+ MODEL
ARTICLE IN PRESS
Journal of Affective Disorders xx (2006) xxx – xxx www.elsevier.com/locate/jad
Review
Mood state effects of chocolate Gordon Parker a, ⁎, Isabella Parker b , Heather Brotchie b a
School of Psychiatry, University of New South Wales, Australia b Black Dog Institute, Sydney, Australia
Received 22 September 2005; received in revised form 2 February 2006; accepted 3 February 2006
Abstract
Background: Chocolate consumption has long been associated with enjoyment and pleasure. Popular claims confer on chocolate the properties of being a stimulant, relaxant, euphoriant, aphrodisiac, tonic and antidepressant. The last claim stimulated this review. Method: We review chocolate's properties and the principal hypotheses addressing its claimed mood altering propensities. We distinguish between food craving and emotional eating, consider their psycho-physiological underpinnings, and examine the likely ‘ positioning’ of any effect of chocolate to each concept. Results: Chocolate can provide its own hedonistic reward by satisfying cravings but, when consumed as a comfort eating or emotional eating strategy, is more likely to be associated with prolongation rather than cessation of a dysphoric mood. Limitations: This review focuses primarily on clarifying the possibility that, for some people, chocolate consumption may act as an antidepressant self-medication strategy and the processes by which this may occur. Conclusions: Any mood benefits of chocolate consumption are ephemeral. © 2006 Elsevier B.V. All rights reserved. Keywords: Keywor ds: Chocolate; Chocolate consumption; Negative mood state; Depression; Atypical depression
Contents
1. 2.
3.
4.
Introduction . . . . . . . . . . . . . . . . Chocolate . . . . . . . . . . . . . . . . . . 2.1. Chocolate composition . . . . . . . 2.2. Chocolate craving . . . . . . . . . . 2.3. Psychoactive properties of chocolate Chocolate and neurotransmitter systems . . 3.1. Dopamine . . . . . . . . . . . . . . 3.2. Serotonin . . . . . . . . . . . . . . 3.3. Opioids . . . . . . . . . . . . . . . From pleasure to aversion: a PET study . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
0 0 0 0 0 0 0 0 0 0
⁎ Corresponding
author. Postal address: Black Dog Institute, Prince of Wales Hospital, Randwick, Sydney 2031, Australia. Tel.: +61 2 93824372; fax: +61 2 93824343.
[email protected] .edu.au ( (G. G. Parker). E-mail address: g.parker@unsw 0165-0327/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi: do i:10.1016/j.jad.2006.02.007 10.1016/j.jad.2006.02.007
JAD-03219; No of Pages 11
ARTICLE IN PRESS 2
G. Par Parker ker et al. / Jour Journal nal of Affec Affective tive Diso Disord rders ers xx (200 (2006) 6) xxx – xxx
5. 6. 7. 8.
Oro rose sen nso sory ry pro rope pert rtie iess of ch cho ocol olaate and th thee ple leas asur uree pri rin nci cip ple . . . . . A homeostatic hypothesis . . . . . . . . . . . . . . . . . . . . . . . . Emotional eating . . . . . . . . . . . . . . . . . . . . . . . . . . . . Food craving as addiction . . . . . . . . . . . . . . . . . . . . . . . . 8.1 .1.. Ch Cho oco cola late te cr craavi vin ng and em emo oti tio ona nall ea eati tin ng as as sep sepaara rate te phen eno omen enaa 9. An evolutionary perspective . . . . . . . . . . . . . . . . . . . . . . 10. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
If any man has drunk a little too deeply from the cup of physical pleasure; if he has spent too much time at his desk that should have been spent asleep; if his fine spirits have become temporaril temporarilyy dulled; if he finds the air too damp, the minutes too slow, and the atmosphere too heavy to withstand; if he is obsessed by a fixed idea which bars him from any freedom of thought: if he is any of these poor creatures, we say, lett hi le him m be gi give ven n a go good od pi pint nt of am ambe berr-fl flav avor ored ed chocola cho colate.. te....an ..and d marv marvels els will be perf performe ormed. d. — Anthelmee Brillat Anthelm Brillat-Savar -Savarin in (1755–1826). Chocola Choc olate te is che cheape aperr tha than n th thera erapy py and you do don't n't need an appointment. — Unknown.
It is commonly claimed that chocolate has the ca pacity to lift spirits, to create highs and make people feel good. In an earlier review of atypical depression and its cons constitu tituent ent fea feature ture of hype hyperpha rphagia gia (Par Parker ker et al.,, 2002 al. 2002), ), we not noted ed the ca capa pacit city y of ca carb rbohy ohydr drat ates es (including chocolate) to have a comforting effect and to al also so pr prom omot otee ‘fee feell good’ sensatio sensations ns thr through ough the release of multiple gut and brain peptides. Others have argued that carbohydrate craving (including chocolate craving) in atypical depression (Moller, (Moller, 1992) 1992) and in seas se asona onall af affe fect ctiv ivee dis disor order der (S (SAD) AD) (Wu Wurtm rtman an and Wurtman, 1989) 1989) is a form of self-medication and, in havi ha ving ng an im impac pactt on br brain ain neu neuro rotra trans nsmi mitte tters rs hav havee antidepressant antidep ressant benefit benefits. s. Publ Pu blic icat atio ion n of ou ourr ea earl rlie ierr pa pape perr (Pa Park rker er et al al., ., 2002)) le 2002 led d to co cons nsid ider erab able le me medi diaa at atte tent ntio ion n an and d inquiri inq uiries es both from those with a moo mood d dis disord order er and thee gen th gener eral al pub public lic as to whe whethe therr cho choco colat latee ac actua tuall lly y wass an an wa antid tidep epre ress ssant ant,, so en encou coura ragin ging g a lit liter eratu ature re review. Use of the single word ‘chocolate’ in search engines eng ines ide identif ntified ied an ext extraor raordina dinarily rily bro broad ad lite literat rature ure warran war ranting ting inte integra gration tion and enc encoura ouraging ging the cur curren rent t revi re view ew.. Wh Whil ilee we ha had d an ag agno nost stic ic vi view ew ab abou out t choco ch ocolat latee hav havin ing g any ant antide idepre press ssant ant pr prope opert rties ies,, it
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
0 0 0 0 0 0 0 0 0
became apparent that there were many other topic compon com ponen ents ts of eq equal ual im impo port rtanc ance, e, re resu sulti lting ng in thi thiss review rev iew goin going g bey beyond ond the init initial ial sim simple ple obje objectiv ctive. e. As noted not ed by one jo jour urnal nal as asse sess ssor or,, as aspe pects cts ha have ve so some me signific sign ificance ance for our unde underst rstandi anding ng of the biol biology ogy of depre dep ress ssion ion,, an and d fo forr the con conce cept pt of ‘atypical depres depres-sion’ in particular. In this overview, we consider why people crave and eatt ch ea choc ocol olate ate,, pa part rtic icula ularl rly y in th thee co conte ntext xt of fo food od cravings and emotional eating, and chocolate's mood state effects. In the many comprehensive reviews (e.g., Bruinsma and Taren, 1999; Christensen, 1993; Ganley, 1989; Mercer and Holder, 1997; Ottley, 2000; Pelchat, 2002; Rogers and Smit, 2000; Weingarten and Elston, 1990), 1990 ), exp explan lanati ations ons of cho chocol colate ate and car carboh bohydr ydrate ate cravin cra vings gs var vary y wide widely ly and inc include lude sel self-m f-medi edicat cation, ion, homeostatic homeosta tic correc correction, tion, hedonic experience, addiction to psy psycho choact active ive sub substa stances nces and emo emotion tional al eat eating ing theories, theorie s, but it may be possible to integra integrate te the dispar disparate ate theorie theo ries. s. We will arg argue ue that cho chocola colate te ‘craving’ and chocolate ‘emotio emotional nal eating’ are are two se separ parate ate phe phe-nomena — althou although gh the they y ca can n co co-e -exis xistt in the sa same me individual — and and th that at an any y mo mood od st stat atee ef efffec ects ts of chocolate are as ephemeral as holding a chocolate in one's mouth. 2. Chocolate
2.1. Chocolate composition
Chocolate Chocola te is man manufa ufactur ctured ed fro from m coc cocoa oa mas masss (th (thee base product produced by the processing of the cocoa bean), cocoa butter (the natural fat from the cocoa co coa bean) b ean) and added sugar. Cocoa butter melts at room temperatur at uree to pr prov ovid idee th thee cr crea eamy my “mel eltt in th thee mo mout uth h” sensatio sens ation. n. Eur Europea opeans ns adde added d sug sugar ar to appe appeal al to thei their r palate when chocolate was introduced from America. Dark Dar k choc chocolat olatee con contain tainss thes thesee thre threee ele elemen ments, ts, mil milk k chocolate contains extra milk solids and fats, and white chocolate is akin to milk chocolate without the cocoa base.
