Expectations and associations that heal: immunomodulatory placebo effects and its neurobiology

Brain, Behavior, and Immunity 20 (2006) 430–446

www.elsevier.com/locate/ybrbi

Named Series: Brain Mechanisms of Placebo

Expectations and associations that heal: Immunomodulatory placebo eVects and its neurobiology

Gustavo Pacheco-López, Harald Engler, Maj-Britt Niemi, Manfred Schedlowski ¤

Chair of Psychology and Behavioral Immunobiology, Institute for Behavioral Sciences, ETH Zurich, 8092 Zurich, Switzerland

Received 28 April 2006; received in revised form 29 May 2006; accepted 29 May 2006

Abstract

The use of placebo may have accompanied healing and medical practices since their origins (Plato; Charmides, 155–156). Recentexperimental data indicate that we would be well advised to further consider placebo eVects in future therapeutic strategies, with a betterknowledge of their potency, psychological basis and underlying neurobiological mechanisms. Current research in the areas of pain,depression and Parkinson’s disease has uncovered some of the potential neurobiological mechanisms of placebo eVects. These data indi-cate that conscious expectation and unconscious behavioral conditioning processes appear to be the major neurobiological mechanismscapable of releasing endogenous neurotransmitters and/or neurohormones that mimic the expected or conditioned pharmacologicaleVects. To date, research on placebo responses aVecting immune-related diseases is scarce, but there are consistent indications that skinand mucosal inXammatory diseases, in particular, are strongly modulated by placebo treatments. However, the brain’s capability to mod-ulate peripheral immune reactivity has been impressively demonstrated by paradigms of behavioral conditioning in animal experimentsand human studies. Thus, placebo eVects can beneWt end organ functioning and the overall health of the individual through positiveexpectations and behavioral conditioning processes.© 2006 Elsevier Inc. All rights reserved.

Keywords: Placebo; Nocebo; Immune; Conditioning; Expectation

1. Introduction

The right knee of a 76-year-old man had been givingtrouble for Wve years; he was diagnosed as having osteoar-thritis and later submitted to surgery. Two years after theoperation, his knee no longer bothered him (Talbot, 2000).Normally in such a case routine arthroscopic lavage/debridement surgery would have been performed. How-ever, in this clinical trial, the man was submitted to sham orplacebo surgery (Moseley et al., 1996). Similarly strong pla-cebo eVects on osteoarthritis after placebo surgery haverecently been replicated (Moseley et al., 2002). Are placeboeVects a myth or a panacea?

The placebo eVect is well known in biomedical researchand practical medicine, but the understanding of its neuro-

* Corresponding author. Fax: +41 44 6321355.E-mail address: [email protected] (M. Schedlowski).

0889-1591/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.bbi.2006.05.003

biology and neuropsychology is still limited. The lack of asolid theoretical framework has somewhat limited a sys-tematic approach and attempts to delineate its boundaries.However, recent research in the areas of pain, depressionand Parkinson’s disease has uncovered some of the neuro-biological mechanisms of placebo eVects. These data indi-cate that conscious expectation and unconsciousbehavioral conditioning appear to be the major neurobio-logical mechanisms. These processes are capable of releas-ing endogenous neurotransmitters that mimic the expectedor conditioned pharmacological eVects, i.e., the placeboeVect. Experimental evidence for the placebo eVects inimmune-related diseases is scant and ambiguous. However,research on the neurobiology and neuropsychology of pla-cebo eVects is becoming an active and productive area ofscience with the Wnal aim of understanding their healingpotential, as well as their limitations, and of delineatingcorrect and ethical use. In the near future it can be expected

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 431

that placebo eVects will contribute to pharmacological ther-apies in an eVort to create the optimum healing situationfor each individual. As a trend pioneer, Brain Behavior andImmunity launched its Wrst Named Series entitled “BrainMechanisms of Placebo” (Wager and Nitschke, 2005) involume 19. This contribution closes this Wrst series of arti-cles, in which a number of important mechanisms of pla-cebo eVects have been revealed: namely, (1) expectancyaVects pain experience, not simply pain reporting (Lorenzet al., 2005); (2) expectation (open vs. hidden stimulations)enhances autonomic responses to stimulate subthalamiclimbic regions in Parkinson patients (Lanotte et al., 2005);(3) individual variations in the neurochemistry of the anal-gesic placebo eVect are related to the individual experienceof pain itself (“need”) and the internal aVective state of theindividual (“belief”) (Zubieta et al., 2006); (4) alteringexpectancies of upcoming aversive events depends onrobust functional associations within the brain regionsimplicated in prior work on placebo eVects (Sarinopouloset al., 2006); and (5) placebo treatment aVects early nocicep-tive processing, however, another component of placeboeVects in reported pain occurs later, either in evaluation ofpain or cognitive judgments of pain reports (Wager et al.,2006).

1.1. Basic concepts of placebo

Placebo conceptualization, placebo employment andethical considerations in placebo research have dramati-cally changed during history, especially in the 20th century(Papakostas and Daras, 2001). There are at least two dis-tinct uses for placebos: in research as tools for reducingbias in clinical trials, and in clinical practice for therapeuticpurposes. In both cases, major ethical issues may arisewhenever two key considerations for the patient, therapeu-tic beneWcence and autonomy, are violated (Simon, 2000;Miller et al., 2005).

