doi.org/ 58. Bouton ME. Context and behavioral processes in extinction. Learn Mem. 2004;11(5):485–94. https://doi.org/10.1101/lm.78804. 10.1016/j.appet.2017.05.014. Experiment showing that findings of conditioned food cue reactivity in laboratory settings 59. Bouton ME. Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychol Bull. 1993;114(1):80– translate to real-life circumstances. 99. https://doi.org/10.1037/0033-2909.114.1.80. 43. Coppin G, Nolan-Poupart S, Jones-Gotman M, Small DM. 60.� Schepers ST, Bouton ME. Hunger as a context: food seeking that is Working memory and reward association learning impairments in inhibited during hunger can renew in the context of satiety. Psychol obesity. Neuropsychologia. 2014;65:146–55. https://doi.org/10. Sci. 2017;28(11):1640 –8. https://doi.org/10.1177/ 1016/j.neuropsychologia.2014.10.004. 0956797617719084. Two experiments in rodents 44.� Zhang Z, Manson KF, Schiller D, Levy I. Impaired associative demonstrating that satiety can provide a “context” and lead learning with food rewards in obese women. Curr Biol. to renewal of (operant) responding. 2014;24(15):1731–6. https://doi.org/10.1016/j.cub.2014.05.075. 61. Mann T, Tomiyama a J, Westling E, Lew A-M, Samuels B, Study indicating differences in associative learning in obese Chatman J. Medicare’s search for effective obesity treatments: diets vs. normal-weight participants, only when food (not money) are not the answer. Am Psychol. 2007;62(3):220–33. https://doi. was used as a reward. org/10.1037/0003-066X.62.3.220. 45. Davidson TL, Martin AA. Obesity: cognitive impairment and the 62. Wing RR, Phelan S. Long-term weight loss maintenance. Am J Clin failure to “eat right”. Curr Biol. 2014;24(15):R685–7. https://doi. Nutr. 2005;82(1):222S–5S. Doi: 82/1/222S [pii] org/10.1016/j.cub.2014.06.031. 63. Jansen A, Broekmate J, Heymans M. Cue-exposure vs self-control in 46. Lissek S, Powers AS, McClure EB, Phelps EA, Woldehawariat G, the treatment of binge eating: a pilot study. Behav Res Ther. Grillon C, et al. Classical fear conditioning in the anxiety disorders: 1992;30(3):235–41. https://doi.org/10.1016/0005-7967(92)90069-S. a meta-analysis. Behav Res Ther. 2005;43(11):1391–424. https:// 64. Jansen A, van den Hout MA, de Loof C, Zandbergen J, Griez E. A doi.org/10.1016/j.brat.2004.10.007. case of bulimia succesfully treated by cue exposure. Behav Ther 47. Hermans D, Baeyens F, Vervliet B. Generalization of acquired emotion- Exp Psychiatry. 1989;20(4):327–32. https://doi.org/10.1016/0005- al responses. In: Handbook of Cognition and Emotion; 2013. p. 117–34 7916(89)90064-5. 48. Grillon C, Davis M. Fear-potentiated startle conditioning in 65. Martinez-mallen E, Castro-fornieles J, La L, Moreno E, Morer A, humans: explicit and contextual cue conditioning following paired Font E, et al. Cue exposure in the treatment of resistant adolescent versus unpaired training. Psychophysiology. 1997;34:451–8. bulimia nervosa. Int J Eat Disord. 2007;40:596–60 1. https://doi.org/ https://doi.org/10.1111/j.1469-8986.1997.tb02389.x. 10.1002/eat20423. 49. Roefs A, Quaedackers L, Werrij MQ, Wolters G, Havermans R, 66. Toro J, Cervera M, Feliu MH, Garriga N, Jou M, Martinez E, et al. Nederkoorn C, et al. The environment influences whether high-fat Cue exposure in the treatment of resistant bulimia nervosa. Int J Eat foods are associated with palatable or with unhealthy. Behav Res Disord. 2003;34(2):227–34. https://doi.org/10.1002/eat.10186. Ther. 2006;44(5):715–36. https://doi.org/10.1016/j.brat.2005.05.007. 67. Rosen JC, Leitenberg H. Bulimia nervosa : treatment with exposure 50. Siep N, Roefs A, Roebroeck A, Havermans R, Bonte M, Jansen A. and response prevention. Behav Ther. 1982;13:117–24. https://doi. Fighting food temptations: the modulating effects of short-term org/10.1016/S0005-7894(82)80055-5. cognitive reappraisal, suppression and up-regulation on 68. Schmidt U, Marks I. Cue exposure to food plus mesocorticolimbic activity related to appetitive motivation. responsepreventionof binges for bulimia: a pilot study. NeuroImage. 2012;60(1):213–20. https://doi.org/10.1016/j. 1988;7(5):663–72. https://doi.org/10.1002/1098-108X(198809)7: neuroimage.2011.12.067. 5<663::AID-EAT2260070510>3.0.CO;2-X. 51. van den Akker K, Havermans RC, Jansen A. Effects of occasional 69. Boutelle KN, Zucker NL, Peterson CB, Rydell SA, Cafri G, reinforced trials during extinction on the reacquisition of condi- Harnack L. Two novel treatments to reduce overeating in over- tioned responses to food cues. J Behav Ther Exp Psychiatry. weight children: a randomized controlled trial. J Consult Clin 2015;48:50–8. https://doi.org/10.1016/j.jbtep.2015.02.001. Psychol. 2011;79(6):759–71. https://doi.org/10.1037/a0025713. 52. van den Akker K, van den Broek M, Havermans RC, Jansen A. 70. Boutelle KN, Zucker N, Peterson CB, Rydell S, Carlson J, Harnack LJ. Violation of eating expectancies does not reduce conditioned de- An intervention based on Schachter’s externality theory for overweight sires for chocolate. Appetite. 2016;100:10–7. https://doi.org/10. children: the regulation of cues pilot. J Pediatr Psychol. 2014 1016/j.appet.2016.02.004. May;39(4):405–17. https://doi.org/10.1093/jpepsy/jst142. 53. Jansen A, Stegerman S, Roefs A, Nederkoorn C, Havermans R. 71. Mount R, Neziroglu F, Taylor CJ. An obsessive-compulsive view Decreased salivation to food cues in formerly obese successful di- of obesity and its treatment. J Clin Psychol. 1990;46(1):68–78. eters. Psychother Psychosom. 2010;79(4):257–8. https://doi.org/ 72. Schyns G, Roefs A, Mulkens S, Jansen A. Expectancy violation, 10.1159/000315131. reduction of food cue reactivity and less eating in the absence of 54. Van Gucht D, Baeyens F, Vansteenwegen D, Hermans D, Beckers hunger after one food cue exposure session for overweight and T. Counterconditioning reduces cue-induced craving and actual obese women. Behav Res Ther. 2016;76:57–64. https://doi.org/ cue-elicited consumption. Emotion. 2010;10(5):688–95. https:// 10.1016/j.brat.2015.11.007. doi.org/10.1037/a0019463. 73. Schyns G, Roefs A, Smulders FTY, Jansen A. Cue exposure ther- 55. Van Gucht D, Vansteenwegen D, Beckers T, Hermans D, Baeyens apy reduces overeating of exposed and non-exposed foods in obese F, Van den Bergh O. Repeated cue exposure effects on subjective adolescents. J Behav Ther Exp Psychiatry. 2018;58:68–77. https:// and physiological indices of chocolate craving. Appetite. doi.org/10.1016/j.jbtep.2017.09.002. 2008;50(1):19–24. https://doi.org/10.1016/j.appet.2007.05.003. 74.� Schyns G, van den Akker K, Roefs A, Hilberath R, Jansen A. What 56. Van Gucht D, Vansteenwegen D, Beckers T, Van den Bergh O. works better ? Food cue exposure aiming at the habituation of eating desires or food cue exposure aiming at the violat ion of Return of experimentally induced chocolate craving after extinction in a different context: divergence between craving for and expecting overeating expectancies? Behav Res Ther. 2018;102:1–7. https:// Curr Addict Rep (2018) 5:223–231 231 doi.org/10.1016/j.brat.2017.12.001. Experiment on the working 82. Craske MG, Kircanski K, Zelikowsky M, Mystkowski J, Chowdhury N, Baker A. Optimizing inhibitory learning during mechanisms of food cue exposure therapy, showing that explicitly tackling overeating expectancies might not be exposure therapy. Behav Res Ther. 2008;46(1):5–27. https://doi. necessary for treatment success. org/10.1016/j.brat.2007.10.003. 75. Schyns G, van den Akker K, Roefs A, Houben K, Jansen A. Weight 83. Craske MG, Treanor M, Conway CC, Zbozinek T, Vervliet B. loss and binge eating reduction after eight sessions food cue expo- Maximizing exposure therapy: an inhibitory learning approach. sure therapy In Rev. Behav Res Ther. 2014;58:10–23. https://doi.org/10.1016/j.brat. 76. Bouton ME, Woods AM, Pineño O. Occasional reinforced trials during 2014.04.006. extinction can slow the rate of rapid reacquisition. Learn Motiv. 