ARTICLE IN PRESS G. Parker et al. / Journal of Affective Disorders xx (2006) xxx – xxx
3
2.2. Chocolate craving
3.1. Dopamine
Chocolate is the most commonly craved food and, for Chocolate for most chocolate craver cravers, s, non-chocolate substitutes are inadequate inadequa te (Weingarten and Elston, 1991). 1991). Chocolate is not a natural product, and thus its appeal depends on its individual individu al constitu constituents ents and their unique combina combination. tion. Chocolate is sweet, raising the possibility of confusion between chocolate craving cravi ng and sweet craving, cravin g, but it also contains fat (Rozin (Rozin et al., 1991). 1991). Similarly, other foods that are commonly craved (e.g., ice cream, doughnuts, cakes, biscuits) biscuits) also taste sweet and might be misjudge misjudged d as rich in sugar, but most of their calories are provided by fat. Drewnowski et al. (1992) have therefore suggested that the term ‘carbohy carbohydrate drate craving’ is a misnomer when applied to such foods.
The mesolimbic dopaminergic system is involved in mostt dru mos drugs gs of mis misuse use exc except ept benz benzodia odiazepi zepines nes,, with increased levels of dopamine in the nucleus accumbens central to mediating the rewards of positive reinforcement (Koob (Koob and Le Moal, 2001). 2001). The positive positive rei reinnforcing effects of stimulation are remembered for long periods, being underpinned by the activation of brain dopamine systems. Rewards from food and from drugs may depend on similar substrates for motivational processes (Pelchat, (Pelchat, 2002), 2002), with the role of dopamine recognized as critical in anticipation as well as in withdrawal (Lingford-Hughes and Nutt, 2003). 2003). It has been argued that, in modern society, society, humans are driven to eat less by interna inte rnall ener energy gy def deficit icitss and mor moree by the anti anticip cipator atory y pleasure of eating (Pinel, 1990). 1990). The contribution of the dopaminergic dopamin ergic system to chocolate craving craving and eating is, however howe ver,, like likely ly to be gene general ral rather than chocolatechocolatespecific.
2.3. Psychoactive properties of chocolate
Becaus Beca usee ch choco ocolat latee cr crav aving ing ha hass so some me fe feat ature uress of addic ad dictio tion, n, at attem tempts pts ha have ve bee been n ma made de to id ident entif ify y an any y psychoactive ingredients. Several candidates have been identi ide ntifie fied d (th (thee bio biogen genic ic sti stimul mulant ant ami amines nes caf caffei feine, ne, theobro the obromin mine, e, tyr tyrami amine ne and phe phenyle nylethyl thylami amine) ne),, but their concentrations are too low to have a significant psychoactive effect and they are also present in higher concentr conc entrati ations ons in non non-cr -crave aved d foo foods ds (BNF Nutr Nutritio ition n Bulleti Bul letin, n, 1998 1998;; Heth Hetherin erington gton and Mac Macdiar diarmid, mid, 199 1993; 3; Hurst et al., 1982; Rozin et al., 1991; Weingarten and Elston, 1991; Ottley, 2000). 2000). Comparisons Compar isons of subject subjectss ingestin ingesting g milk chocolate, dark da rk ch choc ocola olate te,, whi white te ch choco ocolat latee an and d co cocoa coa pow powder der (pow (p owder dered ed coc cocoa oa ma mass ss wi with th so some me co coco coaa bu butte tterr extracted) have demonstrated that milk chocolate is most preferred. If psychoactive substances were involved, then cocoa powder should equally satisfy craving and dark chocolate should be the most preferred (Michener (Michener and Rozin, 1994). 1994). Chocola Choc olate te cont contains ains two ana analogu logues es of anan anandam damine ine similar to the cannabinoid responsible for euphoria from cannabis. However, any association with pleasure from chocolatee is likely to be indirect as the analogues inhibit chocolat breakdown of endogenously produced anandamine (di (di Tomaso et al., 1996). 1996). 3. Chocolate and neurotransmitter systems
Chocolate may intera Chocolate interact ct with a number of neurotra neurotransnsmitter mit ter sys system temss (i (incl ncludi uding ng dop dopam amine ine,, ser seroto otoni nin n and endor en dorphi phins ns)) tha thatt co contr ntribu ibute te to ap appe petit tite, e, re rewa ward rd an and d mood mo od re regu gulat latio ion. n. So Some me re relev levan antt lit liter eratu ature re is no now w considered.
3.2. Serotonin
The ma The many ny ro role less of se sero roto toni nin n in incl clud udee in infl flue uenc ncee over sleep, appetite, impulse control and mood elevation. A number of specific serotonergic contributions to ch choco ocolat latee ea eatin ting g hav havee bee been n pr prop opose osed, d, an and d a lin link k between serotonin, mood and craving has become part of the ‘folklore’ of the psychology of eating (Rogers (Rogers and Smit, 2000). 2000). A popular hypothesis is that chocolate or carbohydrate craving addresses serotonin deficiency cie ncy,, par partic ticula ularly rly in dep depres ressed sed ind indivi ividua duals ls who attempt ‘self-medication’. Studies of seasonal affective disorder and atypical depression provide some support forr th fo this is hy hypo poth thes esis is (Mo Molle llerr, 19 1992 92;; Wur urtm tman an an and d Wurtman, Wurt man, 1989 1989). ). The latter authors suggested that serotonin acts through a biofeedback mechanism to regulate ula te car carboh bohydr ydrate ate con consum sumpti ption. on. The ami amino no aci acid d tryptophan tryptop han normal normally ly circula circulates tes in low concentr concentrations, ations, but when carbohydrates enter the bloodstream, they stimulate insulin production and facilitate the uptake of most mo st am amin ino o aci acids ds,, ex excl clud udin ing g tr trypt yptoph ophan, an, int into o the tissue tis sues. s. The pro proport portion ion of cir circula culating ting tryp tryptoph tophan an increase cre asess rel relati ative ve to oth other er ami amino no acid acids, s, fac facilita ilitating ting its passage across the blood– brain barrier, where it is converted into serotonin. Variations in serotonin levels in tur turn n inf influen luence ce car carbohy bohydra drate te inta intake. ke. Wurtman Wurtman and Wurtman Wurt man (1989) hypot hypothes hesize ized d tha that, t, in so some me cy cycli clicc conditions conditio ns such as seasonal affective disorder, disorder, there is a disturb dis turbanc ancee of the fee feedbac dback k mec mechani hanism sm bet between ween serotonin and carbohydrate intake in which the desire for carbohydrates persists. Additional consumption of
ARTICLE IN PRESS 4
G. Par Parker ker et al. / Jour Journal nal of Affec Affective tive Diso Disord rders ers xx (200 (2006) 6) xxx – xxx
carbohydrates may be necessary to lift serotonin levels and thus mood. If the ser seroton otonin in hypo hypothes thesis is for car carboh bohydra ydrate te and chocolate craving was valid and self-medication effectively tiv ely cor correc rects ts a ser seroton otonin in def deficie iciency ncy,, then cra craving ving would be a plausible biological response to ‘ depression’. However How ever,, link linkss betw between een depr depressi ession on and appe appetite tite are complex. comp lex. In ‘clini clinical cal depr depressi ession on’, par partic ticula ularl rly y the melancholic sub-type, many patients report decreased appetite and recovery is associated with restoration of appetite and food intake, but the converse pattern is often found in the non-melancholic disorders (Parker (Parker et al., 1994). 1994). For example, carbohydrate craving can occur in the ‘atypical’ depressive sub-type, with hyperphagia and weight gain being included in DSM-IV diagnostic criteria. It is also a common response to those with nonclinica cli nicall depr depress ession. ion. For exam example, ple, pati patients ents rou routine tinely ly consult con sulting ing the their ir pri primar mary y car caree phys physicia ician n (Par Parker ker and Brown, 1982) 1982) were asked to rate their likely behavioural responses to two depressogenic stimuli (a break-up of an import imp ortant ant rel relatio ationshi nship, p, and inc increas reased ed cri critic ticism ism fro from m some so meon onee im impo port rtan antt to the them) m).. Am Amon ong g th thos osee wh who o predicted they would experience a depressed mood, 52% reported they would be likely to eat more and only 11% reported that they would eat less, and more thought that increased eating would be effective (33%) rather than ineffective (18%). Thus, appetite loss may be a prototypic symptom of some depressive disorders, while appe ap peti tite te in incr crea ease se or fo food od cr crav avin ing g ma may y mo more re be a prototypic comfort eating response to certain expressions of depression. An examination of the literature (Hammersley (Hammersley and Reid, 1997); 1997); Rogers, Rogers, 1995; Young, 1991) 1991) does indicate that the serotonin account is flawed. First, Benton First, Benton and Donohoe (1999) (1999) reviewed 30 studies of blood amino acid aci d pro profil files es tak taken en af after ter sub subjec jects ts con consum sumed ed mea meals ls varying in their carbohydrate content. Only when the protein component of the meal was less than t han 2% did the resulting amino acid profile favor a rise in serotonin levels in the brain. Chocolate contains 5% of its calorie content as protein (and ice cream 8%), which would be sufficient to negate any serotonin effect. Second, even extreme dietary manipulations of tryptophan result in physiological changes that are too slow to account for mood moo d eff effects ects that are described described duri during ng or soo soon n aft after er eating (Young (Young et al., 1985). 1985). Third, Ottley (2000) has demonstrated that fat slows the absorption of carbohydrate, and commonly craved foods such as chocolate, ice cream and biscuits generally have a high fat content. Fourth, mood changes after carbohydrate ingestion have been reported even in the absence of changes in serotonin synthesis and release (Wurtman (Wurtman and Wurtman,