In order to avoid misunderstandings, it is necessary toclearly deWne concepts and nomenclature of the placebophenomenon. In a group of untreated patients with a givendisease, it is expected that a certain percentage of patientswill improve after a few weeks. With the administration ofan inactive pill, the percentage may be higher, and evenhigher if active pharmacological treatment is employed.One possibility is that an entire spectrum of “non-speciWc”factors, such as the natural course of the disease, the regres-sion towards the mean, and other overt or covert inXuencesmay contribute to this healing process over the time courseof an untreated illness (Ernst and Resch, 1995). Second, inthe case of treatment with an active drug, the patientsimproved because of the “speciWc” eVects of the treatmentthat can be scientiWcally studied, isolated, and predictedthrough speciWcally designed studies: the randomized con-trolled trials (Kaptchuk, 1998). Finally, in the case of“inert” therapy, it is obvious that the inactive pill, becauseit lacked a speciWc pharmacologic eVect, acted therapeuti-cally through its “symbolic power” and its impact on the

patient’s imagination, beliefs, expectations, emotions, andprevious experiences (Brody, 1985; Kaptchuk, 1998).Patients were under the impression that they were taking anactive or speciWc drug with known eYcacy for their condi-tion. Treatment with an inert sugar pill lacking speciWcactivity represents a typical example of what is called pla-cebo therapy, and the “dummy” pill is called a placebo(from the Latin: I shall please) (Shapiro, 1960). Besides theadministration of a drug, this placebo therapy can take theform of physical (i.e., manipulation, context), psychological(i.e., conversation, psychotherapy), or any other kind oftreatment. The elicited and measurable response after theadministration of a placebo is called a placebo response orplacebo eVect. The “true” placebo eVect may be estimatedas the diVerence between the magnitude of the placeboresponse minus the magnitude of the “time eVect” onuntreated patients (Ernst and Resch, 1995), or by compar-ing the eVects of “open” (i.e., aware) vs. “hidden” (i.e.,unaware) treatments (Colloca et al., 2004; Colloca andBenedetti, 2005). Nocebo (from the Latin: I shall harm) andnocebo eVects are terms used in cases where patients attachnegative meanings and emotions to treatment or have ver-bally induced expectations that it is harmful (Benson, 1997;Hahn, 1997; Spiegel, 1997).

Thus, most, if not all, medical procedures are associatedwith a complex psychosocial context that can inXuencethe therapeutic outcome. Placebo research is essentially thestudy of the psychosocial context that surrounds thepatient and contributes to healing (Benedetti et al., 2004).

1.2. Features of placebo eVects

Placebo eVects can be very speciWc (Kirsch, 1997), andthis speciWcity depends on the information available to therecipient (Flaten et al., 1999; Flaten et al., 2004). For exam-ple, placebos can have opposite eVects on heart rate orblood pressure, depending on whether they are given astranquilizers or stimulants (Kirsch, 1997). Opposite verbalsuggestions have been found to inXuence the placeboresponse in the desired direction regardless of the eVect ofthe active agent (Luparello et al., 1968; Kirsch, 1998; Flatenand Blumenthal, 1999; McMillan, 1999), suggesting thateven side eVects may be triggered by opposite verbal sug-gestions (Kaptchuk et al., 2006).

One would probably not expect placebo eVects to exceedthe homeostatic limits of the body. However, this does notmean that they are not clinically relevant. In addition,fascinating observations have been made showing that pla-cebo eVects can persist over 8 weeks (Coryell and Noyes,1988), 6 months (Boissel et al., 1986), or even over 30months (Traut and Passarelli, 1957). Moreover, a placeboresponse has been reported to resist 10 extinction trials(Montgomery and Kirsch, 1997). Thus, comparing thepharmacological eVect with the placebo eVect might revealdiVerent outcomes for acute and chronic pathologies. Fur-thermore, there is evidence to suggest that placebos can beaddictive, causing withdrawal symptoms when treatment is

432 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

discontinued (Wolf, 1959; Brody, 1980). This is in line withcurrent knowledge which claims that some placebotherapies are able to induce the release of neurotransmit-ters, similar to that induced pharmacologically (see below).