84. Salkovskis PM, Hackmann A, Wells A, Gelder MG, Clark DM. Belief 2004;35(4):371–90. https://doi.org/10.1016/j.lmot.2004.05.001. disconfirmation versus habituation approaches to situational exposure 77. Culver NC, Stevens S, Fanselow MS, Craske MG. Building phys- in panic disorder with agoraphobia: a pilot study. Behav Res Ther. iological toughness: some aversive events during extinction may 2006;45(5):877–85. https://doi.org/10.1016/j.brat.2006.02.008. attenuate return of fear. J Behav Ther Exp Psychiatry. 2018;58: 85. Deacon BJ, Kemp JJ, Dixon LJ, Sy JT, Farrell NR, Zhang AR. 18–28. https://doi.org/10.1016/j.jbtep.2017.07.003. Maximizing the efficacy of interoceptive exposure by optimizing 78. Morís J, Barberia I, Vadillo MA, Andrades A, López FJ. Slower inhibitory learning: a randomized controlled trial. Behav Res Ther. reacquisition after partial extinction in human contingency learning. 2013;51(9):588–96. https://doi.org/10.1016/j.brat.2013.06.006. J Exp Psychol Learn Mem Cogn. 2017;43(1):81–93. https://doi. 86. Koch EI, Spates CR, Himle JA. Comparison of behavioral and org/10.1037/xlm0000282. cognitive-behavioral one-session exposure treatments for small an- 79. Boutelle KN, Knatz S, Carlson J, Bergmann K, Peterson CB. An imal phobias. Behav Res Ther. 2004;42(12):1483–504. https://doi. open trial targeting food cue reactivity and satiety sensitivity in org/10.1016/j.brat.2003.10.005. overweight and obese binge eaters. Cogn Behav Pract. 87. Raes AK, Koster EHW, Loeys T, De Raedt R. Pathways to change 2016;24(3):363–73. https://doi.org/10.1016/j.cbpra.2016.08.003. in one-session exposure with and without cognitive intervention: an 80. Foa EB, Kozak MJ. Emotional processing of fear: exposure to exploratory study in spider phobia. J Anxiety Disord. 2011;25(7): corrective information. Psychol Bull. 1986 Jan;99(1):20–35. 964–71. https://doi.org/10.1016/j.janxdis.2011.06.003. 81. Foa EB, McNally RJ. Mechanisms of change in exposure therapy. In: Rapee M, editor. Current controversies in the anxiety disorders. New York: The Guilford Press; 1996. p. 329–43. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Current Addiction Reports Springer Journals

Learned Overeating: Applying Principles of Pavlovian Conditioning to Explain and Treat Overeating

Free
9 pages
Loading next page...
 
/lp/springer_journal/learned-overeating-applying-principles-of-pavlovian-conditioning-to-0IMqu47sTb
Publisher
Springer International Publishing
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Psychiatry; Neurology
eISSN
2196-2952
D.O.I.
10.1007/s40429-018-0207-x
Publisher site
See Article on Publisher Site

Abstract

Purpose of Review This review provides an overview of recent findings relating to the role of Pavlovian conditioning in food cue reactivity, including its application to overeating and weight loss interventions. Recent Findings Both in the laboratory and in real life, cue-elicited appetitive reactivity (e.g., eating desires) can be easily learned, but (long-term) extinction is more difficult. New findings suggest impaired appetitive learning in obesity, which might be causally related to overeating. The clinical analogue of extinction—cue exposure therapy—effectively reduces cue-elicited cravings and overeating. While its working mechanisms are still unclear, some studies suggest that reducing overeating expec- tancies is important. Summary Pavlovian learning theory provides a still undervalued theoretical framework of how cravings and overeating can be learned and how they might be effectively tackled. Future studies should aim to elucidate inter-individual differences in Pavlovian conditioning, study ways to strengthen (long-term) extinction, and investigate the working mechanisms of cue exposure therapy. . . . . Keywords Pavlovian conditioning Food cue reactivity Overeating Obesity Food cue exposure therapy Introduction halting, and reversing the obesity “pandemic” has become a priority in public health. Overweight and obesity prevalence has reached dramatic pro- It has been proposed that the “obesogenic” environment is a portions. It is estimated that currently, around 2.1 billion indi- main contributor to weight gain and obesity, for it promotes a viduals worldwide are overweight (body mass index [BMI] > sedentary lifestyle and provides an abundance of easy-to-get 2 2 25 kg/m ), of which 600 million are obese (BMI > 30 kg/m ) high-calorie foods (e.g., [9, 10, 11� ]. Specifically, it is assumed [1, 2]. Given that obesity is associated with an increased risk that food-associated cues play a major role in overconsumption for diabetes, cardiovascular disease, psychological problems by evoking appetitive responses (such as food cravings), moti- and various mental illnesses [3–6], and societies face explod- vating one to (over)eat. Human Pavlovian learning in such “food ing healthcare costs due to obesity-associated morbidity [7, 8], cue reactivity” has however received little attention in both sci- entific studies and obesity interventions, even though experts have emphasized its importance in overeating and its contribu- tion to the difficulty in achieving long-term weight loss [12–15]. This article is part of the Topical Collection on Food Addiction This review provides an overview of recent findings on Karolien van den Akker and Ghislaine Schyns equally contributed to the Pavlovian learning in food cue reactivity and food intake, includ- manuscript. ing its application to weight loss interventions. * Karolien van den Akker karolien.vandenakker@maastrichtuniversity.nl * Ghislaine Schyns Food Cue Reactivity and Pavlovian ghislaine.schyns@maastrichtuniversity.nl Conditioning Faculty of Psychology and Neuroscience, Department of Clinical The obesogenic environment includes numerous food cues, Psychological Science, Maastricht University, P.O. Box 616, 6200, MD Maastricht, The Netherlands such as the smell of freshly baked bread when walking past 224 Curr Addict Rep (2018) 5:223–231 a bakery or the sight of delicious Italian ice cream when hiking Though these findings seem in line with a role for though the park. Exposure to food cues activates a central Pavlovian conditioning in food cue reactivity, a relevant ques- appetitive state [16], which consists of diverse responses: psy- tion is whether the findings are ecologically valid. The labo- chological (craving, urge, or desire to eat), physiological pre- ratory experiments are generally conducted under highly con- paratory (e.g., increased salivation and insulin release), and trolled conditions with very few distractors present and using neurocognitive (e.g., brain activation patterns, allocation of arbitrary stimuli as CSs. In contrast, in daily life, CSs concern attentional resources). The physiological preparatory re- non-arbitrary naturalistic stimuli, and the CS–US associations sponses (cephalic phase responses) aid in digestion, absorp- will, for example, be more spread over time and embedded tion, and metabolism [17, 18], and overall, stronger cue- within a more distracting environment. In a recent study, we elicited appetitive responding is thought to promote therefore examined whether appetitive conditioning also oc- (over)eating [11� ]. Indeed, increased levels of food cue reac- curs under real-life circumstances, using times of day as con- tivity (e.g., cue-elicited desires to eat) have been associated ditioned stimuli [42� ]. Participants were tested over a period of with overeating, unsuccessful dieting, a higher BMI, and eat- 15 days in their natural environment, using a smartphone ap- ing psychopathology [11� , 19–24]. plication for administering a questionnaire (eating expectan- Although food cue reactivity has a strong genetic component cies, eating desires) and for instructing them to consume choc- [25], Pavlovian conditioning plays an important role as well. olate provided by the experimenter. At one specific