1989; Rosenthal et al., 1989; Cabellero et al., 1988), 1988), sugge sug gesti sting ng tha thatt oth other er or add addit ition ional al me mecha chanis nisms ms ar aree involved. 3.3. Opioids
Heroin addicts experience experience sweet cravings if heroin is not available (Weiss, (Weiss, 1982), 1982), suggesting an overlap in the relevant neural systems (W (Weingarte eingarten n and Elston, 1991). 1991). In add additio ition, n, cond conditio itions ns such as pre pregnan gnancy cy,, men menstr strual ual cycling, cycl ing, alcoholism alcoholism and eati eating ng diso disorder rderss that are associated with craving involve altered endogenous opioid peptide (EOP) levels (Mercer and Holder, 1997). 1997). On such grounds, Drewnowski (1992) argued that carbohydrate craving is more strongly linked with the opioid rather than the serotonergic system, while while Mercer and Holder (1997) (1997) have proposed an opioidergic theory of food cravings. Research with vertebrates has shown a relatio rel ationshi nship p bet between ween cent central ral EOP acti activit vity y and foo food d intake int ake,, and ar argue gued d tha thatt fo food od cr cravi aving ngss me medi diate ate the opioid-ingestion opioid-in gestion link. Thee op Th opio ioid id sy syst stem em ap appe pear arss to pl play ay a ro role le in th thee palatability of preferred foods (Si (Si et al., 1986), 1986), releasing opioids such as P-endorphins as food is eaten which could cou ld en enhan hance ce the pl plea easur suree of ea eati ting ng (Ottley Ottley,, 2000 2000). ). Opioids released in response to ingestion of sweet and other oth er pa pala latab table le fo foods ods (Bl Blas ass, s, 198 1986; 6; Fu Fulle llert rton on et al al., ., 1985)) can increase central opioidergic activity, in turn 1985 stimulating the immediate release of beta-endorphin in the hypothalamus and producing an analgesic effect that is naltrexone-reversible (Blass, (Blass, 1986). 1986). Similarly, Similarly, sucrose immediately reduces crying in infants during a painful hospital hosp ital pro proced cedure ure sugg suggesti esting ng that tas taste, te, rat rather her tha than n post-ingestive factors, provokes release of EOP (Blass and Hoffm Hoffmeyer eyer,, 1991 1991). ). Exogenous Exogeno us opioids generally increase food intake. The mu receptor is central in opiate addiction, but kappa receptor stimulation reduces dopamine function in the nucleus accumbens and may lead to dysphoria (Ling(Lingford-H for d-Hughe ughess and Nutt Nutt,, 2003 2003). ). In Injec jecti tions ons of mu and kapp ka ppaa ag agon onis ists ts ma may y ha have ve a bi biph phas asic ic ef effe fect ct,, fi firs rst t increas incr easing ing but then dec decrea reasing sing feeding (Gulati (Gulati et al., 1991), 1991 ), while low doses may increase consumption and high doses decrease consumption (Triscari (Triscari et al., 1989; Ramarao Ramara o and Bhargava, 1989). 1989). Together, these findings indicate indi cate inte interac raction tionss with othe otherr appe appetite tite-re -regula gulating ting processes. It has been suggested that EOP enhances dopaminergic activity in the mesolimbic pathways to alter the reward value of food (Cooper (Cooper and Kirkman, 1993). 1993 ). Although the precise role of EOP in regulating food intake has not yet been established, established, it is clear it acts in concert with other neurochemicals such as serotonin,
ARTICLE IN PRESS G. Parker et al. / Journal of Affective Disorders xx (2006) xxx – xxx
dopamine dopam ine and cho cholec lecyst ystoki okinin nin tha thatt co contr ntrol ol eat eating ing (Cooper and Higgs, 1994). 1994). 4. From pleasure to aversion: a PET study
Small et al. (2001) undertook (2001) undertook successive PET scans on volunteers who ate chocolate beyond satiety. Their researc res earch h mode modell hel held d the sensory sensory stim stimulus ulus and act of eating as constant, while motivation to eat and reward valuee wer valu weree mod modifie ified d by fee feeding ding.. Dif Differ fering ing str structu uctural ral groups gro ups wer weree sele selectiv ctively ely rec recruit ruited ed depe dependin nding g whet whether her chocol cho colate ate was eat eaten en wit with h ple pleasu asure re and mot motiva ivatio tion n (subcallosal (subcal losal region, caudomedial orbitof orbitofrontal rontal cortex, insula/ ins ula/oper operculu culum, m, str striatu iatum m and mid midbra brain) in) as aga against inst when whe n sat satiate iated d (pa (parahi rahippoc ppocamp ampal al gyr gyrus, us, caud caudolat olateral eral orbitofrontal cortex and prefrontal regions). The authors concluded that results argued for two differing motivational tion al sys system temss bein being g invo involved lved — “one orchest orchestrating rating approach and another avoidance behaviours”. 5. Or Oros osen enso sory ry pr prop oper erti ties es of ch choc ocola olate te an and d th thee pleasure principle
The or The oros osens ensor ory y pr prope opert rtie iess of fo food od,, me media diated ted by palatability, have a significant influence over eating. Moment-to-moment, eating is controlled predominantly by the orosensory effects that provide positive feedback, feedb ack, and the post-ingestive effects which provide negative feedback. Effects in the mouth are stimulatory, while the entry entr y of foo food d int into o the stom stomach ach is inhi inhibito bitory ry.. Pos Positi itive ve feedback is ‘the stimulation of eating by eating’ and its strength is influenced by palatability (Yeomans, (Yeomans, 1996), 1996), which is defined by the hedonic or affective response to the taste, flavor, aroma and texture of food (Rogers, ( Rogers, 1990). 1990 ). Pal Palatab atabilit ility y is dete determi rmined ned by inna innate te res respons ponses es such as inborn taste preference for sweets, fat and salt but a dislike for bitter tastes, and by learning (for example exa mple,, eff effect ectss on moo mood). d). Pos Post-i t-inges ngestive tive eff effects ects increase gradually with the amount eaten, and eating stops when they outweigh the orosensory effects. However, this satiety mechanism is not infallible, as commonly experienced by chocolate cravers attempting to impose self-r sel f-restr estrain aint. t. Chal Challeng lenges es of sel self-r f-restr estrain aintt rel relate ate to a ‘moreishness ’ quality (i.e., a desire for more), which is a function of palatability. Craving is experienced during abstinen abst inence, ce, but ‘moreishness ’ is exp experi erienc enced ed wh while ile eating. Ther Th eree is a wel welll-sup suppor ported ted vie view w (Weingart eingarten en and Elston, 1990; Michener and Rozin, 1994; Rodin et al., 1991;; Heth 1991 Hetheri eringto ngton n and Macd Macdiarm iarmid, id, 1993 1993)) that the orosensory properties of chocolate and the desire for senso se nsory ry gra gratif tifica icatio tion n ar aree suf suffic ficien ientt to exp explai lain n the
5