Researchers have been attempting to identify theplacebo-prone personality. Early work with demographicand personality variables has produced controversialresults (Lasagna et al., 1954; Moertel et al., 1976; McNairet al., 1979). Of course, conclusions drawn from such datacan only be applied strictly to the speciWc patient popula-tion. However, so far a personality proWle that representsthe placebo-responder has not yet been reliably identiWed.A recent study blamed this lack on methodological issuesand conducted an experiment revealing pessimists to bemore likely than optimists to follow negative verbal sugges-tions – “you will ingest a pill, which will make you feelunpleasant symptoms” — and to develop strong noceboeVects. However, no diVerences were found among person-alities when uncertainty about the treatment assignment(placebo or active drug) or neutral verbal suggestion pre-ceded the ingestion of an inactive pill (Geers et al., 2005a).The authors concluded that personality and situationalvariables seem to interact to determine placebo responding.In addition, it has been suggested that motivational pro-cesses aVect the placebo response (Geers et al., 2005b). Inthis regard, a number of psychological variables seem toaVect the placebo response: concern and an uncomfortablefeeling (Todd, 1987), anxiety (Melzack, 1988) and the char-acter of pain; experimental pain in healthy adults is theleast sensitive to placebo eVects, while pain related to anxi-ety, as in heart disease, is the most sensitive at up to 90%(Haour, 2005). In addition, the patient’s attitude toward thetreatment seems to be crucial, in particular, patient suggest-ibility, the desire to be “a good patient” (Bailar, 2001),adherence in cancer (Pizzo et al., 1983) and in coronaryheart disease patients (Gallagher et al., 1993; Horwitz et al.,1990), awareness and expected outcome of the treatment,and compliance (de Craen et al., 1999a) have been demon-strated to have a positive impact on placebo response. Theperceived group assignment (placebo or active drug) hasbeen shown to play an important role (Benedetti, 2005).For instance, in human fetal mesencephalic transplantationfor patients with Parkinson’s disease, the perceived assign-ment had a more powerful impact on both quality of lifeand motor function than did the actual treatment (McRaeet al., 2004). Also, when placebo responders were told thatthe treatment allocated to them had been a placebo, mostof them relapsed and had to be prescribed the “real” medi-cation (Leuchter et al., 2002); patients who learn that theyhave been deceived with a placebo can worsen (Wrolstad,2002). An important Wnding in this context is the fact thatmany participants (31%) actually changed their belief as totheir group allocation during the trial (Rees et al., 2005).Controversial Wndings have been reported concerning thedesire for relief (Price et al., 1999). The desire manipulationof one study relied on participants’ intrinsic desire for painrelief, whereas that of another study was related to being

told that the response to the drug meant that the partici-pants had more favorable personality characteristics(Jensen and Karoly, 1991). Apparently, some forms ofdesire/motivation are more eVective than others in promot-ing placebo responses, and behavioral conditioningstrengthens these eVects (Haour, 2005).

Several authors have stressed the importance of thephysician–patient relationship (Papakostas and Daras,2001; Colloca et al., 2004). In particular, interpersonal skills(Oh, 1991), persuasive inXuence (Shapiro, 1971), and thetime spent with the patient (Solomon, 2002; Kaptchuket al., 2006) have been identiWed as the variables that mod-ify placebo eVects. Moreover, the physician’s expectationsand attitudes (SandroV, 1980), his or her enthusiasm (Grylland Katahn, 1978), and optimistic expectation (Uhlenhuthet al., 1966) were positively correlated with the patient’splacebo response (Graz et al., 2005). However, the decep-tion that is inherent in the use of placebos may damage thedoctor–patient relationship in subtle ways. Thus, there isreason for caution when dealing with the notion that place-bos cannot hurt (Bailar, 2001).

1.3. Placebo features aVecting placebo eVects

Regarding the nature of the placebo itself, it is well docu-mented that certain features support its eVectiveness,among them are: the number (de Craen et al., 1999b;Moerman, 2000) and the color of the pills (Schapira et al.,1970), the frequency (Braverman, 1986; Ilnyckyj et al.,1997), the route of administration and formulation (Haour,2005), and the mode of administration (Levine and Gor-don, 1984). For instance, surgery is said to have the mostpotent placebo eVect (Finneson, 1969; Moseley et al., 2002),and injections are more eVective than tablets. Also thepotency of the corresponding active agent (Evans, 1974), awell-known brand name (Branthwaite and Cooper, 1981),new vs. old treatment, novelty (Shapiro, 1971), and phar-macological treatment before placebo stimulus all triggerstronger eVects (Wager and Nitschke, 2005). In addition,the setting in which the placebo is given (e.g., white coat,Guess et al., 2002; hospital, Voudouris et al., 1990) makes adiVerence.

1.4. Limitation of placebo eVects

Hróbjartsson and Gøtzsche analyzed more than a hun-dred studies covering a wide range of medical problems. Intotal, forty clinical conditions were examined, from asthmaand smoking to menopause, marital discord and schizo-phrenia (Hróbjartsson and Gøtzsche, 2001). These authorsaveraged all these studies and, because there were relativelyfew studies in this sample that provided evidence in favor ofplacebo eVects, the negative view prevailed. But if oneattempted the same thing for virtually any powerful drug,the result would be the same. However, a recent re-analysisof those studies shows that when disorders are amenable toplacebos and the design is adequate to detect the eVects,

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 433

placebo eVects are robust and even come close to the treat-ment eVect (Wampold et al., 2005).

It should be mentioned that any kind of therapy thatworks — be it a drug, a surgical intervention, or behavioraltherapy — will help patients with some conditions and notothers. There is no such thing as a universal remedy, a real-life cure-all, a panacea. Certainly, some people have claimedthat placebos are just this. Beecher (1955) was largelyresponsible for launching the idea that placebos can aVectvirtually every medical condition, which may be one reasonwhy placebo eVects have so often been dubbed, unhelpfully,as “non-speciWc”. If Hróbjartsson and Gøtzsche had con-tented themselves with exposing the myth about placeboeVects being a panacea, then the path would have been openfor a more realistic assessment of placebo eVects, distin-guishing between those conditions that are placebo-respon-sive and those that are not.

