Tobacco Documents Online

I I I I I I I I I I I I I I I I Appetite. 1993, ?1, 233-246 "Chocolate Addiction": a Preliminary Study of its Description and its Relationship to Problem Eating MARION M. HETHERINGTON and JENNIFER I. MACDIARMID Department of Psychology, University of Dundee Definitions of chocolate addiction and its potential relationship to dieting and problem eating were investigated in 50 individuals who identified themselves as "chocoholics". Respondents were interviewed and completed a battery of ques- tionnaires on food cravings, eating. weight. dieting and depression. On average this sample consumed about 12 (60-g) bars of chocolate per week and craved chocolate about six times per week. Cravings and amount consumed were not significantly related but amount consumed was significantly correlated with disinhibition (r=0•3). Most (76%) respondents had definitions of chocolate addiction that centred on a lack of control around chocolate and regarded the "addictive" factor in chocolate as orosensory (i.e. taste, smell, texture). Unlike most others, dieters and secret eaters experienced negative affect following consumption of chocolate. Consumers who preferred to eat in secret reported a higher degree of aberrant eating. The extent to which the behaviour of "chocolate addicts" resembles that of eating disordered individuals and other addictions remains to be clarified. INTRODUCTION Chocolate is identified as the single most craved food in studies of food cravings (Hill et al., 1991; Rodin et al., 1991; Rozin et al., 1991; Weingarten & Elston, 1991). More than most foods, chocolate has come to signify luxury, reward and pleasure (James, 1979). It is not surprising that chocolate is considered pleasurable by most people since it combines positive orosensory qualities, including a highly preferred sweet taste and creamy texture (Drewnowski & Greenwood, 1983) with positive connotations developed from childhood of gift-giving and special occasions (Barthel. 1989). It has been suggested that chocolate cravings have a biological basis, associated with depressed mood states (Schuman et al., 1987; Lester & Bernard. 1991), the menstrual cycle (Rozin et al., 1991; Smith & Sauder, 1969) and psycho- pharmacological effects (Weil & Rosen, 1983). However, equally plausible is that the desire for chocolate simply derives from the desire for sensory gratification (Rodin et al., 1991). "Chocolate addiction" or "chocoholism" are terms commonly used in the popular media (Take a Break, 1992; Sunday Express, 1992; Voak, 1992). Reports of excessive or exclusive consumption of chocolate have been associated with powerful cravings analogous to those associated with substance abuse. Since chocolate provides pleasure, contains psychoactive ingredients and may be Address correspondence to Dr Marion M. Hetherington. Department of Psychology, The Unirersity, Dundee DDI 4HN, Scotland. 0195--6663 93;060?33+ 14 $08.00 0 i 1993 Academic Press Limited I

I I I I I I I I I I I I I I I '134 M. ?v1. HETHERItiGTON AND J, 1. MACDIARMID considered a"forbidden fruit" (James. 1990). it is possible that some individuals abuse or binge on chocolate, as others might abuse other substances such as drugs or alcohol. Coffee intake may be driven by a desire for caffeine and its effects (Hughes et al.. 1992). Chocolate, as well as coffee and tea, is a source of caffeine and therefore potentially open to abuse. However, chocolate contributes relatively little to the adult intake of caffeine (Kanarek & Marks-Kaufman. 1991) and Rozin et al. (1991) have argued against a possible link between chocolate addiction and caffeine content (or other xanthines). This suggests that addiction to chocolate may be due to pieusure- seeking unrelated to pharmacological effects. Our understanding of addiction has shifted in recent years away from a physical dependence model to a positive-incentive model (Pinel, 1990). If the primary motivation underlying substance abuse. or addictive behaviours such as gambling, is pleasure-seeking or sensation-seeking (Anderson & Brown, 1984), then this widens considerably the range of abuse targets, e.g. work, exercise, eating and sex. There have been at least two case reports of treating chocolate addiction in the literature (Michell et al., 1989: Weil & Rosen. 1983). Some comparisons have been made between eating disorders and addictions (Beresford & Hall. 1989; Brisman & Siegel. 1984: Cooper, 1989: Flood. 1989: Orford, 1985; Vandereycken. 1990). Although there are shared characteristics, and some eating disordered patients present with substance abuse (Hatsukami et al., 1986), clinical diagnoses of the eating disorders are independent of the clinical criteria for addictive behaviours (APA. 1987). The present study set out to investigate the behaviour of individuals who identify themselves as "chocolate addicts" or "chocoholics". The main objectives of this study were tD discover how these individuals define chocolate addiction, what behavioural factors are associated with the consumption of chocolate in this group and what, if any, characteristics of this behaviour reflect true addictions. Psychological categories for identifying addictive behaviours were taken from Brown's taxonomy (Brown. 