motivation for ingesting chocolate, and that the sight and smell of chocolate is sufficient to trigger cravings. Chocolate has a high hedonic rating, based on the set of extremely extrem ely appealing sensory characteristics characteristics (as previously noted) and with an attractive aroma and unique flavour (Drewnowski (Drewnowski and Greenwood, 1983). 1983). Its appeal can be argued on the basis of anticipatory pleasure rather than tha n ref reflect lecting ing inte interna rnall ener energy gy def deficit icitss (Pinel, 1990). 1990). Satiety is somewhat taste-specific. taste-specific. As satiety occurs for one food, the positive-incentive value and palatability of that food declines, but if a new food is offered it can promote further eating (Pinel, 1990). 1990). Experiments with rats have demonstrated that increased concentrations of sucrose result in a rise in the ‘hedonic breakp breakpoint oint ’ after which whi ch ing ingest estion ion dec decrea reases ses.. How Howev ever er,, the hed hedoni onicc responses to sweet and fat complexes of ‘real foods’ are interdependent, with human studies demonstrating thatt com tha combini bining ng suga sugarr with dairy cre cream am or ful full-cr l-cream eam milk increases the hedonic rating of each (Drewnowski (Drewnowski and Greenwood, 1983). 1983). Heightened responsiveness to increased palatability is often cited as a major factor in the development of obesity and even rats on a highly palatable high-fat or ‘supermarket ’ diet will overeat to thee po th poin intt of ob obes esit ity y (Drew Drewnows nowski ki and Gre Greenwo enwood, od, 1983). 1983 ). Chocolate provides preferred tastes and texture that have innate appeal and the combinations increase the hedonic rating, challenging the satiety ceiling that might otherwise apply. The com compos posite ite oros orosenso ensory ry pro proper perties ties of choc chocola olate te outweigh more simple explanations explanations of its role in appetite and satiety. For instance, if a caloric deficit motivates choc ch ocol olat atee cr crav aving ing,, th then en mi milk lk ch choc ocol olat atee an and d wh whit itee chocolate should have equal appeal, but they do not. If psychoactive substances or magnesium deficit are the basis of craving, then milk chocolate and unsweetened cocoa powder should have equal appeal, and other foods contain con taining ing the these se sub substan stances ces in high higher er conc concentr entrati ations ons should be craved, but they are not. If the appeal is the unique sensory combination of chocolate, then chocolate is the only way to satisfy that craving (Michener (Michener and Rozin, 1994). 1994). 6. A homeostatic hypothesis
One view regards craving for chocolate and carbohydrates as a homeostatic response to dietary deficiencies ci es.. For ex exam ample ple,, ch choco ocola late te cr crav aving ing has als also o bee been n proposed as a response to a magnesium deficit. However, while a magnesium supplement has been demonstrated to reduce chocolate craving, certain foods high in magnesium are neither craved nor satisfy craving for chocolate (Michener (Michener and Rozin, 1994). 1994). Further, several
ARTICLE IN PRESS 6
G. Par Parker ker et al. / Jour Journal nal of Affec Affective tive Diso Disord rders ers xx (200 (2006) 6) xxx – xxx
studies have failed to establish that dieters do experience any greater craving than non-dieters (Wei non-dieters (Weingarten ngarten and Elston, 1991; Cohen et al., 1987; Lappalainen et al., 1990) instead, 1990) instead, they tend to experience less craving. So far, no support for the homeostatic hypothesis can be provided. 7. Emotional eating
Lyman (1982) (1982) has observed that food preference is altered across a range of mood states, with preference for ‘ junk food’ increasing during negative mood states and an d pr pref efer eren ence ce fo forr he heal althy thy fo food odss inc incre reas asing ing du duri ring ng positive mood states. Reviews by by Ganley (1989) and Christensen Chris tensen (1993) detail detail nume numerou rouss ass associ ociatio ations ns between multiple mood states (stress, boredom, depression, loneliness, social self-doubt, discord, frustration, ange an gerr and an anxie xiety) ty) and co copin ping g by em emoti otiona onall ea eati ting ng (Wein eintra traub ub and Aron Aronson son,, 196 1969; 9; Taylo aylorr et al. al.,, 1996 1996;; Hudson and Williams, 1981; Hill et al., 1991; Rozin et al., 1991). 1991). A review by Ganley (1989) (1989) indicated that stress-associated eating (i.e., emotional eating) is more common in those who are overweight or obese but that mixed results in the research means that the direction of the ass associa ociation tion rem remains ains uncl unclear ear.. Cli Clinica nicall obs observa ervation tion suggests that, while weight gain caused by emotional eating may exacerbate aversive mood states, the latter may ma y tr trigg igger er a cyc cycle le of fu furt rther her em emoti otiona onall ea eatin ting g and continu con tinued ed weig weight ht gain gain.. Suc Such h ass associa ociation tionss are high highly ly dependent on the current emotional state of the subjects. For example, a study by Ruderman by Ruderman (1983) demonstrated (1983) demonstrated thatt obes tha obesee wom women en ate sign signific ificantl antly y les lesss whe when n high highly ly anxious than when mildly anxious. However, such links do not support the hypothesis that carbohydrate craving is necessarily an attempt to alleviate aversive mood states, or that all obesity is the result of emotional eating, or that stress has a unilateral rela re lati tion onsh ship ip wi with th fo food od in inta take ke.. In fa fact ct,, th thee na natu tura rall tendenc ten dency y duri during ng str stress ess is for decreased decreased foo food d inta intake, ke, reflecting activation of the sympathetic nervous system (Schachte Schachterr et al., 1968), 1968), whereby blood is diverted from the digestive system and other systems not required for defense. A study by Stone and Brownell (1994) (1994) sup ported this proposition, showing show ing that subjects were more likely to eat less in response to stress, and with the decrease in proportion to the severity of stress. Although eating when stressed or anxious is inconsistent with the intrinsic physiological stress response, it occurs in real life and is supported empirically with stress thought to disrupt restraint and post-ingestional satiety feedback. Meisel et al. (1990) showed (1990) showed a marked increase in body weight in female Syrian hamsters when socially stressed
by being caged in groups compared those housed individually and with the additional finding of greatly enlarged adrenal glands. Morley glands. Morley et al. (1983) showed (1983) showed an increase in sucrose ingestion in response to stress when rats were subjected to a ‘ pinch-tail test ’ and suggested that th at en endo doge geno nous us op opio ioid id pe pept ptid ides es as we well ll as ot othe her r neuropeptides played a role in the central regulation of stress str ess-in -induc duced ed ea eatin ting. g. As fur furthe therr evi evide dence nce of the variable associations between appetite and stress, animal model mod el and hum human an vol volunt unteer eer stu studie diess of dep depres ressio sion n demonstrate demonstr ate decrea decreased sed respons responsiveness iveness to reward (sucrose solution) under conditions of chronic mild stress, but the opposite when the animals are provided with sweeter pellets (Willner (Willner et al., 1998). 1998). The latter effect could coul d ref reflect lect intr intrinsi insicc reg regulat ulatory ory mec mechani hanisms sms bein being g compro com promis mised ed whe when n con concen centra tratio tions ns of cer certai tain n foo food d compone com ponents nts are markedly markedly high higher er than in nat nature ure.. The high hedonic value of fat and carbohydrate concentrations in certain foods, including chocolate, may overstimu sti mulat latee pos positi itive ve fe feedb edback ack and and,, tog togeth ether er wit with h the learned lear ned exp expecta ectation tionss of all allevia eviation tion of aver aversive sive moo mood, d, serve to override normal post-ingestive satiety feedback mechanisms. Any attempt to study the effects on mood following ingestion of carbohydrate is confounded by a range of methodological variables. These include time in relation to ea eatin ting g (i (i.e .e., ., dur durin ing, g, im immed mediat iatel ely y aft after er,, or lat later er), ), whether whet her subj subjects ects are cra craver verss or nonnon-cra craver vers, s, and the actual actu al moo mood d sta state te (e. (e.g., g., dep depres ression sion,, anxi anxiety ety,, ennu ennui). i). Several studies illustrate these difficulties. In one study, non-carbohydrate cravers and lean subjects experienced an increase in fatigue and sleepiness after a carbohydrate-ri drat e-rich ch snac snack, k, while obes obesee crav cravers ers expe experien rienced ced improvement in mood (Lieberman (Lieberman et al., 1986; Spring et al., 1982/3). 1982/3). While Wurtman and Wurtman (1989) showed show ed that car carbohy bohydra drate te con consum sumptio ption n pro produce duced d an initial init ial tem tempor porary ary rel relief ief fro from m dys dysphor phoria, ia, Johnson Johnson and Larson (1982) found (1982) found marked post-binge dysphoria following occasional and transient relief. Thayer (1987) demonstrated demonstr ated a longerlonger-term term effect of reduce reduced d energy energy,, creating creatin g a cycle that contributed to the developm development ent and mainten mai ntenance ance of depr depress ession. ion. In a ret retrosp rospecti ective ve stu study dy,, Hetherin Heth erington gton and Mac Macdiar diarmid mid (199 (1993) 3) reporte reported d that any mood improvement was during consumption consumption only, with negative moods returning immediately after eating.. In a la ing later ter study study, Macdiarmid Macdiarmid and Hetherin Hetherington gton (1995) reported (1995) reported that ‘chocolate addicts’ were more de pressed than controls, and their negative mood did not improve impr ove aft after er choc chocolat olatee inta intake. ke. In thos thosee two stu studies dies,, chocolate eating resulted in a slight increase in contentment whilst eating chocolate, but the authors suggest ge sted ed th that at th this is re refl flec ecte ted d sa sati tiet ety y ra rath ther er th than an an