time of Every time palatable food is consumed, the food can be asso- day (CS+), participants always received a signal to complete ciated with cues in the internal or external environment, and the questionnaire and the instruction to consume chocolate, these cues then promote cue reactivity upon their next encoun- whereas on another time of day (CS−), participants only com- ter. In Pavlovian terms, food entering the digestive system is an pleted the questionnaire. Results indicated considerably unconditioned stimulus (US), and once a cue (conditioned stim- heightened eating desires and eating expectancies to the ulus; CS) has become associated with the US, it can stimulate chocolate-associated time of day (vs. CS−) from the fifth appetitive responses, i.e., food cue reactivity (conditioned re- day forward [42� ]. Thus, consistent with the findings of labo- sponse: CR), which promotes food intake [13, 21, 26]. For ratory studies, this study shows that cue-elicited eating desires example, consider Carl who repeatedly eats crisps in the eve- can be learned with relative ease in the natural environment. ning when watching his favorite TV show: this context (CS; Associative learning processes might therefore play a critical watching his favorite TV show in the evening) may become role in the experience of many daily eating desires. associated with eating crisps (US) and if this is the case, just the Given the potentially important role of Pavlovian learning in CS, that is, just the look of Carl’s favorite TV show, will elicit a the onset and maintenance of obesity, several recent studies desire for the crisps and promote its intake. Theoretically, any have examined whether overweight individuals differ in appe- titive Pavlovian learning from normal-weight individuals. cue can become associated with palatable food intake, includ- ing the sight and smell of food, a hormonal or other internal Interestingly, the findings point towards impairments in acquir- physical state like hunger or satiety, specific rituals, cognitions, ing appetitive responses (US expectancies, US desires, CS lik- emotions, and so on [21, 27–29]. Several laboratory studies ing/preference) to food-associated stimuli in overweight and have examined this Pavlovian learning account of food cue obese individuals [34� , 43, 44� ]. In two studies, this was reactivity by repeatedly pairing an initially neutral stimulus reflected by reduced differential responding (i.e., smaller differ- (CS+) with tasty foods (such as a piece of chocolate or a sip ences in responding between the CS+ and CS−)[34� , 44� ]. of milkshake), and comparing the appetitive responses to this These findings are intriguing and raise the question if and stimulus with the appetitive responses to another stimulus (CS how such impairments are causally linked to overeating. It −) which is never followed by food intake. Stimuli that have might be that heightened CS− responding (which drove the been examined in these studies include trays [30, 31], boxes deficit reported in [44� ] reflects an impaired ability to inhibit [32], geometric figures [33, 34� ], objects [35], negative emo- appetitive responses to stimuli signaling non-reward in an ap- tions [36], and (virtual) environments/contexts [37–39]. A con- petitive context, which, in the natural environment, might plau- sistent finding is that after conditioning, the CS+ (vs. CS−) sibly increase reactivity and overeating to stimuli not associated evokes conditioned responding, as reflected by heightened psy- with food intake [45, 46]. The findings might also indicate a chological (e.g., explicit eating expectancies, eating desires, tendency towards overgeneralization in overweight and obese CS+ liking) (e.g., [30, 38, 39], psychophysiological (e.g., sali- individuals [34� , 44� ]: a reduced ability to discriminate between vation) [34� , 35, 40], and behavioral responding (e.g., approach predictive and non-predictive stimuli is a core feature of over- tendencies, place preferences, food intake/choice) (e.g., [30, 36, generalization, and heightened responding to the CS− could 37, 41]). Findings also suggest that appetitive conditioning usu- result from overgeneralization from the CS+ to the CS− [47]. ally occurs quickly; in most studies, only 3 to 6 CS–US pairings In the example of Carl, overgeneralization of appetitive were sufficient, while one study even reports evidence for suc- responding might imply that while he originally learns to asso- cessful acquisition after only one CS–US pairing [35]. ciate the eating of crisps to a particular context (e.g., sitting on Curr Addict Rep (2018) 5:223–231 225 the couch in the evening and watch his favorite TV show), he is expectancies, eating desires) reduce with non-reinforced CS also cue reactive to a wider variety of stimuli that are similar to presentations (e.g., [32, 51]). However, studies have also con- the original CS+ but are never actually paired with eating (e.g., sistently suggested that a complete extinction of eating desires other people’s living rooms, or watching TV in general). is difficult to achieve: towards the end of extinction, eating Overgeneralization thus can easily lead to a large number of desires to the CS+ often remain higher than to the CS− (e.g., reactivity-evoking stimuli, promoting frequent cravings and [30, 31, 51]—even when experimentally eliminating US ex- overeating [34� ]. Further, reduced learning about the predictive pectancies through explicit instructions after acquisition [52]. value of a CS promotes the formation of context–US associa- This difficulty to completely extinguish eating desires might tions [48]. This can lead to contextual, rather than strong cue- partly explain why it is so difficult to stick to one’sdiet. Back specific, reactivity (e.g., to Carl’s living room). to the example of Carl: watching his crisps-associated favorite Another study suggesting altered appetitive learning in TV show in the evening while no longer eating crisps will take obese individuals found an opposite pattern: successful acqui- some time for desires for crisps to reduce, and as long as the sition of a swallowing response to a CS+ vs. CS− in over- cued desires are still present, not eating the crisps will be weight but (unexpectedly) not in normal-weight individuals difficult. If extinction of appetitive responses (such as eating was found. One may argue that findings reporting reduced desires) is slow, this could diminish one’s chances to diet learning could have been plausibly caused by overweight/ successfully [53]. In addition, it has been argued that partial obese individuals using implicit or explicit strategies that can (in contrast to continuous) reinforcement of food cues might lead to reduced learning about CS–US contingencies, like be frequently practiced by a significant portion of unsuccess- being reluctant to report appetitive desires [49], having ful dieters who try to abstain from eating in response to certain mindsets that suppress appetitive responding [50], or suffering food cues but regularly fail [32]. Partial reinforcement is well from a greater cognitive load due to thoughts about eating, known to result in slowed extinction of appetitive responses dieting, and weight [44� ]. Thus, moreresearchisneededto (the partial reinforcement extinction effect)—possibly leading examine differences in the acquisition of appetitive responses to a greater difficulty to extinguish appetitive responding and in overweight/obese versus normal-weight individuals, taking hence, more risk of lapses during one’s dieting attempts. into account these potential explanations, while preferably Why are cue-elicited desires seemingly so difficult to ex- assessing sensitive physiological indices of appetitive learning tinguish? It has been proposed that the difficulty to extinguish in addition to self-report measures. eating desires could indicate the presence