1988): salience, conflict, relief, relapse, tolerance and withdrawal. Since few previous studies have addressed the issue of chocolate addiction, a self- report approach (semi-structured interview) was combined with data collection from structured and validated questionnaires. METHODS Subjects Subjects were recruited through an advertisement placed in a local newspaper (covering the regions of Tayside and Fife. North East Scotland), which asked the question "are you a chocolate addict?". In response to the advert 75 subjects claiming to be "chocolate addicts" contacted the university willing to take part in the study. Of the 75 respondents, 54 (72%) of them completed the first part of the study, filling out a questionnaire on cravings and 50 (67%) were subsequently interviewed in more depth about their chocolate addiction, and administered a series of questionnaires. Procedure Consumers who responded to the initial advertisement were sent a modified version of a cravings questionnaire (Weingarten & Elston, 1991) asking more details about dieting and the extent to which cravings interfered with their lives. I

I I I I I I I I I I I I 1 I I I CHOCOLATE ADDICTION ~ 235 In the next part of the study we interviewed respondents more specifically about their chocolate addiction. The aim of this was to quantify consumption of chocolate and to record the subjects' beliefs, attitudes and feelings about their addiction. All interviews were tape-recorded after written informed consent and all tapes subse- quently transcribed for examination. Finally. respondents completed questionnaires to assess disordered eating, res- trained eating, body dissatisfaction and depression, namely: the Dutch Eating Behaviour Questionnaire (DEBQ: Van Strien et al.. 1986). the Eating Attitudes Test (EAT-40: Garner & Garfinkel. 1979). the Eating Inventory (EI: Stunkard & Messick, 1985), the Body Shape Questionnaire (BSQ: Cooper et al.. 1987) and the Beck Depression Inventory (BDI: Beck et al., 1961). RESt; LTS Participant Profile The mean (SEM) age of the respondents who completed the study was 39•8 (2•2) years, with a range of 14-83 years. The majority (92%) who responded and took part in the study were female. Mean (SEM) body mass index (BMI) of the sample was 25•3 (0•8) kg!m=, with a range of 16•4--41 .0. A majority of the respondents were just above the normal range of 20•0-24•9 (Garrow, 1981). Crarings As expected. the most intensely craved food reported by questionnaire was chocolate (94%) with the only other foods being "sweet foods" (2%), and pizza (4%). The mean frequency of craving chocolate per month was 24•7t2•2 times, ranging from I to 90 times per month. For the next strongest craving 20% craved no other food except chocolate, 28% craved sweet food (confectionery, biscuits, cakes and ice- cream), 19% craved individual foods (bread, pasta and cheese), 13% craved savoury meals (pizza. spicy foods, salad, fish and chips) and 10% craved salty snacks (peanuts and crisps). Frequency of craving for these foods was only 13•6t 1•9 per month. When asked to identify a substitute for their most craved food, 62% stated that there was none. Of those who reported substitutes, these were other forms of confectionery (32%), salty foods (4%) and fruit (2%). Respondents reported giving in to cravings for their most craved food 86-5% ± 2•7 of the time. The majority (86%) indulged their cravings 75% of the time or more. Affective response to consuming chocolate following a craving was 51 % positive (happier, satisfied, calm, recharged, energetic) and 49% negative (guilty, disap- pointed, unattractive, depressed, headachey, dissatisfied, sick). Female respondents were asked to report any relationship between cravings and menstrual cycle; 58% of all women (currently menstruating. pre-menopausal women) noticed an increase in cravings in the premenstrual phase. Reported Consumption The mean (SEM) consumption of chocolate per week reported at interview was 12•5 (1 •6) bars, with a range of I to 70 bars per week (see Figure 1). where a bar was I