ARTICLE IN PRESS G. Parker et al. / Journal of Affective Disorders xx (2006) xxx – xxx
improv impr oveme ement nt in mo mood, od, as ra rati tings ngs of dep depre ress ssio ion n or feeling relaxed were not affected and, following consumption, ratings of guilt were increased. In their review, Roger Rogerss an and d Sm Smit it (2 (200 000) 0) sugge suggest st tha thatt th this is inducement of guilt in the eating of chocolate stems from fro m the cul cultura turall amb ambival ivalence ence sur surroun rounding ding choc chocolat olatee consumption. As a corollary, other studies have demonstrated that resisting resis ting the craving produced a more positive emotional tone (Hill (Hill et al., 1991) 1991) and that amelioration of mood was ac achie hieved ved by eli elimi mina natin ting g re refi fined ned suc sucro rose se and caffeine (Chr (Christ istens ensen en et al. al.,, 1985 1985;; Chri Christe stense nsen n et al. al.,, 1989; Kreitsch et al., 1988; Christensen and Burrows, 1990), 1990 ), leading to the suggestion suggestion that continued ingestion ingestion of carbohydrates may contribute to the maintenance of dysp dy spho hori ricc mo mood od.. Fi Fina nall lly y, a re revi view ew by Reid Reid an and d Hammersley Hammer sley (1999) (1999) examin examining ing all stud studies ies on huma human n subjects since 1983 dealing with the effects of carbohydrates hydr ates on aro arousal usal,, fou found nd half the stud studies ies rep reporte orted d some effect on alertness after consumption of a carbohydrate snack with individuals feeling relaxed or sleepy, but with no specific effect on mood. Various psychological theories of emotional eating have ha ve be been en pr propo opose sed. d. On Onee ar argu gued ed fo forr fa failu ilure re in sa sattisfaction of basic needs (Timmerman (Timmerman and Acton, 2001), 2001), proposing that when individuals perceive life situations to be stressful they draw on self-care resources which may be inte interna rnall (e. (e.g., g., sel self-e f-estee steem) m) or exte externa rnall (e. (e.g., g., social support). support). If needs remain unmet over a prolonge prolonged d period of time, then a resource deficit exists and food consump con sumption tion may be use used d to supp suppleme lement nt the def deficit icit.. Another Anot her the theory ory is that emotional emotional eati eating ng res results ults from confusion and apprehension in recognizing and accurately responding to emotional and visceral states related to hunger and satiety (Van (Van Strien, 2000). 2000). A composite ‘escape theory’ of eating (Heatherton (Heatherton and Baumeister, 1991)) ar 1991 argu gues es th that at ea eati ting ng is an ‘es esca cape pe fr from om se self lf ’. Heatherton et al. (1992) hypothesized (1992) hypothesized that distress, by threatening self-image, acts as a trigger for disinhibition, motivating motivati ng the individual to escape from self-awareness self-awareness whic wh ich h is av aver ersi sive ve wh when en th they ey en enco coun unte terr ne nega gati tive ve information about self. They avoid meaningful thought and instead, a cognitive shift redirects attention away from unpleasant thoughts about self and towards food cues cu es in th thee en envir vironm onment ent.. Th This is als also o ha hass the ef effe fect ct of disengaging restraints and inhibitions that are normally supported support ed by mindful mindfulness. ness. In sum summar mary y, it appe appears ars that emo emotion tional al eat eating ing fai fails ls to pro produce duce any rea reall or las lasting ting ben benefit efit to psy psychol chologiogicall an ca and d mo mood od st state ates, s, an and d tha thatt in incr crea ease sed d or re repea peated ted emoti em otiona onall eat eating ing ma may y act actual ually ly con contr tribu ibute te to moo mood d dysphoria.
7
8. Food craving as addiction
Food cra Food craving ving shares shares som somee fea feature turess with drug addiction, including similar neurotransmitter neurotransmitter substra substrates, tes, and terms such as ‘ chocoholic’ and ‘ sugar addiction’ are in common use, but there is no consensus that food cravings qualify as an addiction. The psychoactive substances in chocolate lack such propensities, and chocolate cravers classing themselves as ‘ chocoholics’ identify the orosensory properties as the ‘addictive’ factor (Hether(Hetherington ingt on and Mac Macdiar diarmid mid,, 1993 1993). ). Thi Thiss inte interpr rpretat etation ion is compatible with positive-incentive theories of acquired addicti add iction, on, in whic which h the cra craved ved sub substan stance ce is des desire ired d for its hedonic experience (Pinel, (Pinel, 1990), 1990), as opposed to abstinence models that involve a substance being physiologically ica lly req require uired d to give relief from an aver aversive sive state. Expecta Exp ectation tion models are bas based ed on lea learni rning ng abou aboutt the positive reinforcement of the desired substance, with cravingss triggered by stimuli such as the sight and smell craving of palatable foods which elicit desire for, and subsequent subsequent consumption of that food (Cornell (Cornell et al., 1989). 1989). The abstinence model may have some relevance to a tendency by those with low blood sugar to report irritability or poor mood while completing cognitive tasks, with wi th su such ch a mo mood od st stat atee st stim imula ulatin ting g co cons nsump umptio tion n of pal palat at-ablee hig abl high h car carbohy bohydra drate te foo food d tha thatt rel releas eases es endo endorphi rphins ns (Bent Benton, on, 2002 2002). ). Recog Recognizi nizing ng that thataa prop proporti ortion on of carb carbohyohydrate cravers report feeling ‘ alert ’ and ‘ calm’ after eating a snack, it is reasonable to concede that physiological benefits mediated by the opioid system have relevance to some individuals and in certain circumstances. 8.1. Ch 8.1. Choc ocol olat atee cr crav avin ing g an and d em emot otio iona nall ea eati ting ng as separate phenomen phenomena a
The literature tends to confound concepts of chocolate craving, carbohydrate craving, obesity, emotional eating and self-medication. We suggest it is useful to distinguish disting uish clearly between two separa separate te phenome phenomena: na: first, ‘chocolate craving’, an and d se seco cond nd,, car carboh bohydr ydrate ate craving in the context of ‘emotional eating’. The two phenomena can, however, co-exist in the same individual by virtue of the dual status of chocolate as being desired desire d specif specifically ically for its unique sensor sensory y experien experience ce and being more generally craved as a carbohydrate at times of emotional eating. There is some support for the hypothesis that each phenomenon is driven by different motivations, activating differ different ent neurotr neurotransmitt ansmitters, ers, and producin producing g differ different ent outcomes. In essence, chocolate craving is driven by a desire for hedonic reward, and dopamine (being responsible sib le for rew reward ard)) is the pre predom dominan inantt neur neurotra otransm nsmitte itter r
ARTICLE IN PRESS 8
G. Par Parker ker et al. / Jour Journal nal of Affec Affective tive Diso Disord rders ers xx (200 (2006) 6) xxx – xxx
release relea sed d af after ter ea eatin ting g cho choco colat latee fo forr the pu purp rpose ose of a pleasurable sensory experience. In contrast, emotional eating is characterized by carbohydrate craving and is motivated by desire for the comforting effect of opioids (endorphins) to alleviate dysphoria and other negative states. Suppor Sup portt for the dif differ ferenti ential al role of dopa dopamine mine and opioid opi oid sys system temss in choc chocolat olatee cra craving ving and emo emotion tional al eat eating ing is sug suggest gested ed in sev several eral stu studies dies.. Crav Craving ing for choc chocolat olatee has been shown to be more driven by a desire for the dopamin dop aminerg ergic ic rew reward ardss of con consum sumptio ption n rat rather her than to avoid negative consequences of abstinence (Stewart (Stewart et al., al ., 198 1984; 4; Wi Wise se,, 198 1988; 8; Coo Coope perr and Kir Kirkma kman, n, 199 1993; 3; Ko Koob ob and an d Le Mo Moal al,, 20 2001 01;; Pe Pelc lcha hat, t, 20 2002 02;; Pi Pine nel, l, 19 1990 90). ). Drewnowski Drewnow ski (1992) (1992) reporte reported d that inf infusi using ng the opio opioid id antagon ant agonist ist nalo naloxone xone red reduce uced d cal caloric oric inta intake ke in bing bingee eaters, as well as the taste preference for sweet high-fat foods such as biscuits and chocolate in both binge eaters and an d in co contr ntrols ols,, co comp mpar ared ed wit with h an op opioi ioid d ag agoni onist st or sa sali line ne infusion. The author suggested that endogenous opioid peptides may be involved in mediating taste responses and