of separate response To summarize, food cue reactivity has been shown to be systems: possibly, US expectancies stem from a system that related to overeating and weight gain and can partly be learned serves to prepare an organism for the incoming US, whereas through Pavlovian learning principles. Laboratory studies have eating desires might stem from another (hedonic/evaluative) system that is based on the mere activation of the US repre- shown that relatively few CS–US pairings are sufficient to form CS–US associations, and that—consequently—encounters of sentation in memory and that is less sensitive to extinction [30, CSs enhance food cue reactivity. While this model has also 52, 54–56]. The implication is that to reduce acquired eating been established in real-life circumstances, it remains of interest desires more effectively, it may be fruitful to investigate tech- to study whether overweight individuals might be different niques that specifically target this hedonic/evaluative response from normal-weight individuals in their appetitive conditioning system (see e.g., [54]). capacity. Initial evidence suggests that overweight individuals One important characteristic of extinction is that it does not show impairments in the acquisition of appetitive responses, reflect mere “unlearning” of the CS–US relationship; it largely forming an interesting direction for future investigations. reflects new learning of a second—contextually controlled—in- hibitory association (CS–noUS) that competes with the original CS–US association [57]. So, building upon the example of Carl: Extinction of Appetitive Responding practicing extinction of not eating crisps while watching the TV show leads to a novel association: namely that the TV show is no Given that food cue reactivity is partly learned, it may be longer a predictor of crisps intake (CS–noUS), while the original reduced through extinction, which refers to the decline in association between the TV show and crisps intake also remains (appetitive) responding with repeated CS–noUS pairings after intact (CS–US). Consistent with this account, a large amount of acquisition [13]. Extinction of appetitive responses in real life (mostly animal and some human) data show that after extinction can be achieved during a diet: when CSs, such as a chocolate- procedures, seemingly extinguished appetitive responses can re- associated time of day, or watching one’s crisp-associated fa- turn under certain conditions—as demonstrated by phenomena vorite TV show in the evening, are deliberately no longer such as rapid reacquisition (the rapid return of responding with followed by food intake (noUS), food cue reactivity (e.g., new CS–US pairings after extinction, e.g., when eating crisps eating desires) decreases over time. Several laboratory studies while watching the TV show once after a diet), reinstatement have indeed shown that conditioned responses (such as eating (the return of responding when the US is provided after 226 Curr Addict Rep (2018) 5:223–231 extinction – e.g., eating crisps again after a diet), renewal (the are thus exposed to intake-predicting cues (CSs) such as see- return of responding after extinction with a change in context, ing, smelling, and tasting the food, while for example being in e.g., extinction of evening cravings during a holiday and an overeating-associated context, and/or while experiencing a returning home), and spontaneous recovery (the spontaneous specific emotion, though food intake is not permitted (noUS) return of responding when time has elapsed after extinction, and food is thrown away at the end of the session. In line with i.e., after a period of successful dieting) [13, 52, 56]. Extinction a learning-based interpretation of food cue reactivity and over- is dependent on the context for expression, and therefore, a eating, the (small-scale) studies that have been conducted on change in context may promote a return of conditioned responses CET suggest it leads to substantial reductions in cue-elicited [57, 58]. Various types of (internal and external) stimuli can food cravings and binge eating in bulimic patients [63–68]. In provide such contexts [59, 60� ]. For example, one might learn overweight and obese samples, CET has also found to signif- that recent CS–US pairings are part of the acquisition “context,” icantly reduce cue-elicited cravings, binge eating, eating in the and recent CS–alone presentations are part of the extinction con- absence of hunger, and improved (maintenance of) weight text. When providing renewed CS–US pairings, this returns an loss [69–75]. Studies examining its long-term efficacy are individual to the acquisition context, resulting in a rapid return of however sparse with mixed findings [63, 70, 75], suggesting responding [58]. For example, Carl previously repeatedly con- that exposure protocols may require optimization. A number sumed crisps in the evening may have (largely) extinguished his of techniques derived from learning theory may be suitable for or her evening-crisp-cravings by refraining from eating crisps for this (for recent overviews, see [14]). Here, we focus on tech- awhile(CS–alone presentations, extinction). However, after niques that have received (recent) empirical investigation. months of strictly sticking to the crisps-less diet, Carl has some friends over and consumes crisps again in the crisp-associated Foods to Be Included in Cue Exposure Sessions Aconsistent context (reinforcing the original CS–US association). This return finding in CET studies is that after one to seven exposure to this original acquisition context may elicit a renewed craving sessions to a certain high-calorie food, intake of that specific for crisps in the evenings over the next days and risks the lapse food during an ad libitum intake test is decreased, compared turning into a full-blown relapse. The ease of conditioned re- with ad libitum intake after a control lifestyle intervention [72, sponses to re-emerge after extinction might explain the observa- 73, 75]. Several prior studies have additionally examined tion that although a considerable proportion of dieters are able to whether exposure effects generalize to (often personalized) achieve initial weight loss, only few are able to also successfully high-calorie food items not explicitly included during expo- maintain their weight loss—most dieters regain the lost weight sure sessions. These studies found that, generally, exposure (or even more) (e.g., [61, 62]). effects do not transfer to non-exposed foods [72, 73, 75]. Thus, insights into mechanisms of extinction can explain why The implication for CET is straightforward: to achieve clini- cally relevant effects on overeating and weight loss, the ther- diets are often unsuccessful: first, dieters likely experience per- sistent eating desires even after having successfully refrained apist should include all relevant food cues in the exposure from intake for a while due to not reaching (full) extinction sessions, i.e., those that are identified by the subject as cues (which may be exacerbated by partial reinforcement extinction for overeating. effects). Such prolonged eating desires could be (too) hard to resist over a longer period of time—promoting lapses in the diet Occasional Reinforced Extinction It has been argued that oc- that, in their turn, again strengthen the CS–US association. casionally reinforcing the CS–US relationship during expo- Second, when dieters finally achieve (partial) extinction of food sure therapy may lead to a reduced risk of relapse of binge cue reactivity, responding can easily return, thereby promoting and overeating [76]. Occasional reinforcements might reduce (re)lapse. Therefore, the long-term effectiveness of dieting efforts the risk of relapse is by allowing a reinforced trial (CS–US and treatments might be considerably improved by focusing on pairing) to become