Importantly, it has been documented that prefrontalcortex degeneration such as occurs in Alzheimer disease,makes analgesic therapies less eVective (Benedetti et al.,2006). Such an eVect has been attributed to the loss of anexpectation-related mechanism; i.e., the placebo analgesiceVect. In this regard, it will be necessary to determine whichneuro- and psychopathological conditions potentiate oreliminate placebo eVects and to reconsider posology forpatients with such conditions.

2. A theoretical framework for placebo eVects: expectation and conditioning

The psychobiological phenomena ascribed to placeboeVects seem to be due to diVerent mechanisms, includingconscious cognitive factors, such as the patient’s expecta-tion of clinical beneWt, and Pavlovian conditionedresponses (Benedetti et al., 2005). In some instances thesetwo neurobiological processes might contribute together tothe Wnal therapeutic outcome, and in others they mightcontribute diVerentially to a given placebo eVect. Althoughfor methodological and mechanistic reasons it will be essen-tial to precisely analyze these neuropsychological processesof the placebo response, the clinical perspective has beenfocused on the symptom side.

Two alternative theoretical frameworks have tradition-ally been proposed to explain placebo eVects (Byerly, 1976):(1) the mentalistic theory, according to which the patient’sexpectation is the primary basis for the placebo eVect, and(2) the conditioning theory, which states that the placeboeVect is essentially the consequence of previous experiencesand results from associative learning processes (condi-tioned response). Models have been developed to supportone or the other of these theories (Wolf, 1959; Evans, 1985;Wickramasekera, 1985; Ader, 1997; Kirsch, 1997). How-ever, these two mechanisms do not have to be mutuallyexclusive, because both can play a relevant role underdiVerent circumstances, having additive properties or evena synergistic eVect in certain cases. For example, it has beendocumented that expectation and conditioning are the

neurobiological mechanisms in placebo analgesia(Amanzio and Benedetti, 1999). Here, the magnitude of theplacebo analgesia (tolerance to experimentally inducedischemic arm pain) was enhanced when expectation andconditioning were combined. Additionally, Benedetti andcolleagues were able to demonstrate a clear neuropharma-cological dissection of placebo analgesia: expectationactivated the opioid system and conditioning activatedother speciWc subsystems. The placebo analgesia induced byexpectation (positive verbal suggestion: “you will receive apowerful painkiller”) was completely opioid-dependent,thus blocked by naloxone treatment, whereas conditioningprocedures (conditioned stimulus: injection procedure —unconditioned stimulus: ketorolac, a non-steroidanti-inXammatory drug) were naloxone-insensitive. Fur-thermore, if learning processes were not involved in placeboeVects, one would expect that one placebo would be aseVective as another and that a placebo would be equallyeVective no matter when it was administered (Ader, 1997).However, experimental and clinical data clearly indicatethat there is a greater placebo eVect if placebo medication isgiven after a period of eVective drug treatment, when asso-ciative learning might have taken place, than when it isgiven as the “Wrst” medication (Wager and Nitschke, 2005).

Consciousness seems to be another key feature in devel-oping certain kinds of placebo responses. Recent data indi-cate that conditioning procedures can be developedunconsciously resulting in a placebo eVect on hormonalsecretion, whereas verbally induced expectations (i.e., con-sciousness) produce a powerful modulation of pain toler-ance (Benedetti et al., 2003). Benedetti and colleagues foundthat verbally induced expectations of analgesia/hyperalge-sia were eVective in changing pain tolerance (i.e., placeboand nocebo eVects). In addition, they also tested the eVectsof opposing verbal suggestions on growth hormone andcortisol secretion and found no eVect. However, recallingan unconscious association between the subcutaneousinjection procedure (conditioned stimulus) and the phar-macological eVects of sumatriptan (unconditioned stimu-lus) resulted in a conditioned response aVecting growthhormone and adrenocortical secretion, similar to the phar-macological eVects of sumatriptan. The authors concludedthat when conscious expectation is engaged, consciousphysiological pain and motor mechanisms are malleable.On the other hand, when one seeks to inXuence functionthrough unconscious physiological processes, the placeboeVect appears to be mediated by behavioral conditioning.

In clinical settings it is likely that expectations andbehavioral conditioning processes contribute to a givenplacebo eVect. In this regard, conditioned/placebo eVectscould be experimentally dissected from expectation/placeboeVects, by comparing the magnitude of the placebo eVectfrom naïve subjects (i.e., expectation) with that of a groupof subjects who were submitted to behavioral conditioningprior to verbal suggestions (i.e., conditioning). Additionally,based on classical conditioning theory, the conditioningcomponent could be dissected from a given placebo eVect

434 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

by implementing latent inhibition (i.e., pre-exposing the sub-jects to the CS several times before the association phase)or extinction phenomena (i.e., after the association phase,several CS re-exposures without reinforcement). However,each particular experimental setting should be carefullyconsidered in order to be able to dissect the components,and the impact of each, in a particular placebo response.