I I I I I I I I I I I I I I I I I 236 M. M. HETHERINGTON AND J. I MACDIARMID Intake per day FIGURE 1. Frequency (% of respondents) reporting number of 60-g units of chocolate consumption each day. . less than i; ®, at least l: ®, at least 2; Z., at least 3: ::. more than 3. defined as a standard chocolate block of 60 g in weight. Consumption of chocolate did not correlate significantly with frequency of craving. The majority (72%) considered their consumption of chocolate to be excessive. Reasons given for this included: consumption contributed to overweight status, chocolate was considered high fat and of poor nutritional value, and in comparison to others. consumption seemed excessive. Half of the respondents stated that consumption of chocolate interfered with their life in some way. The type of chocolate preferred by the majority (76%) of subjects was milk chocolate with only 8% preferring dark chocolate and 4% preferring white chocolate. The remaining 12% had no stated preference. Chocolate was eaten in a variety of forms with the most preferred forms being block (or solid) chocolate (42%), and then bars of chocolate (38%). Chocolate was also eaten in the form of cakes, biscuits, desserts, ice-cream and drinks (hot chocolate, diet shakes). In addition. cooking chocolate, chocolate spread, and chocolate in cereals were also eaten. Mood When asked to report how they felt when cutting down on chocolate consump- tion, 66% reported negative affect (irritable, deprived, angry, unhappy), 6% reported feeling good when able to do this, 14% reported no particular feelings and 14% had never tried to cut down. Fifty-two per cent recorded negative moods (stressed, lonely, depressed) during cravings for chocolate, 46% reported no specific affect and only 2% felt positively. The majority of respondents described positive feelings before (54%), during (86%) and after (46%) eating chocolate, as distinct from negative or neutral moods. A chi-square analysis was conducted of the positive, negative and neutral responses recorded before, during and after consumption. A greater number reported positive affect during consumption of chocolate than before or after, x=(5)=25•45, p < 0•01. Interestingly, approximately half reported sustaining this positive affect after eating and half reported negative affect. I

I I I I I I I I CHOCOLATE ADDICTION 237 The difference noted between negative affect during cravings and positive affect just prior to consuming chocolate may suggest the power of anticipation. Cravings for chocolate can occur in the presence or absence of that food, whereasjust prior to eating chocolate the individual has the food there and is preparing to eat it. Definitions of AddFction The respondents were asked why they identified themselves as "chocolate addicts". The majority (76%) of respondents attributed this to being unable to resist chocolate. The inability to resist eating chocolate consisted of two components: feeling unable to moderate chocolate consumption andjor the inability to stop eating once eating had been initiated. Other reasons included the amount of chocolate consumed (12%), other people labelling the respondent as an addict (10%) and unspecified (2%). Respondents were also asked to state what it was about chocolate which they considered to be addictive. The majority (72%) identified sensory features of the food (taste, smell and texture) as being addictive. Other reasons given were the chemical composition of chocolate (caffeine, theobromine, unknown factor: 12%), or that chocolate itself was not addictive, but consumption had become habitual (4%), or that it was not considered addictive at all (8%). the remaining 4% did not specify a particular factor. Cravings and Dieting Many respondents (44%) reported no specific pattern of cravings or consumption. while 30% said late evening. 8% said mornings and 18% stated that consumption tended to be after meals. When asked to describe factors which stimulated cravings. subjects' responses were classified according to whether the factor was internal (e.g. mood: 36%, hunger: 8%), external (e.g. watching others, seeing advertisements: 36%), due to habit (4%) or no specific stimulus (16%). Participants were asked the question "are you currently on a diet, if no, have you ever been on a diet to lose weight?". Of the sample, 21 were currently on a diet to lose weight and 29 were not currently dieting. Of those not currently dieting, eight had been on a diet at some time and five were "conscientious" eaters (i.e. were not on a specific diet but "watched" what they ate, especially high-fat foods). Also, while dieting, 52% of respondents reported that dieting increased cravings and, for the others, dieting did not influence craving frequency. Conditions of Consumption and Attitudes towards Chocolate Respondents consumed chocolate alone, either by design or through circumstance or with others present. Secretive eating occurred in 40% of the sample. These were respondents who stated that they preferred to eat alone and in secret. When asked to express their general opinion of chocolate, the majority (84%) gave a positive response ("best thing in the world". "fantastic". "wouldn't want it to go out of fashion"), 12% gave a negative response ("wish it had never been invented", "it's bad for you") and 4% had no opinion. Given six descriptions of chocolate which were roughly categorized as positive (energy giving, a luxury), neutral (neither good nor bad, airight as part of a balanced diet) or negative (unhealthy. fattening). subjects were asked to select those which fit their opinion of chocolate. They were able to select