associated with extinction (CS–alone pre- strengthening extinction learning and diminishing the magnitude sentations), leading to a slower rate of reacquisition with of returns of appetitive responses. renewed CS–US pairings (e.g., when one experiences a lapse in one’sdiet) [76]. In CET, one way to implement this tech- nique is by encouraging individuals to occasionally consume Cue Exposure Therapy: Effects, Working small amounts of food during cue exposure exercises, in order Mechanisms, and its Optimization to learn that eating a small amount of food is no longer a cue for overeating [13]. In support, animal and human appetitive Cue exposure therapy (CET) is the direct clinical analogue of conditioning [51, 76], as well as fear conditioning studies [77, experimental extinction. In CET, individuals who suffer from 78], has reported evidence for occasional reinforcements dur- overeating or binge eating are repeatedly and long-lasting ex- ing extinction to reduce relapse of responding (i.e., rapid re- posed to palatable food cues (CSs), while refraining from acquisition). In studies on CET, occasional reinforced extinc- tion has not yet been studied in isolation, but some studies intake (i.e., CS–noUS). During CET sessions, participants Curr Addict Rep (2018) 5:223–231 227 have incorporated this technique in exposure sessions by let- such as high heart rate and sweating, which maximize the expec- ting participants take small bites of high-calorie food previ- tancy of the US to take place—the heart attack. The larger the ously associated with overeating [74� , 75]. These studies pro- mismatch between perceived expectancy of the heart attack to vided indirect support for the effectiveness of this technique occur and the actual absence of such a heart attack (noUS), the by showing that specific overeating expectancies—specified stronger expectancies are violated and hence the better strength- for taking one bite of the food—significantly reduced after ening of the CS–noUS association. Some experimental studies CET vs. control. Nevertheless, examining the pure benefits have indeed provided evidence for the benefits of focusing ex- of adding occasional reinforcements to CET in (sub)clinical posure sessions on expectancy violation, rather than habituation: samples awaits investigation. panic disorder patients who received two sessions of exposure therapy aimed at the violation of danger expectancies reported Working Mechanisms of CET Gaining insight into the possible fewer panic attacks, less anxiety, and less avoidance of fearful working mechanisms of CET could greatly help to optimize cue situations after therapy, compared to patients who received two exposure protocols. It has been proposed that the reduction in sessions of habituation-focused exposure ( [84]; see also [85]). cue-elicited eating desires could be a primary driver underlying It is likely that violation of CS–US expectancies also play a CET’s effectiveness, which is why most prior CET studies have significant role in CET. For example, a dieter may verbalize focused on achieving a reduction (or “habituation”)of eating the expectancy “If I feel exhausted and chocolate is available desires during exposure sessions. That is, exposure sessions are (CS), then I will lose control and eat the entire chocolate box usually terminated when cue-elicited cravings (e.g., [79], arousal (US).” To maximize violation of overeating expectancies dur- [71], or anxiety levels (e.g., [65])) have reached a pre-determined ing CET, a patient is exposed to the box of chocolate when threshold (e.g., 10 on a 100-point scale). Likewise, a large feeling exhausted, while exposure is continued until “losing amount of fear studies aimed to achieve habituation of fear within control” is no longer expected (for example by asking how exposure sessions (within-session habituation or WSH) and be- much time it will take to lose control and continuing exposure tween exposure sessions (between-session habituation or BSH). slightly beyond this point). To answer the question whether It has for long been argued that treatment success was dependent expectancy violation or habituation determines the effective- on successful WSH and BSH [80, 81]. However, research point- ness of CET, it was studied whether the reduction of CS–US ed out that indices of habituation are not consistently related to expectancies (used as index of the violation of CS–US expec- better treatment outcome in anxiety disorders [82]. Instead of tancies) or WSH and BSH of cue-elicited eating desires were focusing on habituation of fear, it has been proposed that expo- associated with the intake of food after exposure [72, 73, 74� , sure sessions should be designed in a way that maximizes the 75]. Data of four experimental studies, similar in design, were violation of CS–US expectancies, since this should effectively aggregated to overcome power issues (see Table 1). It was strengthen the inhibitory CS–noUS association [83]. Specifically, found that greater expectancy violation, but not habituation, exposure sessions should aim at maximizing the expectancy of was modestly but significantly associated with reduced intake the individual’s catastrophic event to occur. For example, expo- of exposed food after exposure. Thus, focusing on the reduc- sure sessions for panic disorder should carefully introduce CSs, tion of eating desires during CET might not be necessary for a Table 1 Correlations between z-scores of kcal intake of foods included (exposed) and not included (non-exposed) in CET and within-session habituation (WSH) and between-session habituation (BSH) of eating desires, and expectancy violation among participants of four studies [72, 73, 74� , 75] WSH of eating desires BSH of eating desires Expectancy violation n =140 n =94 n =150 z intake exposed food − 0.05 − 0.08 − 0.18* z intake non-exposed food 0.02 − 0.06 − 0.12 In all studies [72, 73, 74� , 75], WSH was calculated subtracting the end eating desire from the session’s peak eating desire (measured on a 100 mm Visual Analogue Scale [VAS]), and, in case of multiple exposure sessions, the average was calculated. BSH was calculated for three studies (which included 2 or more exposure sessions) by subtracting peak desires in the last sessions from peak desires in the first session. Expectancy violation was calculated by subtracting post-intervention expectancies from pre-intervention expectancies (measured on 100 mm VAS); in studies where expectancies about eating specific foods were asked, scores were averaged. As expectancies were assessed using a 5-point scale in [73], scores were transformed (1 = 0; 2 = 25; 3 = 50; 4 = 75; 5 = 100) *p < .05; sample sizes of each correlation vary due to measurement differences between studies To compare the intake during the taste test in all studies, z-scores were calculated. Some differences between studies are worth mentioning: general exposed foods were included in the taste tests described in [72, 73], while foods in [74� , 75]were personally selected.In [75], one general exposed food item was additionally included during the taste test, but no general non-exposed food item was included as comparison and therefore not included in the present investigation. Pure kcals were converted into z-scores for the data of [72, 74� , 75], while the intake in [73] was converted into a percentage of daily required energy intake for children, and then converted into z-scores 228 Curr Addict Rep (2018) 5:223–231 food cue exposure session to be effective. Instead, the reduc- initially neutral stimuli—both in the laboratory and in real tion of CS–US expectancies might be an important treatment life—can easily elicit conditioned appetitive responding after a target, like in anxiety disorders. few stimulus-food pairings. Extinction of some appetitive re- However, to critically evaluate whether expectancy violation sponses (e.g., desires) seems more difficult to achieve, and even and/or habituation are causally related to treatment success, it is after successful extinction, appetitive responding remains prone necessary to manipulate them. In a recent study, two exposure to relapse—perhaps explaining the general difficulty to achieve conditions were compared to a no-treatment control condition successful long-term weight loss. Cue exposure therapy, the clin- [74� ]. Participants in the exposure conditions either received ex- ical equivalent of extinction, seems effective in lowering cue- posure explicitly aimed at the habituation of eating desires (ex- elicited cravings, overeating, and binge eating, but still little is posure habituation condition) or exposure explicitly aimed at the known about the most effective exposure strategies and its work- violation of CS–US expectancies (exposure expectancy violation ing mechanisms. Reductions in overeating expectancies seem to condition). In all conditions, food intake was measured, while be important. Future research may aim to elucidate inter- expectancy violation and habituation was also measured in both individual differences in Pavlovian learning (e.g., as a function exposure conditions. Results indicated that although expectancy of weight status) and their consequences for eating behavior, to violation was stronger in the exposure expectancy violation vs. study how successful long-term extinction of appetitive exposure habituation condition during exposure sessions, the re- responding is best achieved, as well as to examine CET’s work- duction from pre- to post-therapy was equally strong in both ing mechanisms. conditions. Moreover, consumption during an ad libitum intake Acknowledgements This work is supported by the Netherlands test was similarly reduced across the two exposure conditions Organization for Scientific Research (NWO): Vici Grant 453.10.006, relative to the control condition. These results might question awarded to Anita Jansen. the importance of explicitly manipulating the violation of expec- tancies during exposure (see also [52]). They indicate that al- Compliance with Ethical Standards though greater reductions of CS–US expectancies are related to better outcome, it might not be essential to explicitly tackle these Conflict of Interest The authors declare that they have no conflict of CS–US expectancies during cue exposure to reduce them suc- interest. cessfully, since focusing on the reduction of eating desires led to Human and Animal Rights and Informed Consent All reported studies/ similarly reduced expectancies and food intake after the interven- experiments with human or animal subjects performed by the authors tion. Interestingly, similar findings have also been documented in have been previously published and complied with all applicable ethical the anxiety domain. For example, patients who completed a standards. purely behavioral treatment for animal phobias profited as much Open Access This article is distributed under the terms of the Creative from the treatment as patients who received a cognitive- Commons Attribution 4.0 International License (http:// behavioral treatment, in which dysfunctional cognitions were creativecommons.org/licenses/by/4.0/), which permits unrestricted use, challenged by behavioral tests [86, 87]. Moreover, the (similar) distribution, and reproduction in any medium, provided you give appro- priate credit to the original author(s) and the source, provide a link to the change in maladaptive cognitions in both treatment conditions Creative Commons license, and indicate if changes were made. was found to mediate the improvement [87], which generally fits with the findings in CET: changing overeating expectancies is important, while explicitly challenging these expectancies does References not seem to be crucial. To recapitulate, studies indicate that CET is a promising intervention to target cue-elicited cravings, overeating, binge Papers of particular interest, published recently, have been eating, and weight, although long-term effects are not yet highlighted as: studied. CET may be further optimized by incorporating in- � Of importance sights from learning theory. The working mechanism of CET is still unknown, though initial evidence suggests that the vi- 1. NCD Risk Factor Collaboration (NCD-RisC). Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled olation of US expectancies is necessary. It is clear that more analysis of 1698 population-based measurement studies with 19·2 work in this area is needed. million participants. Lancet. 2016;387(10026):1377–96. https:// doi.org/10.1016/S0140-6736(16)30054-X. 2. Ng M, Fleming T, Robinson M, Thomson B, Graetz N, Margono C, et al. Global, regional, and national prevalence of overweight and obe- Conclusion sity in children and adults during 1980-2013: a systematic analysis for the global burden of disease study 2013. Lancet. 2014;384(9945):766– Pavlovian learning theory provides a theoretical framework of 81. https://doi.org/10.1016/S0140-6736(14)60460-8. how cravings and overeating can be learned and how they might 3. GuhDP, ZhangW, BansbackN, AmarsiZ, BirminghamCL, Anis AH. The incidence of co-morbidities related to obesity and be effectively tackled. Experimental studies indeed show that Curr Addict Rep (2018) 5:223–231 229 overweight: a systematic review and meta-analysis. BMC Public 23. Staiger P, Dawe S, McCarthy R. Responsivity to food cues in bu- limic women and controls. Appetite. 2000;35(1):27–33. https://doi. Health. 2009;9(88) https://doi.org/10.1186/1471-2458-9-88. org/10.1006/appe.2000.0327. 4. Luppino FS, de Wit LM, Bouvy PF, Stijnen T, Cuijpers P, Penninx BWJH, et al. Overweight, obesity, and depression: a systematic 24. van den Akker K, Stewart K, Antoniou EE, Palmberg A, Jansen A. review and meta-analysis of longitudinal studies. Arch Gen Food cue reactivity, obesity, and impulsivity: are they associated? Psychiatry. 2010;67(3):220–9. https://doi.org/10.1001/ Curr Addict Reports. 2014;1(4):301–8. https://doi.org/10.1007/ archgenpsychiatry.2010.2. s40429-014-0038-3. 5. Pulgarón ER. Childhood obesity: a review of increased risk for 25. Carnell S, Haworth CMA, Plomin R, Wardle J. Genetic influence physical and psychological comorbidities. Clin Ther. 2013;35(1): on appetite in children. Int J Obes. 2008;32(10):1468–73. https:// A18–32. https://doi.org/10.1016/j.clinthera.2012.12.014. doi.org/10.1038/ijo.2008.127. 6. Scott KM, Bruffaerts R, Simon GE, Alonso J, Angermeyer M, De 26. Pavlov IP. Conditioned reflexes: an investigation of the physiolog- Girolamo G, et al. Obesity and mental disorders in the general ical activity of the cerebral cortex. London: Oxford University population: results from the world mental health surveys. Int J Press; 1927. Obes. 2008;32(1):192–200. https://doi.org/10.1038/sj.ijo.0803701. 27. Davidson TL, Kanoski SE, Tracy AL, Walls EK, Clegg D, Benoit 7. Cawley J, Meyerhoefer C. The medical care costs of obesity: an SC. The interoceptive cue properties of ghrelin generalize to cues instrumental variables approach. J Health Econ. 2012;31(1):219– produced by food deprivation. Peptides. 2005;26(9):1602–10. 30. https://doi.org/10.1016/j.jhealeco.2011.10.003. https://doi.org/10.1016/j.peptides.2005.02.014. 8. Leung HT, Reeks LM, Westbrook RF. Two ways to deepen extinc- 28. Vohs KD, Wang Y, Gino F, Norton MI. Rituals enhance consump- tion and the difference between them. J Exp Psychol Anim Behav tion. Psychol Sci. 2013;24(9):1714–21. https://doi.org/10.1177/ Process. 2012;38(4):394–406. https://doi.org/10.1037/a0030201. 0956797613478949. 9. Goran MI, Weinsier RL. Role of environmental vs. metabolic fac- 29. Wardle J. Conditioning processes and cue exposure in the modifi- tors in the etiology of obesity: time to focus on the environment. cation of excessive eating. Addict Behav. 