preferences, and especially for sweet, fatty foods. Benton Ben ton (20 (2002) 02) rep reporte orted d tha thatt poor mood stim stimulat ulates es eati eating ng of pal palatab atable le high car carbohy bohydra drate te food lead leading ing to end endorp orphin hin release, and Fullerton et al. (1985) demonstrated demonstrated that opioid antagonists can suppress stress-induced eating. This differentiating model is supported by other lines of evid evidenc ence. e. Weinga Weingarten rten and Els Elston ton (19 (1991) 91) found found a majority of chocolate cravers reported there was no nonchocolate substitute when they craved chocolate, but, when whe n cra cravin ving g car carbohy bohydra drate, te, then any swee sweet-f t-fat at food (including chocolate) was consumed. consumed. Schuman et al., 1987 found found som somee dis distinc tinction tion betw between een cho chocola colate te and sweet cravings, while Rozin while Rozin et al., 1991 concluded 1991 concluded that, although craving for chocolate is partly related to sweet craving, the major component is independent of sweet craving. Schum Schuman an et al al.. (1 (1987 987)) investigated ‘chocoholics’, former alcoholics (since they crave sweet food) and non non-cr -cravi aving ng con contr trols ols.. Whe When n sub subjec jects ts wer weree re re-grouped according to whether they ‘self-medicated’ or not, 32% of chocoholics were ‘self-medicators’ com pared to 13% of the controls and 23% of the former alcohol alc oholics ics.. Cho Chocola colate te was not grea greatly tly pre prefer ferred red ove over r other oth er swee sweets ts for sel self-m f-medic edicatio ation, n, even amo among ng choc chocooholics. hol ics. Sel Self-m f-medic edicatio ation n occu occurre rred d mos mostt com commonl monly y in response to depression, tension and irritability, less often in response to anxiety, and least often to anger. Chocolate produces a unique effect: when craving chocolate specifically, only chocolate will satisfy that craving. Yet, when experiencing an aversive mood state, any carbohydrate will suffice in an attempt to achieve relief, with chocolate no better than ice cream, cake or
potato chips. Anecdotal reports from ‘chocoholics’ who are also emotional emotional eat eaters ers indi indicate cate tha that, t, as cho chocoho cohollics, they crave chocolate only. However, when feeling ‘stressed ’, they will crave any carbohydrate, and chocolate is not preferred preferred over any other carbohydrate. Since choco ch ocola late te ha hass the hig highe hest st of he hedo donic nic ra rati tings ngs and produces rewarding and pleasurable feelings, it might seem reasonable to expect that these feelings would be an affective panacea. However, when experiencing an aversive mood state, the pleasurable feelings unique to the experience of eating chocolate are not what is required, and may be ineffective in ameliorating negative affective states. In such negative states, any carbohydrate may be ingested as a comfort food. 9. An evolutionary perspective
Addiction has been modeled as a cycle of progressively-increasing dysregulation of brain reward systems, resulting in compulsive use and loss of control over drug taking. taki ng. Acco Accordi rding ng to to Koob Koob an and d Le Mo Moal al (2 (200 001) 1) the normal ‘limitation of reward’ function fails to operate within wit hin the nor norma mall ho home meost ostat atic ic ra rang ngee an and d fo form rmss an allostatic state (i.e., a chronic deviation of the reward set point). An evolutionary model hypothesizes that food craving, overeating and addiction emerge from dysregulation of endogenous systems whereby higher than normal concentrations of certain foods provide excessive sti stimul mulatio ation n and over overrid ridee reg regulat ulatory ory fee feedbac dback k systems. Ther Th eree ar aree two key in innat natee fa facto ctors rs as asso soci ciat ated ed wit with h eating eati ng that cont contribu ribute te to this dysregula dysregulation tion.. Fir First, st, the natural environment in early human existence was unlikely lik ely to hav havee pr provi ovide ded d a re regul gular ar an and d re relia liabl blee fo food od suppl sup ply y. To ad adjus justt to var varia iable ble per period iodss of pl plent enty y and famine fam ine,, ada adaptiv ptivee adva advantag ntages es woul would d hav havee acc accrue rued d to those with flexibi flexibility lity in appetite and and satiety mechanism mechanisms, s, allowing consumption of greater amounts during times of plenty and building of temporary stores of body fat to carry through times of famine (Hammersley (Hammersley and Reid, 1997). 1997 ). Individuals lacking such flexibility would have been less likely to survive. Extreme expressions of ‘flexibility ’ would have been maladaptive. For example an ob obes esee ind indivi ividua duall wou would ld be at a di disa sadva dvant ntage age in hunting food and escaping from predators. In modern western west ernized ized soc society iety whe where re food is rea readily dily available available,, suc such h flexibi fle xibility lity is pote potentia ntially lly mal malada adaptiv ptive, e, lead leading ing ind individ ividuals uals to react to the stacked supermarket shelves as if it were a time of ple plenty nty and like likely ly to be fol followe lowed d by fam famine ine.. In In the absence of famine, shopping and food intake require some degree of cognitive restraint to avoid unwanted weight gain.
ARTICLE IN PRESS G. Parker et al. / Journal of Affective Disorders xx (2006) xxx – xxx
Second, hum Second, humans ans have inna innate te tas taste te pre prefer ferenc ences es for sweet, fat and salt. This ensured that foods ingested in their natural form would provide sufficient calories and other essential nutrients. Body mechanisms that evolved for survival in a natural environment, with appetite and satiety mechanism ‘calibrated’ to regulate ingestion of natural foods in most individuals, are unlikely to have adapted to accommodate the higher concentrations of caloriess in processed food. In developed countries, such calorie taste preferences and lack of adaptation operate in an environment with ready availability of processed foods that are cal calorie orie-su -super persatu saturat rated ed and cont contain ain high con con-centrations of sugar, starch, fat and salt. Processed food might also be deficient in other essential elements such as magnesium, thus failing to satisfy needs for these nutrient nutr ientss and cra craving ving will not be sat satisf isfied. ied. For suc such h crav cr aving ings, s, ch choco ocolat latee of offe fers rs a uni unique que com combin binat ation ion of sweetn swe etness ess,, cre creamy amy tex textur ture, e, cha charac racter terist istic ic tas taste te and tantalizing tantali zing aroma — the perfect orosensory experience to seduce the palate. Endoge End ogenou nouss mec mechan hanism ismss tha thatt hav havee evo evolve lved d to respond to natural substances, whether external (food) or inte interna rnall (ne (neurot urotran ransmi smitter tters), s), can be com comprom promise ised d when wh en th thee bod body y is su subje bject cted ed to su subs bstan tances ces wh which ich ar aree chem ch emica ically lly si simi milar lar bu butt in mo more re con conce centr ntrat ated ed fo form rm.. Ju Just st as the natural function of dopamine and opioid systems can be compromised by exogenous opiates and other illicit drugs, dru gs, the nat natura urall fun functi ction on of app appeti etite te and sat satiet iety y mechani mec hanisms sms can be chal challeng lenged ed by calo calorierie-supe supersa rsatur turated ated processed food. Rather than assume a state of addiction or presence of psychoactive chemicals, a simpler answer may be that the behaviour is either a compromise or a dysregu dys regulati lation on of natu natural ral endo endogeno genous us mec mechani hanisms sms.. In essence ess ence,, cer certain tain neur neurotra otransm nsmitte itterr sys systems tems are act activat ivated ed in dysfunc dys function tional al way wayss by inap inappro propria priate te ana analogu logues es or by concent con centrat rations ions of the fac factor torss (in (includ cluding ing foo foods) ds) whic which h exceed exc eed the evo evoluti lutionar onary y capa capacity city and over overrid ridee reg regulat ulatory ory feedback mechanisms. Such foods do not have to be intrinsically intrins ically addictive through any ‘ psychoactive’ properties, but simply have the capacity to activate neurotransmitters in a dysfunctional way. High concentrations of sugar and fat, by virtue of their ability to release dopamine and opioids, have the potential to compromise such endogenous mechanisms. Rather than view them (and chocolate specifically) as ‘substances of abuse’, it migh mi ghtt be mo more re app appro ropr priat iatee to co cons nside iderr th them em as ab abus user erss of endogenous mechanisms. 10. Conclu Conclusion sion
For most people chocolate invokes anticipatory and consummatory consumm atory pleasure, and is therefore an indulgen indulgence. ce.