I I I I I ~ I I I I I I I I I 2-3g M. M. HETHERINGTON AND 1. I. MACDIARMID more than one description. A frequency distribution for these descriptions was generated and subjected to a chi-square analysis (contingency table). Attitudes towards chocolate were not equally distributed across the six outcomes, x=(5)= 15•03, p<0•01. The most frequently cited description of chocolate was that it was "fatten- ing" (25%); however, the second most frequently selected description was "alright as part of a balanced diet" (19%), followed by "energy-giving" (16%), "a luxury" (16%), "unhealthy" (14%) and "neither good nor bad" (10%). Psti•chological Variables The group means of scores on each of the questionnaires are within the normal range for control subjects published in the literature (see Laessle et al., 1989 for standard scores on restraint items) with two exceptions (Table 1). The mean score for the Body Shape Questionnaire in the community has been estimated at 81 •5 (Cooper et al., 1987); however, 60% of the present sample scored above this mean. Also, a score of 4 on the Beck Depression Inventory indicates mild depression (Steer et al., 1982), and 84% of the present sample scored above 4. A score of between 14 and 20 indicates moderate depression and 30% of the present sample scored above 14. Therefore, the majority of the sample were dissatisfied with their body shape and one- third of the sample reported moderate levels of depression. Conelations between reported amount of chocolate consumed and questionnaire scores gave only one significant result. Amount consumed correlated positively with the disinhibition item of the Eating Inventory, r(47)=0•314, p<0-05, indicating that reported amount was related to a disinhibited eating style which is influenced by the presence of palatable foods, social circumstances and emotional factors. Correlations between frequency of cravings and questionnaire scores resulted in one significant finding that frequencies of cravings were positively correlated with the external eating item of the Dutch Eating Behaviour Questionnaire, r(45)=0•448, p<0•0l. This suggests a link between cravings and environmental factors such as social occasions and the presence of tempting food. The percentage follow-through of cravings for this sample was significantly negatively correlated with a number of questionnaire scores. These negative correla- tions suggest that those subjects who fulfil their cravings most of the time, have the lowest scores on questionnaires reflecting problem eating, body dissatisfaction and depression. Dieters and Non-dieters Dieters were classified as those who were currently on a diet (n=21) and non- dieters as those not currently on a diet (n = 24), and the remaining subjects who were not clearly either dieters or non-dieters were excluded from comparisons. Dieters (12•2 f 1•7 bars per week) did not eat significantly more chocolate than non-dieters (9•4 ± 0•9 bars per week), nor did they report significantly greater cravings than non-dieters (dieters: 6•5 f 0•9 times per week; non-dieters: 5•9 f 0•8 times per week). However, dieters were significantly heavier than non-dieters. From multivar- iate analyses of variance with body mass index as a co-variate, it was clear that dieters scored significantly more highly on the EAT and on the restraint items of the DEBQ and El. I

I TABt_e: I Scores (mean f SEM) for secret and non-secret eaters on Body Shape Questionnaire (BSQ), Eating Attitudes T est (EAT), Beck n Depression Invetttory (BDI), Dutch Eating Behaviour Questionnaire (DEBQ) restrained eating item (r), emotional eating item (e) and x external eating item (x), Eating Inventory (EI), disinhihition item (d), cognitive re.rtraint item (cr) and hunger ilent (h) ~ 0 r BMI BSQ EAT BDI DEBQ:r DEBQ:e DEBQ:x El:d El:cr D EL:h -~ m Whole sample 25-2 t 0-8 92-8 f 5-8 15 •5 f 1-5 11-4f 1-2 2-74f0-14 3-32t0-16 3•46t0-11 9•4 + 0-6 7-3 t 0-7 D 7-9 1 0-6 d Secret eaters 26-6 f 1-4 116-4 f 8-5 20•5 t 2-6 147f2-2 2-99f0-18 3•87f0-13 3-74f0-13 11-9 f 0-8 9-0 t 1-I 9-9 t 0 7 Non-secret catcrs 24-3 t 1•0 77-7 t 6-6 12- I t 1•6 9•I f I-2 2•57t0-19 2•92±0-23 3-26±0-15 7-5 f 0-8 6-0 f 0-9 6•5 t 0-8 ~ F(1,42) 3-46 9-10 812 3-56 1-72 10-38 4-93 15-26 3-93 9-76 O p n.s. <0-004 <0-007 n.s. n.s. <0•002 <0-03 <0.0001 <0-05 <0•001 z M66E910Z