1990;15(4):387–93. Obes Res. 2000;8(5):407–9. https://doi.org/10.1038/oby.2000.50. https://doi.org/10.1016/0306-4603(90)90047-2. 10. Swinburn B, Egger G, Raza F. Dissecting obesogenic environ- 30. Van Gucht D, Vansteenwegen D, Van den Bergh O, Beckers T. ments: the development and application of a framework for identi- Conditioned craving cues elicit an automatic approach tendency. fying and prioritizing environmental interventions for obesity. Prev Behav Res Ther. 2008;46(10):1160–9. https://doi.org/10.1016/j. Med. 1999;29:563–70. https://doi.org/10.1006/pmed.1999.0585. appet.2007.05.003. 11.� Boswell RG, Kober H. Food cue reactivity and craving predict 31. Papachristou H, Nederkoorn C, Havermans R, Bongers P, Beunen eating and weight gain: a meta-analytic review. Obes Rev. S, Jansen A. Higher levels of trait impulsiveness and a less effective 2016;17:159–77. https://doi.org/10.1111/obr.12354. Meta- response inhibition are linked to more intense cue-elicited craving analysis showing the importance of cue-elicited cravings for for alcohol in alcohol-dependent patients. Psychopharmacology. eating behaviour and weight gain. 2013;228(4):641–9. https://doi.org/10.1007/s00213-013-3063-3. 12. Boggiano MM, Dorsey JR, Thomas JM, Murdaugh DL. The 32. van den Akker K, Havermans RC, Bouton ME, Jansen A. How partial Pavlovian power of palatable food: lessons for weight-loss adher- reinforcement of food cues affects the extinction and reacquisition of ence from a new rodent model of cue-induced overeating. Int J appetitive responses. A new model for dieting success? Appetite. Obes. 2009;33(6):693–701. https://doi.org/10.1038/ijo.2009.57. 2014;81:242–52. https://doi.org/10.1016/j.appet.2014.06.024. 13. Bouton ME. Learning and the persistence of appetite: extinction 33. Andreatta M, Pauli P. Appetitive vs. aversive conditioning in and the motivation to eat and overeat. Physiol Behav. humans. Front Behav Neurosci. 2015;9:1–8. https://doi.org/10. 2011;103(1):51–8. https://doi.org/10.1016/j.physbeh.2010.11.025. 3389/fnbeh.2015.00128. 14. Jansen A, Schyns G, Bongers P, van den Akker K. From lab to clinic: 34.� van den Akker K, Schyns G, Jansen A. Altered appetitive condi- extinction of cued cravings to reduce overeating. Physiol Behav. tioning in overweight and obese women. Behav Res Ther. 2017;99: 2016;162:174–80. https://doi.org/10.1016/j.physbeh.2016.03.018. 78–88. https://doi.org/10.1016/j.brat.2017.09.006. Study 15. Woods SC. The eating paradox: how we tolerate food. Psychol Rev. suggesting appetitive conditioning impairments in overweight 1991;98(4):488–505. https://doi.org/10.1037/0033-295X.98.4.488. vs. normal-weight individuals. 16. Bouton ME. Learning and behavior: a contemporary synthesis. 35. Blechert J, Testa G, Georgii C, Klimesch W, Wilhelm FH. The Sunderland: Sinauer Associates, Inc.; 2007. Pavlovian craver: neural and experiential correlates of single trial 17. Power ML, Schulkin J. Anticipatory physiological regulation in naturalistic food conditioning in humans. Physiol Behav. 2016;158: feeding biology: cephalic phase responses. Appetite. 2008;50(2– 18–25. https://doi.org/10.1016/j.physbeh.2016.02.028. 3):194–206. https://doi.org/10.1016/j.appet.2007.10.006. 36. Bongers P, Jansen A. Emotional eating and Pavlovian learning: 18. Woods SC, Ramsay DS. Pavlovian influences over food and drug evidence for conditioned appetitive responding to negative emo- intake. Behav Brain Res. 2000;110(1–2):175–82. https://doi.org/ tional states. Cogn Emot. 2017;31(2):284–97. https://doi.org/10. 10.1016/S0166-4328(99)00194-1. 1016/j.appet.2015.02.018. 19. Ferriday D, Brunstrom JM. “Ijustcan”thelp myself’: effects of 37. Birch LL, McPhee L, Sullivan S, Johnson S. Conditioned meal food-cue exposure in overweight and lean individuals. Int J Obes. initiation in young children. Appetite. 1989;13(2):105–13. https:// 2011;35(1):142–9. https://doi.org/10.1038/ijo.2010.117. doi.org/10.1016/0195-6663(89)90108-6. 20. Ferriday D, Brunstrom JM. How does food-cue exposure lead to 38. Astur RS, Carew AW, Deaton BE. Conditioned place preferences in larger meal sizes? Br J Nutr. 2008;100(6):1325–32. https://doi.org/ humans using virtual reality. Behav Brain Res. 2014;267:173–7. 10.1017/S0007114508978296. https://doi.org/10.1016/j.bbr.2014.03.018. 21. Jansen A. A learning model of binge eating: cue reactivity and cue 39. van den Akker K, Jansen A, Frentz F, Havermans RC. Impulsivity exposure. Behav Res Ther. 1998;36(3):257–72. https://doi.org/10. makes more susceptible to overeating after contextual appetitive 1016/S0005-7967(98)00055-2. conditioning. Appetite. 2013;70:73–80. https://doi.org/10.1016/j. appet.2013.06.092. 22. Jansen A, Theunissen N, Slechten K, Nederkoorn C, Boon B, Mulkens S, et al. Overweight children overeat after exposure to 40. Meyer MD, Risbrough VB, Liang J, Boutelle KN. Pavlovian condi- food cues. Eat Behav. 2003;4(2):197–209. https://doi.org/10. tioning to hedonic food cues in overweight and lean individuals. 1016/S1471-0153(03)00011-4. Appetite. 2015;87:56–61. https://doi.org/10.1016/j.appet.2014.12.002. 230 Curr Addict Rep (2018) 5:223–231 41. Bongers P, van den Akker K, Havermans R, Jansen A. Emotional to eat chocolate. Behav Res Ther. 2008;46(3):375–91. https://doi. org/10.1016/j.brat.2008.01.003. eating and Pavlovian learning: does negative mood facilitate appe- titive conditioning? Appetite. 2015;89:226–36. https://doi.org/10. 57. Bouton ME. Context, ambiguity , and unlearning : sources of relapse 1016/j.appet.2015.02.018. after behavioral extinction four mechanisms of relapse. 2002;52: 976– 986. doi:https://doi.org/10.1016/S0006-3223(02)01546-9. 42.� Akker K, Van Den HRC, Jansen A. Appetitive conditioning to specific times of day. Appetite. 2017;116:232–8. https://doi.org/ 58. Bouton ME. Context and behavioral processes in extinction. Learn Mem. 2004;11(5):485–94. https://doi.org/10.1101/lm.78804. 10.1016/j.appet.2017.05.014. Experiment showing that findings of conditioned food cue reactivity in laboratory settings 59. Bouton ME. Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychol Bull. 1993;114(1):80– translate to real-life circumstances. 99. https://doi.org/10.1037/0033-2909.114.1.80. 43. Coppin G, Nolan-Poupart S, Jones-Gotman M, Small DM. 60.� Schepers ST, Bouton ME. Hunger as a context: food seeking that is Working memory and reward association learning impairments in inhibited during hunger can renew in the context of satiety. Psychol obesity. Neuropsychologia. 2014;65:146–55. https://doi.org/10. Sci. 2017;28(11):1640 –8. https://doi.org/10.1177/ 1016/j.neuropsychologia.2014.10.004. 0956797617719084. Two experiments in rodents 44.� Zhang Z, Manson KF, Schiller D, Levy I. Impaired associative demonstrating that satiety can provide a “context” and lead learning with food rewards in obese women. Curr Biol. to renewal of (operant) responding. 2014;24(15):1731–6. https://doi.org/10.1016/j.cub.2014.05.075. 61. Mann T, Tomiyama a J, Westling E, Lew A-M, Samuels B, Study indicating differences in associative learning in obese Chatman J. Medicare’s search for effective obesity treatments: diets vs. normal-weight participants, only when food (not money) are not the answer. Am Psychol. 2007;62(3):220–33. https://doi. was used as a reward. org/10.1037/0003-066X.62.3.220. 45. Davidson TL, Martin AA. Obesity: cognitive impairment and the 62. Wing RR, Phelan S. Long-term weight loss maintenance. Am J Clin failure to “eat right”. Curr Biol. 2014;24(15):R685–7. https://doi. Nutr. 2005;82(1):222S–5S. Doi: 82/1/222S [pii] org/10.1016/j.cub.2014.06.031. 63. Jansen A, Broekmate J, Heymans M. Cue-exposure vs self-control in 46. Lissek S, Powers AS, McClure EB, Phelps EA, Woldehawariat G, the treatment of binge eating: a pilot study. Behav Res Ther. Grillon C, et al. Classical fear conditioning in the anxiety disorders: 1992;30(3):235–41. https://doi.org/10.1016/0005-7967(92)90069-S. a meta-analysis. Behav Res Ther. 2005;43(11):1391–424. https:// 64. Jansen A, van den Hout MA, de Loof C, Zandbergen J, Griez E. A doi.org/10.1016/j.brat.2004.10.007. case of bulimia succesfully treated by cue exposure. Behav Ther 47. Hermans D, Baeyens F, Vervliet B. Generalization of acquired emotion- Exp Psychiatry. 