2.1. Expectation framework

Based on placebo studies with Parkinson’s patients, ithas been proposed that the placebo eVect is mediated by thebrain reward circuitry (de la Fuente-Fernández et al., 2001;de la Fuente-Fernández and Stoessl, 2002; de la Fuente-Fernández et al., 2004). These authors emphasized thatreward expectations, such as expectation of clinicalimprovement, likely played an important role in the pla-cebo eVect. Expectation may be closely tied to a tonic acti-vation of nigro-striatal dopaminergic neurons, whichproject to the dorsal and ventral striatum and prefrontalcortices. Prior to reward, in the expectation phase, there isuncertainty and this is reXected in sustained dopaminergicactivation, which is maximized when the probabilityof reward is 0.5. It is known that with a 0.5 probability ofreward, 29% of dopaminergic cells are tonically activated(Fiorillo et al., 2003). Certainty, either of occurrence(pD1.0) or of non-occurrence (pD .0), leads to virtually notonic activation. There is also phasic dopaminergicactivation, which takes place after reward, and this isstrongest when the reward has come as a surprise. Havingsome uncertainty appears to heighten expectation in thisbrain reward circuitry model. Based on this information,the following neurobiological placebo mechanism hasbeen proposed (de la Fuente-Fernández, 2004; de laFuente-Fernández et al., 2004): when an interaction (e.g.,positive verbal suggestion) creates a reward possibility, cer-tain cortical neurons get activated in relation to rewardprobability. These cells send direct excitatory glutamatergicinputs to dopaminergic cell bodies along with indirectinhibitory gamma amino butyric acid inputs. The combina-tion of these signals arriving via direct and indirect path-ways at the dopaminergic neurons contributes to theprobability of tonic activation. Furthermore, it has beenreported that neurons in the prefrontal cortex, nucleusaccumbens and the caudate-putamen display tonic activa-tion during expectation of reward (Schultz, 1998). Also,there is increasing evidence that the placebo response is tiedto stimulation of the brain’s reward and motivation cir-cuitry (de la Fuente-Fernández, 2004). However, the behav-ioral eVects and the neural basis of expectation dependedon the level of certainty (Ploghaus et al., 2003). In thisregard, certain negative expectation has been associatedwith the emotional state of fear, activating the rostral ante-rior cingulate cortex and posterior cerebellum, and induc-ing hypoalgesia when a noxious stimulus is present. Incontrast, uncertain negative expectation is associated withanxiety, activating the ventromedial prefrontal cortex,

mid-cingulate cortex and hippocampus, and increasingpain sensitivity (Rhudy and Meagher, 2000). Recapitulat-ing, a conscious belief may induce a change in the activityof the prefrontal cortex, which feeds back to stimulatedopamine outXow in the medial forebrain bundle Xowingfrom the ventral tegmentum. Receptors in the dorsal stria-tum (caudate and putamen), in the ventral striatum(nucleus accumbens) and further up the line in the anteriorcingulate, as well as the medial orbitofrontal and the dorso-lateral prefrontal cortices, record this tonic dopaminechange. At the same time, activation of the prefrontal cor-tex through conscious expectation may stimulate midbrainopioid output systems in the periaqueductal grey matter. Inthis way, a placebo may beneWt conditions with pain, aswell as depression and Parkinson’s disease. Thus, it appearsthat a placebo-induced belief and conscious expectation setup a receptive Weld for the solace that accompanies thesense of safety in being assigned to a compassionate andprotective caregiver. The result is relief from the pain ofseparation distress and a reduction in the stress response. Inneurobiological terminology, when this occurs, there is pre-frontal cortical and paralimbic dampening of deeper limbicamygdalar fear conditioning (Fricchione and Stefano,2005).

2.2. Conditioning framework

The notion that a placebo is a conditioned stimulus isnot new. Pavlov (1927) described what appears to havebeen conditioned morphine eVects, and several investiga-tors in the 1950s elaborated theoretical frameworks identi-fying the placebo eVect as a conditioned response (Wolf,1950; Skinner, 1953).

In some cases the response to placebo treatment “lookslike” the response to a conditioned stimulus (CS) (Ader,1997). In behavioral terms, the physiological eVects uncon-ditionally elicited by pharmacologic agents represent theunconditioned response (UR), the drug itself being theunconditioned stimulus (US). The environmental or behav-ioral stimuli that are coincidentally or purposely associatedwith the voluntary or involuntary administration of a drugmay become the CS. Such conditioned stimuli should beneutral with respect to eliciting the unconditioned eVect ofthe active drug. These conditioned stimuli could include thebottle containing the drug, the room (or its color or smell)in which medication is taken, the place of treatment or theperson administrating it, and the injection or “pill” itself.The repeated association of CS–US eventually enables theCS to elicit a conditioned response (CR), in most of thecases an approximation of the response unconditionallyelicited by the US. Thus, the response to an inert or thera-peutically irrelevant substance (i.e., a placebo) has beendescribed as a conditioned response.

Several experiments, mostly with animals, have speciW-cally addressed the placebo eVect as a conditioning phenome-non. Pioneer Wndings indicated that, under appropriatetemporal conditions, rats would respond to an injection of

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 435

saline (CS) with suppression of the behavioral reactions(placebo eVect) that characterized the previous response toscopolamine hydrobromide (US) (Herrnstein, 1962). Simi-larly, after one or more injections (CS) of D-amphetamine(US), a conditioned increase in activity behavior (placeboeVect) was observed after an injection of saline, and it wasnoted that the magnitude of the conditioned/placeboresponse was a function of the number of times the activedrug had previously been injected (Ross, 1963; Pihl andAltman, 1971). Nowadays, much progress has been made inelucidating the neuronal and molecular events that take placeduring association and consolidation of the memory trace inclassical conditioning paradigms (Kandel et al., 1983; Tullyet al., 1990; Glanzman, 1995; Menzel and Muller, 1996;Berman and Dudai, 2001; Bermúdez-Rattoni, 2004).