I I I I I I I I I 240 M. M. HETHERINGTON AND J. I, MACDIARMID Both dieters and non-dieters reported positive affect immediately before and during eating chocolate, but afterwards 63% of the non-dieters and only 24% of the dieters retained this positive mood [Figure 2(a)]. The distribution of positive, negative and neutral affect after consumption significantly differed between the groups. with more dieters than non-dieters reporting negative affect following consumption of chocolate. X'(2)= 7• 165, p<0-05. Secret and Non-secret Eaters Subjects were classified as secret eaters if they preferred to eat chocolate alone and in secret (n= 20) and the others as non-secret eaters. Secret eaters (13•3 f 2•0 bars per week) did not eat more chocolate than non-secret eaters (12•0t2•2 bars per week). nor were cravings different between the two groups (secret = 6•7 ± 1-2 times per week: non-secret = 5•9 ± 0•6 times per week). Fifty-five per cent of the secret eaters were currently dieting, compared to 33% of the non-secret eaters. Secret eaters scored significantly higher on a number of questionnaires (see Table 1). Secret eaters were significantly more dissatisfied with their bodies than non- secret eaters, they had higher scores on the EAT. on the emotional and external eating items of the DEBQ and on the disinhibition and cognitive restraint items of the El. Secret eaters demonstrated evidence of more disturbed eating patterns than non- secret eaters. There were no significant correlations between the amount of chocolate eaten and the questionnaire scores among the secret or non-secret eaters. Positive correlations were found within the secret eaters between the frequency of cravings and BMI, r(16) =0•51., p<0•05, and DEBQ (external eating), r(16)=0•65, p<0•01. The majority (80%) of secret eaters felt that their consumption of chocolate interfered in their life, this compared to only 30% of the non-secret eaters. Eighty per cent of the secret eaters described negative affect when cutting down or avoiding chocolate. Similarly, negative affect was reported by 57% of the non-secret eaters. Of the non-secret eaters, 23% said that they had never tried to cut down on chocolate, 100 80 ~ ~ 60 c v ~ a 40 L 20 0 before FtcuRF 2. Percentage of respondents reporting positive affect before, during and after consumption of chocolate. (a) Frequency of dieters and non-dietcrs reporting positive affect. •, dieter: G, non-dieter. (b) Frequency of secret and non-secret eaters reporting positive affect. , secret; ::, non-secret. during Positive mood after 100 20 (b) i i i before during after Positive mood

I I I I I t I I I I I I i I CHOCOLATE ADDICTION' -. 241 but all of the secret eaters reported that they had tried to avoid chocolate at some time. The percentage of subjects reporting positive affect before, during and after eating chocolate is shown in Figure 2(b). Seventy per cent of the secret eaters reported positive mood before eating, which increased to 90% during consumption and fell to 25% of the secret eaters after consumption. Fewer of the non-secret eaters exper- ienced positive mood before (43%) and during (83%) consumption of chocolate; however, more (60%) of the non-secret eaters maintained positive affect after consuming chocolate. Again, using a contingency table to assess the distribution of positive, negative and neutral affect after consumption, a significantly different distribution was observed. More secret eaters than non-secret eaters reported specifically negative affect following consumption of chocolate. y=(2)=6•537, p < 0•05. DISCUSSION It is clear from this preliminary study that self-identified "chocolate addicts" are not a homogeneous group, but have various patterns of consumption, cravings, and attitudes towards chocolate. Although the use of self-identified "chocoholics" may be problematic, our findings are consistent with those from other studies of addictions (Coventry & Brown, 1993; Fisher, 1993; Griffiths, 1991). Respondents in the present study consumed a larger than average amount of chocolate per week. The national average for the U.K. is 9 kg of chocolate confectionery each year per head of the population (Chocolate Market Review, 1991), which is about three bars of chocolate each week. Subjects in the present sample consumed almost four times this amount. According to market research data organized by region (British Market Research Bureau, 1990), Scotland has a higher percentage of heavy users (consumers regularly purchasing more than the national average of chocolate) than other regions in the U.K. None the less, the majority of subjects in this study believed that their intake was excessive and half of the subjects felt that chocolate consumption interfered with their life in some way, particularly for those who ate in secret. Perceptions of excessive consumption and interference are central to our understanding of chemical dependence (APA, 1987), providing some evidence of addictive features. Another characteristic of addictions shared by our subjects was the experience of cravings. Frequency of cravings (six times per