1989;20(4):327–32. https://doi.org/10.1016/0005- al responses. In: Handbook of Cognition and Emotion; 2013. p. 117–34 7916(89)90064-5. 48. Grillon C, Davis M. Fear-potentiated startle conditioning in 65. Martinez-mallen E, Castro-fornieles J, La L, Moreno E, Morer A, humans: explicit and contextual cue conditioning following paired Font E, et al. Cue exposure in the treatment of resistant adolescent versus unpaired training. Psychophysiology. 1997;34:451–8. bulimia nervosa. Int J Eat Disord. 2007;40:596–60 1. https://doi.org/ https://doi.org/10.1111/j.1469-8986.1997.tb02389.x. 10.1002/eat20423. 49. Roefs A, Quaedackers L, Werrij MQ, Wolters G, Havermans R, 66. Toro J, Cervera M, Feliu MH, Garriga N, Jou M, Martinez E, et al. Nederkoorn C, et al. The environment influences whether high-fat Cue exposure in the treatment of resistant bulimia nervosa. Int J Eat foods are associated with palatable or with unhealthy. Behav Res Disord. 2003;34(2):227–34. https://doi.org/10.1002/eat.10186. Ther. 2006;44(5):715–36. https://doi.org/10.1016/j.brat.2005.05.007. 67. Rosen JC, Leitenberg H. Bulimia nervosa : treatment with exposure 50. Siep N, Roefs A, Roebroeck A, Havermans R, Bonte M, Jansen A. and response prevention. Behav Ther. 1982;13:117–24. https://doi. Fighting food temptations: the modulating effects of short-term org/10.1016/S0005-7894(82)80055-5. cognitive reappraisal, suppression and up-regulation on 68. Schmidt U, Marks I. Cue exposure to food plus mesocorticolimbic activity related to appetitive motivation. responsepreventionof binges for bulimia: a pilot study. NeuroImage. 2012;60(1):213–20. https://doi.org/10.1016/j. 1988;7(5):663–72. https://doi.org/10.1002/1098-108X(198809)7: neuroimage.2011.12.067. 5<663::AID-EAT2260070510>3.0.CO;2-X. 51. van den Akker K, Havermans RC, Jansen A. Effects of occasional 69. Boutelle KN, Zucker NL, Peterson CB, Rydell SA, Cafri G, reinforced trials during extinction on the reacquisition of condi- Harnack L. Two novel treatments to reduce overeating in over- tioned responses to food cues. J Behav Ther Exp Psychiatry. weight children: a randomized controlled trial. J Consult Clin 2015;48:50–8. https://doi.org/10.1016/j.jbtep.2015.02.001. Psychol. 2011;79(6):759–71. https://doi.org/10.1037/a0025713. 52. van den Akker K, van den Broek M, Havermans RC, Jansen A. 70. Boutelle KN, Zucker N, Peterson CB, Rydell S, Carlson J, Harnack LJ. Violation of eating expectancies does not reduce conditioned de- An intervention based on Schachter’s externality theory for overweight sires for chocolate. Appetite. 2016;100:10–7. https://doi.org/10. children: the regulation of cues pilot. J Pediatr Psychol. 2014 1016/j.appet.2016.02.004. May;39(4):405–17. https://doi.org/10.1093/jpepsy/jst142. 53. Jansen A, Stegerman S, Roefs A, Nederkoorn C, Havermans R. 71. Mount R, Neziroglu F, Taylor CJ. An obsessive-compulsive view Decreased salivation to food cues in formerly obese successful di- of obesity and its treatment. J Clin Psychol. 1990;46(1):68–78. eters. Psychother Psychosom. 2010;79(4):257–8. https://doi.org/ 72. Schyns G, Roefs A, Mulkens S, Jansen A. Expectancy violation, 10.1159/000315131. reduction of food cue reactivity and less eating in the absence of 54. Van Gucht D, Baeyens F, Vansteenwegen D, Hermans D, Beckers hunger after one food cue exposure session for overweight and T. Counterconditioning reduces cue-induced craving and actual obese women. Behav Res Ther. 2016;76:57–64. https://doi.org/ cue-elicited consumption. Emotion. 2010;10(5):688–95. https:// 10.1016/j.brat.2015.11.007. doi.org/10.1037/a0019463. 73. Schyns G, Roefs A, Smulders FTY, Jansen A. Cue exposure ther- 55. Van Gucht D, Vansteenwegen D, Beckers T, Hermans D, Baeyens apy reduces overeating of exposed and non-exposed foods in obese F, Van den Bergh O. Repeated cue exposure effects on subjective adolescents. J Behav Ther Exp Psychiatry. 2018;58:68–77. https:// and physiological indices of chocolate craving. Appetite. doi.org/10.1016/j.jbtep.2017.09.002. 2008;50(1):19–24. https://doi.org/10.1016/j.appet.2007.05.003. 74.� Schyns G, van den Akker K, Roefs A, Hilberath R, Jansen A. What 56. Van Gucht D, Vansteenwegen D, Beckers T, Van den Bergh O. works better ? Food cue exposure aiming at the habituation of eating desires or food cue exposure aiming at the violat ion of Return of experimentally induced chocolate craving after extinction in a different context: divergence between craving for and expecting overeating expectancies? Behav Res Ther. 2018;102:1–7. https:// Curr Addict Rep (2018) 5:223–231 231 doi.org/10.1016/j.brat.2017.12.001. Experiment on the working 82. Craske MG, Kircanski K, Zelikowsky M, Mystkowski J, Chowdhury N, Baker A. Optimizing inhibitory learning during mechanisms of food cue exposure therapy, showing that explicitly tackling overeating expectancies might not be exposure therapy. Behav Res Ther. 2008;46(1):5–27. https://doi. necessary for treatment success. org/10.1016/j.brat.2007.10.003. 75. Schyns G, van den Akker K, Roefs A, Houben K, Jansen A. Weight 83. Craske MG, Treanor M, Conway CC, Zbozinek T, Vervliet B. loss and binge eating reduction after eight sessions food cue expo- Maximizing exposure therapy: an inhibitory learning approach. sure therapy In Rev. Behav Res Ther. 2014;58:10–23. https://doi.org/10.1016/j.brat. 76. Bouton ME, Woods AM, Pineño O. Occasional reinforced trials during 2014.04.006. extinction can slow the rate of rapid reacquisition. Learn Motiv. 84. Salkovskis PM, Hackmann A, Wells A, Gelder MG, Clark DM. Belief 2004;35(4):371–90. https://doi.org/10.1016/j.lmot.2004.05.001. disconfirmation versus habituation approaches to situational exposure 77. Culver NC, Stevens S, Fanselow MS, Craske MG. Building phys- in panic disorder with agoraphobia: a pilot study. Behav Res Ther. iological toughness: some aversive events during extinction may 2006;45(5):877–85. https://doi.org/10.1016/j.brat.2006.02.008. attenuate return of fear. J Behav Ther Exp Psychiatry. 2018;58: 85. Deacon BJ, Kemp JJ, Dixon LJ, Sy JT, Farrell NR, Zhang AR. 18–28. https://doi.org/10.1016/j.jbtep.2017.07.003. Maximizing the efficacy of interoceptive exposure by optimizing 78. Morís J, Barberia I, Vadillo MA, Andrades A, López FJ. Slower inhibitory learning: a randomized controlled trial. Behav Res Ther. reacquisition after partial extinction in human contingency learning. 2013;51(9):588–96. https://doi.org/10.1016/j.brat.2013.06.006. J Exp Psychol Learn Mem Cogn. 2017;43(1):81–93. https://doi. 86. Koch EI, Spates CR, Himle JA. Comparison of behavioral and org/10.1037/xlm0000282. cognitive-behavioral one-session exposure treatments for small an- 79. Boutelle KN, Knatz S, Carlson J, Bergmann K, Peterson CB. An imal phobias. Behav Res Ther. 2004;42(12):1483–504. https://doi. open trial targeting food cue reactivity and satiety sensitivity in org/10.1016/j.brat.2003.10.005. overweight and obese binge eaters. Cogn Behav Pract. 87. Raes AK, Koster EHW, Loeys T, De Raedt R. Pathways to change 2016;24(3):363–73. https://doi.org/10.1016/j.cbpra.2016.08.003. in one-session exposure with and without cognitive intervention: an 80. Foa EB, Kozak MJ. Emotional processing of fear: exposure to exploratory study in spider phobia. J Anxiety Disord. 2011;25(7): corrective information. Psychol Bull. 1986 Jan;99(1):20–35. 964–71. https://doi.org/10.1016/j.janxdis.2011.06.003. 81. Foa EB, McNally RJ. Mechanisms of change in exposure therapy. In: Rapee M, editor. Current controversies in the anxiety disorders. New York: The Guilford Press; 1996. p. 329–43.

Journal

Current Addiction ReportsSpringer Journals

Published: Apr 21, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off