Regarding immunomodulatory placebo eVects, it hasbeen reported that peripheral antigenic stimulation canwork as an US, and is thus able to be associated with aspeciWc external stimulus (CSDplacebo). Metalnikov andChorine are generally credited with having conducted theWrst studies on behaviorally conditioned immune eVects(Metalnikov and Chorine, 1926). However, Luk’ianenko(1961) cites the dissertation of Makukhin (1911) at theUniversity of St. Petersburg, and the report by Voronovand Riskin (1925) as perhaps the Wrst ones demonstrating“conditioned/placebo leukocytic reaction”. Considerableattention was given to the question of conditioned immuneeVects by Soviet investigators (reviewed in: Luk’ianenko,1961). For instance, it was reported that after repeatedinjections with saline (CS), rabbits displayed conditionedsuppression of the antibody response (placebo eVect) to asubsequent injection of the same antigen (Dolin and Kry-lov, 1952; Dolin et al., 1960). In 1975, new interest in thebehavioral conditioning of immunity was generated by theseminal publication of Ader and Cohen on conditionedimmunosuppression (Ader and Cohen, 1975), and this wasformally the beginning of PsychoNeuroImmunogy as amodern (inter)discipline. Afterwards, this and otherresearch groups continued documenting the capabilities of

the central nervous system to modulate peripheral immunefunctions on demand, when a previous association isevoked. To date, a number of innate and adaptive immuneresponses have been shown to be modulated by behavioralconditioning protocols (Table 1), in which conditionedimmunomodulating responses could be conceptualized asplacebo eVects (reviewed in: Brittain and Wiener, 1985;Ader and Cohen, 1991; Markovic et al., 1993; Ader andCohen, 2001; Hucklebridge, 2002; Pacheco-López et al.,2006). Various theoretical frameworks have been proposedto explain the mechanism behind this phenomena (Hiram-oto et al., 1997; Ader and Cohen, 2001; Bovbjerg, 2003;Pacheco-López et al., 2006). Additionally, neuroendocrineresponses in rodents and humans have been shown to besubject to behavioral conditioning, which might also con-tribute to certain immunomodulatory placebo eVects(Ader, 1976; Janz et al., 1996; Pacheco-López et al., 2004).For instance, human adrenal secretion can be elicited afterevoking a given association (CS:taste/US:dexamethasone,nD8) (Sabbioni et al., 1997).

3. Immunomodulatory placebo eVects in humans

So far, few attempts have been undertaken to speciWcallyinvestigate placebo eVects directly modulating peripheralimmune functions in human subjects. However, severalmeta-analyses and reviews on placebo/nocebo eVects indi-cate the susceptibility of various immune-related pathologi-cal conditions to “inert” treatments (Table 2). Among suchpathologies, mucosal inXammatory diseases seem to bestrongly modulated by placebo treatments, e.g., ulcerativecolitis (26.7%: Ilnyckyj et al., 1997), duodenal ulcer (36.2–44.2%: de Craen et al., 1999b), and irritable bowel syn-drome (40%: Patel et al., 2005). In this regard, it hasrecently been proposed that the placebo eVects may bemediated by alteration of one or more components of theacute-phase response (Evans, 2005). However, otherimmune-related diseases, such as multiple sclerosis (LaMantia et al., 1996), Crohn’s disease (Su et al., 2004) and

Table 1Immune parameters aVected by behavioral conditioning paradigms

Conditioned stimulus Unconditioned stimulus Conditioned response

Taste/odor Immunosuppressant drugs Antibody productionLymphocyte proliferationHypersensitivityAllergic responseAllograft rejectionNK-cell activityCytokines

Taste/odorAuditory/visualTouch

Immunostimulating drugs/antigens Skin hypersensitivityNK-cell activityCTL activityNeutrophil activityAntibody productionHistamine releaseAnaphylaxisComplement

436 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

chronic fatigue syndrome (Cho et al., 2005), also seem to besusceptible to placebo treatments.

3.1. Placebo eVect in allergy/asthma

Asthma has been consistently reported to be inXu-enced by psychological factors, including placebo treat-ment. Joyce et al. (2000) reported a pooled placebo eVectin stable asthmatics during long-term drug therapy trials.Out of 143 asthmatic patients receiving placebo treat-ment, approximately 6% showed clinically relevantimprovement in pulmonary function. In this context,since the 19th century, anecdotic case studies havereported the occurrence of allergic symptoms in theabsence of allergens, provoked simply by diVerent place-bos, e.g., a picture of a hay Weld or by an artiWcial rose(MacKenzie, 1886). Several decades later, Ikemi andNakagawa (1962) reported a placebo dermatitis responsein adolescent male subjects elicited either by expectation(verbal information of poisonous tree application: RhusVenicifera, n D 5) or as a result of evoking (blue solutionapplication) a speciWc association (CS: blue solution/US:2% raw extract R. Venicifera application, nD 4). Regard-ing asthma, it was reported that almost 50% of asthmat-