week) for chocolate were not, however, correlated to intake of chocolate, thus, consumption of chocolate was preceded by an identified craving only part of the time. That intake and cravings may be distinct entities is confirmed in studies of smoking (Shiffman, 1984) and alcohol abuse (Edwards & Gross, 1976). Since cravings may be associated with a period of abstinence (see Weingarten & Elston, 1990), it is possible that some of our subjects did not crave because they did not abstain. Most of the present sample were not currently attempting to avoid or reduce intake. Craving frequency was associated with a tendency to eat for emotional reasons. Thus, cravings for chocolate may be driven by a desire to obtain the rewarding consequences of consumption, rather than to avoid negative consequences of abstinence (Stewart et al., 1984; Wise, 1988). Most subjects identified themselves as "chocolate addicts" because they had difficulty resisting chocolate, either during times of abstinence or when trying to limit I

I I I i I I I I I I I I I i 242 M. M. HETHERINGTON AND J I MACDIARMID intake once eating had been initiated. Rogers (1993) has labelled chocolate "a very moreish food", referring to a type of food which is very palatable and "and difficult to resist". Subjects in the present study noted that boxes of chocolates were especially irresistible, that they could not just have a few chocolates, but had to finish the entire box. Overall. participants attributed the addictiveness of chocolate to orosensory features of the food (its taste, smell and texture) rather than to psychoactive ingredients. This is consistent with a positive-incentive model of chocolate addiction and with Rozin et al. (1991) who found little evidence of a biochemical link to chocolate liking or addiction. Internal and external cues were equally likely to stimulate consumption of chocolate in this sample. Few subjects identified a particular pattern of consumption. Of those who did, the pattern of eating tended to be in the late evening which agrees with findings by Hill et al. (1991) that cravers tended to report cravings in the evening. Also, bulimics tend to binge most frequently in the evening (Mitchell et al., 1991). Of the others who recognized a pattern of eating chocolate, this tended to be after a meal. This agrees with the description by Douglas and Nicod (1974) of a traditional British meal where something sweet is eaten to signal its conclusion. In support of previous findings (Rozin et al., 1991), a high percentage of female participants reported an increase in cravings for chocolate premenstrually. In contrast, Weingarten and Elston (1991) found a smaller percentage of women who reported a link between general food cravings and the menstrual cycle. However, they also found that those women who did report a link tended to crave chocolate in particular. Of the respondents who had dieted at some time, just over half reported that dieting increased craving frequency; however, it is not known whether this was related to increased hunger (effects of abstinence) or increased appeal of chocolate when it had been excluded from the diet (a "forbidden food"; see Kales. 1990). Just as eating disordered patients describe certain foods as "dangerous" or as inducing guilt (Sunday et al., 1992), some of the subjects in the present study referred to chocolate as "dangerous". Several subjects preferred to minimize temptation by not keeping chocolate in the home. Participants reported ambivalent attitudes towards chocolate. Overall, most gave a positive appraisal of chocolate, but the description "fattening" was selected by a significant proportion of the participants. This ambivalence has been discussed by Rogers (1993) who has identified conflicting influences on chocolate intake. Choco- late was originally regarded as a luxury item (Barthel, 1989) and is used as a treat, gift or reward. However, nutritional wisdom dictates that as a society, intake of fat and sugar (specifically in confectionery and soft drinks) should be reduced (NACNE, 1983). Thus, chocolate is likely to produce conflict and ambivalence. Amount of chocolate consumed was correlated with the disinhibition item of the Eating Inventory, suggesting a link between intake and the extent to which eating may be influenced by disinhibitors (e.g. palatable foods, emotional and social factors). Hill et al. (1991) demonstrated a strong relationship between cravings and disinhibited eating, in particular individuals identified as "strong cravers" displayed a tendency to eat in response to negative mood states. The relationship between disinhibition and consumption in the present study may reflect both a general tendency for a disinhibited eating style and a particular vulnerability to the irresistibi- lity of chocolate. Thus, "chocolate addicts" may respond to a variety of disinhibiting