9
When tak When taken en in re resp spons onsee to a dys dyspho phoric ric state state as an ‘ emoti emotional onal eating’ stra strate tegy gy it ma may y pr prov ovid idee so some me transient ‘comforting’ role role bu butt it is mo morre li like kely ly to prolong rather than abort the dysphoric mood. It is not, as some would claim, an antidepressant.
Acknowledgments
This review was supported by an NHMRC Program Grant Gra nt (222 (222708) 708) and an Inf Infras rastruc tructur turee Gra Grant nt fro from m the NSW Department of Health. References Benton, D., 2002. Carbohydrate ingestion, blood glucose and mood. Neuroscience and Biobehavioral Reviews 26, 293–308. Benton, D., Donohoe, R.T., 1999. The effects of nutrients on mood. Public Health Nutrition 2, 403–409. Blass, E.M., 1986. Functional interaction between positive effect of sweet and the negative effect of pain and distress. Appetite 7, 243. Blass, Bla ss, E.M E.M., ., Hof Hoffme fmeyer yer,, L.B L.B., ., 199 1991. 1. Suc Sucros rosee as an ana analge lgesic sic for newborn infants. Pediatrics 87, 215–218. British Nutrition Foundation, 1998. BNF Conference report: stimulating thoughts: caffeine and food. BNF Nutrition Bulletin 23, 226–233. Bruinsma, K., Taren, D.L., 1999. Chocolate: food or drug? Journal of American Dietary Association 99, 1249–1257. Cabellero, B., Finer, N., Wurtman, R.J., 1988. Plasma amino acids and insuli ins ulin n lev levels els in obe obesit sity: y: res respon ponse se to car carboh bohydr ydrate ate int intake ake and tryptophan supplements. Metabolism 37, 672–676. Christensen, L., 1993. Effects of eating behavior on mood: a review of the lit literatu erature. re. Inte Internat rnationa ionall Jour Journal nal of Eati Eating ng Dis Disorde orders rs 14, 171–183. Christense Chris tensen, n, L., Burrow Burrows, s, R., 1990. Diet Dietary ary treatm treatment ent of depres depression. sion. Behavior Therapist 21, 183–193. Christensen, L., Kreitsch, K., White, B., Stagner, B., 1985. Impact of dietary die tary chan change ge on emot emotiona ionall dis distres tress. s. Jou Journal rnal of Abn Abnormal ormal Psychology 94, 565–579. Christensen, L., Kreitsch, K., White, B., 1989. Development, crossvalidation, valid ation, and asses assessment sment of the reliability reliability of the Christensen Christensen Dietar Die tary y Dis Distre tress ss Inv Invent entory ory.. Can Canadi adian an Jou Journa rnall of Beh Behavi aviour oural al Science 21, 1–15. Cohen, I.T., Sherwin, B.B., Fleming, A.S., 1987. Food cravings, mood and the menstrual cycle. Hormones and Behavior 21, 457–470. Cooper, S.J., Higgs, S., 1994. Neuropharmacology of appetite and taste preferences. In: Legg, C.K., Booth, D.A. (Eds.), Appetite, Neural and Behavioral Bases. Oxford University Press, Inc., New York, pp. 212–242. Cooper, S.J., Kirkman, T.C., 1993. Opioid mechanisms in the control of food consumption and taste preferences. In: Herz, A. (Ed.), Handbook of Experimental Pharmacology. Springer-Verlag, Berlin, pp. 321–342. Cornell, C., Rodin, J., Weingarten, H., 1989. Stimulus-induced eating when satiated. Physiology & Behavior 45, 695–704. di Tomaso, E., Beltramo, M., Piomelli, D., 1996. Brain cannabinoids in chocolate. Nature 382, 677–678. Drewno Dre wnowsk wski, i, A., 199 1992. 2. Foo Food d pre prefer ferenc ences es and the opi opioid oid pep peptid tidee system. Trends in Food Science & Technology 3, 97–99.
ARTICLE IN PRESS 10
G. Par Parker ker et al. / Jour Journal nal of Affec Affective tive Diso Disord rders ers xx (200 (2006) 6) xxx – xxx
Drewnowski, A., Greenwood, M.R.C., 1983. Cream and sugar: human preferences for high-fat foods. Physiology & Behavior 30, 629–633. Drewnowski, A., Kurth, C., Ho, J., Saari, J., 1992. Food preferences in human obesity: carbohydrates versus fats. Appetite 18, 207–221. Fullerton, D.T., Getto, C.J., Swift, W.J., Carlson, I.H., 1985. Sugar, opioids, opioi ds, and binge eating. Brain Research Bulletin 14, 673–680. Ganley, R.M., 1989. Emotion and eating in obesity: a review of the literature. International Journal of Eating Disorders 8, 343–361. Gulati Gul ati,, K., Ray, Ray, A.,Sharma A.,Sharman, n, K.K K.K., ., 199 1991. 1. Rol Rolee of diurnal diurnal var variat iation ion and receptor recept or speci specificit ficity y in the opioid opioidergic ergic regulation regulation of food intake in free-fe fre e-fed d and foo food-d d-depr eprive ived d rat rats. s. Phy Physio siolog logy y & Beh Behavi avior or 49, 1065–1071. Hammersley, R., Reid, M., 1997. Are simple carbohydrates physiologically addictive. Addiction Research 5, 145–160. Heatherton, T.F., Baumeister, R.F., 1991. Binge eating as an escape from self-awareness. Psychological Bulletin 110, 86–108. Heatherton, T.F., Herman, C.P., Polivy, J., 1992. Effects of distress on eating: the importance of ego-involvement. Journal of Personality and Social Psychology 62, 801–803. Hetherington, M.M., Macdiarmid, J.I., 1993. “Chocolate addiction”: a prel prelimi iminar nary y stu study dy of its des descri cripti ption on and its rel relati ations onship hip to problem eating. Appetite 2, 233–246. Hill, A.J., Weaver, C.F., Blundell, J.E., 1991. Food craving, dietary restraint and mood. Appetite 17, 187–197. Hudson, Huds on, A., Williams, Williams, S.G. S.G.,, 1981. Eating behavi behavior, or, emotions and overweight. Psychological Reports 48, 669–670. Hurst, W.J., Martin, R.A., Zoumas, B.L., Tarka, S.M., 1982. Biogenic amines in chocolate — a review. Nutrition Reports International 26, 1081–1086. Johnson, C., Larson, R., 1982. Bulimia: an analysis of moods and behavior. Psychosomatic Medicine 44, 341–351. Koob, G.F., G.F., Le Moal, M., 2001. Drug addict addiction, ion, dysregulation dysregulation of reward, and allostasis. Neuropharmacology 24, 97–129. Kreitsch, K., Christensen, L., White, B., 1988. Prevalence, presenting symptoms, sympto ms, and psycho psychologica logicall charact characteristi eristics cs of indivi individuals duals experiencin experi encing g a diet-re diet-related lated mood-d mood-disturb isturbance. ance. Behavi Behavior or Therap Therapy y 19, 593–604. Lappalainen, R., Sjoden, P., Hursti, T., Vesa, V., V., 1990. Hunger/craving responses and reactivity to food stimuli during fasting and dieting. International Journal of Obesity and Related Metabolic Disorders 14, 679–688. Lieberman, H.R., Wurtman, J.J., Chew, B., 1986. Changes in mood after carbohydrate consumption among obese individuals. American Journal of Clinical Nutrition 44, 772–778. Lingford-Hughes, A., Nutt, D., 2003. Neurobiology of addiction and implica imp licatio tions ns for trea treatmen tment. t. Bri Britis tish h Jou Journalof rnalof Psy Psychi chiatr atry y 182 182,, 97–100. Lyman, B., 1982. The nutritional values and food group characteristics of foo foods ds pre preferr ferred ed dur during ing var variou iouss emo emotio tions. ns. The Jou Journa rnall of Psychology 112, 121–127. Macdiarmid Macdi armid,, J.I. J.I.,, Hethe Hetheringto rington, n, M.M. M.M.,, 1995. Mood modula modulation tion by food: an explanation of affect and cravings in ‘chocolate addicts’. British Journal of Clinical Psychology 34, 129–138. Meisel, Meis el, R.L., Hays, T.C., T.C., Del Paine Paine,, S.N. S.N.,, Luttr Luttrel, el, Y.R., Y.R., 1990. Induction of obesity by group housing in female Syrian hamsters. Physiology & Behavior 47, 815–817. Mercer, M.E., Holder, M.D., 1997. Food cravings, endogenous opioid peptides, and food intake: a review. Appetite 29, 325–352. Michener, W., W., Rozin, P., 1994. Pharmacological versus sensory factors in the satiation of chocolate craving. Physiology & Behavior 56, 419–422. Moller, S.E., 1992. Serotonin, carbohydrates and atypical depression. Pharmacology and Toxicology 71 (suppl 1), 61–71.