ics patients (n D 40) displayed asthma-like reactions if agiven inhalation was accompanied by verbal suggestionof allergen inhalation (Luparello et al., 1968; Luparelloet al., 1970). Additionally, in 30% of the subjects suchsymptoms diminish dramatically after inhalation of afalse bronchodilatator. In both cases the subjects werejust exposed to physiological saline (placebo) and thediVerence in symptoms could only be attributable to theinformation provided (expectations), and quite likely toprevious experiences (conditioning). Recently, theseresults were basically replicated and it was determinedthat the suggestibility of each subject directly correlateswith the magnitude of the placebo eVect (5 out of 8 sug-gestible subjects but only 1 out of 9 suggestion-resistantsubjects showed a reduction in forced expiratory volume)(Leigh et al., 2003). In addition to placebo eVects inducedby expectation, asthma seems to be highly sensitive toassociative learning protocols. In an early case report,two asthmatic patients suVering from skin sensitivities tohouse-dust extract and grass pollen were exposed to theseallergens by inhalation (Dekker et al., 1957). After aseries of conditioning trials, they experienced allergicattacks after inhalation of the neutral solvent used todeliver the allergens. This work showed not only fast

Table 2Placebo eVects in inXammatory and immune-related diseases

a Meta-analysis.b Review.

Disease Placebo eVects Methodological issues Reference

Asthma Pooled placebo eVect in stable asthmatics during long-term drug therapy trials was small but measurable. A modest number of patients receiving placebo (6%) showed changes in pulmonary function that can be considered clinically relevant.

33 randomized double-blind placebo-controlled clinical trials, 1243 patients.

Joyce et al. (2000)a

Cancer Placebo treatment associated with slight improvement in symptoms such as pain (9%) and appetite (20%) but rarely with positive tumor response (2.4%). A signiWcant part (10–60%) of nocebo eVects is associated with “inert” treatments.

37 randomized placebo-controlled trials: Pain 12 trials, 405 patients; Appetite 5 trials, 368 patients; Tumor size 10 trials, 464 patients.

ChvetzoV and Tannock (2003)b

Crohn’s disease Mean placebo response rate was 19%. Number of oYce visits, duration of study and disease severity at entry inXuenced placebo response.

32 placebo-controlled trials, 1047 patients.

Su et al. (2004)a

Chronic fatigue syndrome

Pooled placebo response was 19.6%. Psychological-psychiatric interventions showed a low placebo response (14%), whereas infectious-immunological and alternative-complementary interventions showed a high placebo response (24%).

29 randomized placebo-controlled trials, 1016 patients.

Cho et al. (2005)a

Duodenal ulcer Pooled healing rate in placebo groups ranged from 36.2 to 44.2%. 79 randomized placebo-controlled trials, 3325 patients.

de Craen et al.(1999a,b)a

Irritable bowel syndrome

Placebo response ranged from 16 to 71% with a population-weighted average of 40%. Study duration and the number of visits/study duration did not signiWcantly inXuence the placebo response.

45 randomized placebo-controlled trials, 3193 patients.

Patel et al. (2005)a

Multiple sclerosis Relapse frequency in remitting-relapsing MS patients decreased compared to baseline. Reduction in the exacerbation rate ranged from 11 to 50% during follow-up. In progressive MS patients, disease evolution during the trials was similar to that observed during the natural course of the disease.

Randomized placebo-controlled trials, 953 patients.

La Mantia et al. (1996)b

Ulcerative colitis Placebo remission rate was 9.1% with a placebo beneWt rate of approximately 26.7%.

38 double-blind placebo-controlled trials.

Ilnyckyj et al. (1997)a

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 437

conditioning of the asthmatic attack (CR), but also tena-cious retention, i.e., lack of extinction. This observation,together with data from animal experiments, resulted inthe early hypothesis that asthma could be conceived of asa learned response (Turnbull, 1962). This view was fur-ther supported by a conditioning protocol (CS: taste/US:dust mite allergen) in nine patients with allergic rhinitis(Gauci et al., 1994). After the association phase, elevatedmast cell tryptase in mucosa was observed when an intra-nasal saline application was given simultaneously withthe CS. Another type of allergic reaction, the delayedtype hypersensitivity response, was tested in sevenhealthy volunteers who received Wve monthly tuberculinskin tests (Smith and McDaniel, 1983). In this condition-ing protocol both tuberculin (US) and saline wereinjected; while the latter one was taken from a green vial(CS¡), tuberculin was drawn from a red vial (CS+). Onthe test day, the color labeling of the substances wasreversed. Although the saline injections did not induce askin reaction (erythema and induration), the severity ofthe symptoms was signiWcantly blunted in all the subjectstested when the tuberculin was drawn from the green vial(placebo eVect). However, a similar protocol using vari-ous allergens (e.g., mite dust, fur) taken from coloredvials did not result in conditioned modulation of skinreactions in the 15 subjects tested (Booth et al., 1995).