Morley, J.E. Morley, J.E.,, Levine Levine,, A.S. A.S.,, Rowla Rowland, nd, N.E., 1983. Stress induced eating. Life Sciences 32, 2169–2182. Ottley, C., 2000. Food and mood. Nursing Standard 15, 46–52. Parker, G., Brown, L.B., 1982. Coping behaviors that mediate between life events and depres depression sion.. Archiv Archives es of Genera Generall Psych Psychiatry iatry 39, 1386–1391. Parker, G., Hadzi-Pavlovic, D., Wilhelm, K., Hickie, I., Brodaty, H., Boyce, P., Mitchell, P., Eyers, K., 1994. Defining melancholia: properties of a refined sign-based measure. British Journal of Psychiatry 164, 316–326. Parker, Parke r, G., Roy Roy,, K., Mitchell, P., P., Wil Wilhelm, helm, K., Mahli Mahli,, G., HadziPavlovic, D., 2002. Atypical depression: a reappraisal. American Journal of Psychiatry 159, 1470–1479. Pelchat, Pelch at, M.L., 2002. Of human bondage: food cravin craving, g, obses obsession, sion, compulsion, and addiction. Physiology & Behavior 76, 347–352. Pinel, J.P.J., 1990. Biopsychology, 3rd edn. Allyn and Bacon, London, pp. 240–242. Ramarao, P., Bhargava, H.N., 1989. Effects of kappa-opioid receptor agonists agoni sts and morphi morphine ne on food intake and urinary output in fooddeprived and non-deprived rats. Pharmacology, Biochemistry & Behavior Behav ior 33, 375–380. Reid, M., Hammersley, Hammersley, R., 1999. The effec effects ts of carboh carbohydrate ydratess on arousal. Nutrition Research Reviews 12, 3–23. Rodin, J., Mancuso, J., Granger, J., Nelbach, E., 1991. Food cravings in relation to body mass index, restraint and estradiol levels: a repeat rep eated ed mea measur sures es st study udy in hea healt lthy hy wom women en.. App Appeti etite te 17, 177–185. Rogers, P.J., 1990. Why a palatability construct is needed. Appetite 14, 167–170. Rogers,, P.J., 1995. Food, mood and appeti Rogers appetite: te: a psycho psychobiolo biological gical perspective. Nutrition Research Review 8, 243–269. Rogers, P.J., Smit, H.J., 2000. Food craving and food ‘addiction’: a critical review of the evidence from a biopsychosocial perspective. Pharmacology, Biochemistry & Behavior 66, 3–14. Rosenth Ros enthal, al, N.E N.E., ., Genh Genhart, art, M.J M.J., ., Caba Caballer llero, o, B., Jac Jacobse obsen, n, F.M. .M.,, Skwerer, Skwe rer, R.G., Course Coursey y, R.D., Rogers, S., Sprin Spring, g, B.J., 1989. Psychobiological effects of carbohydrate and protein-rich meals in patients with seasonal affective disorder and normal controls. Biological Psychiatry 25, 1029–1040. Rozin, P., Levine, E., Stoess, C., 1991. Chocolate craving and liking. Appetite 17, 199–212. Ruderm Rud erman, an, A.J A.J., ., 198 1983. 3. Obe Obesit sity y, anx anxiet iety y, and foo food d con consum sumpti ption. on. Addictive Behaviors 8, 235–242. Schachter, S., Goldman, R., Gordon, A., 1968. Effects of fear, food depriv dep rivati ation on and obe obesit sity y on eat eating ing.. Jou Journa rnall of Per Person sonali ality ty and Social Psychology 10, 91–97. Schuman, M., Gitlin, M.J., M.J., Fairbanks, L., 1987. Sweets, chocolate, and atypical depressive traits. Journal of Nervous and Mental Disorder 175, 491–495. Si, E.C.C., Bryant, H.U., Yim, G.K.W., 1986. Opioid and non-opioid components of insulin-induced feeding. Pharmacology, Biochemistry & Behavior 24, 899–903. Small, D.M., Zatorre, R.J., Dagher, A., Evans, A.C., Jones-Gotman, M., 2001. Changes in brain activity related to eating chocolate. From pleasure to aversion. Brain 124, 1720–1733. Spring,, B., Maller, O., Wurt Spring Wurtman, man, J., Digma Digman, n, L., Cozolino L., 198283. Effects Effects of prot protein ein and car carboh bohydr ydrate ate meals on moo mood d and performance: interactions with sex and age. J. Psychiatric Res. 17, 155–167 Stewar Ste wart, t, J., de Wi Wit, t, H., Eik Eikelb elboom oom,, R., 198 1984. 4. The rol rolee of unc uncond ondit ition ioned ed and conditioned drug effects in the self-administration of opiates and stimulants. Psychological Review 91, 251–268.
ARTICLE IN PRESS G. Parker et al. / Journal of Affective Disorders xx (2006) xxx – xxx
Stone, A.A. Stone, A.A.,, Browne Brownell, ll, K.D. K.D.,, 1994. The stres stress-eati s-eating ng parado paradox: x: multi multiple ple daily measurements in adult males and females. Psychology and Health 13, 211–219. Tayl aylor or,, G.J G.J., ., Par Parker ker,, J.D J.D.A. .A.,, Bab Babey ey,, R.M R.M., ., Bou Bourke rke,, M.P., 1996 1996.. Relationships between alexithymia and psychological characteristics associated with eating disorders. Journal of Psychosomatic Research Resea rch 41, 561–568. Thayer, R.E., 1987. Energy, tiredness, and tension effects of a sugar snack versus moderate exercise. Journal of Personality and Social Psychology 52, 119–125. Timmerman, G.M., Acton, G.J., 2001. The relationship between basic need satisfaction and emotional eating. Issues in Mental Health Nursing 22, 691–701. Triscari, Tris cari, J., Nelso Nelson, n, D., Vincent, Vincent, G.P G.P., ., Li, C.H. C.H.,, 1989. Effects of centrally and peripherally administered beta endorphin on food intake in rats. International Journal of Peptide and Protein Research 34, 358–362. Van Str Strien, ien, T., 2000 2000.. IceIce-cre cream am cons consumpt umption, ion, ten tendenc dency y tow toward ard overeatin overe ating, g, and pers personal onality ity.. Inte Internati rnational onal Jour Journal nal of Eatin Eating g Disorders 28, 460–464. Weinga eingarten, rten, H.P., H.P., Elsto Elston, n, D., 1990. The phenom phenomonolo onology gy of food cravings. Appetite 15, 231–246. Weingarten, H.P., Elston, D., 1991. Food cravings in a college population. Appetite 17, 167–175.
11
Weintr eintraub, aub, W., Arons Aronson, on, H., 1969. Appli Applicatio cation n of verbal behavior analys ana lysis is to the stu study dy of psy psycho cholog logica icall def defens ensee mec mechan hanism isms. s. Archives of General Psychiatry 21, 739–744. Weiss, G., 1982. Food fantasies of incarcerated drug users. International Journal of Addictions 17, 905–912. Willner, P., Benton, D., Brown, E., Cheeta, S., Davies, G., Morgan, J., Morgan, M., 1998. “Depression” increases “craving” for sweet rewards in animal and human models of depression and craving. Psychopharmacology 136, 272–283. Wise, R.A., 1988. The neurobiology of craving: implications for the understanding and treatment of addiction. Journal of Abnormal Psychology 97, 118–132. Wurtman, R.J., Wurtman, J.J., 1989. Carbohydrates and depression. Scientific American 260, 68–75. Yeomans, M.R., 1996. Palatability and the microstructure of feeding in humans: the appetizer effect. Appetite 27, 119–133. Young, S.N., 1991. Some effects of dietary components (amino acids, carbohydrate, folic acid) on brain serotonin synthesis, mood and behavior. Canadian Journal of Physiology and Pharmacology 69, 893–903. Young, S.N., Smith, S.E., Pihl, R.O., Ervin, F.R., 1985. Tryptophan deplet dep letion ion causes causes a rap rapid id low loweri ering ng of moo mood d in norm normal al mal males. es. Psychopharmacology 87, 173–177.