3.2. Placebo eVects in oncology

ChvetzoV and Tannock (2003) have recently reviewedplacebo eVects in oncology, summarizing results from 37clinical trials that indicated improvement in symptomssuch as pain (9%; 12 clinical trials, nD 405) and appetite(20%; 5 clinical trials, n D 365), but rarely with positivetumor response. Objective response rates (i.e., reductionof tumor size according to World Health Organizationcriteria) to placebo were low but not zero, ranging from2% to 7% (10 clinical trials, nD 464). For instance, in aninterferon �-1b clinical trial (n D 90) for renal cancer aplacebo eVect of 6.6% was reported (i.e., tumor sizereduction); by comparison, the response rate was only4.4% in the interferon �-1b treatment group (Gleaveet al., 1998). Concerning nocebo eVects, 10–60% of cancerpatients were found to develop negative responses (e.g.,nausea and vomiting, abdominal pain, lethargy, drymouth, diarrhea) to placebo therapy (ChvetzoV andTannock, 2003). In this regard, associative learning hasbeen consistently reported in the context of cancer treat-ment, particularly chemotherapy (Bovbjerg, 2003). Che-motherapy agents (e.g., cyclophosphamide) generallyhave immunosuppressive eVects. These agents are typi-cally administered in cycles, with each outpatient treat-ment infusion followed by a period of recovery prior tothe next infusion. From a conditioning perspective, clinictreatment visits can be viewed as “association trials” inwhich the distinctive salient features of the clinic environ-ment (CS) are contingently paired with the infusion of

agents (e.g., cyclophosphamide; US) that have eVects onthe immune system. For instance, immune function wasassessed in 20 cancer patients in the hospital prior to che-motherapy and compared with assessments conducted athome. Proliferative responses to T-cell mitogens werelower for cells isolated from blood samples taken in thehospital (i.e., after recall) than for home samples(Bovbjerg et al., 1990). These results were replicated in 22ovarian patients (Lekander et al., 1995) and 19 pediatricpatients receiving chemotherapy (Stockhorst et al., 2000).However, often chemotherapy patients develop condi-tioned/nocebo nausea (Andrykowski, 1988; Bovbjerget al., 1990; Morrow et al., 1991; Matteson et al., 2002),anxiety (Jacobsen et al., 1993; DiLorenzo et al., 1995) andfatigue (Bovbjerg et al., 2005) responses to reminders ofchemotherapy. These conditioned/nocebo nausea andanxiety responses can also be elicited by thoughts andimages of chemotherapy (Redd et al., 1993; Dadds et al.,1997), raising the possibility that conditioned/noceboeVects may aVect patients during the course of normal lifefor years after treatment.

3.3. Placebo eVects of surgery

The psychosocial context surrounding surgery has beenfound to elicit strong expectations and subsequent placeboeVects (Finneson, 1969; Moseley et al., 2002). Additionally,the anti-inXammatory and analgesic eVects attributable tothe symbolic power of surgical treatment have been clearlydemonstrated in studies regarding the bilateral surgicalextraction of lower third molars (Hashish et al., 1986;Hashish et al., 1988; Ho et al., 1988). Thus, at least somepart of the placebo eVects of surgery might be explained bypsycho-neuro-immune interactions.

The classic example of the placebo eVects of surgerycomes from two studies on the ligation of the bilateralinternal mammary arteries as treatment for angina (Cobbet al., 1959; Dimond et al., 1960). Patients receiving shamsurgery did as well - with 80% of patients substantiallyimproving - as those receiving the active procedure intrials or in general practice. It is noteworthy that theseeVectiveness rates (and those reported by the proponentsof the procedure at the time) are much the same as thoseachieved by contemporary treatments, such as coronaryartery bypass or �-blockers.

The placebo eVect of surgery for Ménière’s disease wasinvestigated by comparing the eVect of a regular endolym-phatic shunt with the eVect of a placebo operation (regularmastoidectomy; nD30). Minor diVerences were observedbetween the active and the placebo group. However, thegreatest diVerence in symptoms was found when comparingpre- and post-operative scores, as both groups showed sig-niWcant improvement; 70% in the placebo group (Thomsenet al., 1981b; Thomsen et al., 1981a). Strikingly, the thera-peutic eVect of such placebo surgery has been followed forup to 9 years after treatment (Thomsen et al., 1983; Bretlauet al., 1989).

438 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

Moseley et al. (1996) initially reported that arthroscopicsurgery for osteoarthritis of the knee displayed pronouncedplacebo eVects. However, the low number of patientsenrolled (arthroscopic debridement nD 2; lavage nD3; pla-cebo surgery nD5) limited any serious conclusion.Recently, these authors have found similar strong and long-lasting placebo eVects of surgery on osteoarthritis in theknee in an extended follow-up study (arthroscopic debride-ment nD 59; lavage nD61; placebo surgery nD 60)(Moseley et al., 2002). In both reports, arthroscopic surgerywas simulated under sedation (i.e., placebo surgery), mak-ing small incisions in the leg, but not removing any tissue.During a two-year follow-up, researchers found no diVer-ences between the three groups. All patients reportedimprovement in the symptoms relating to pain and abilityto use their knees. Throughout this time, patients did notknow whether they had received the “real” or placebo sur-gery. However, patients who had received actual surgerydid not report less pain or better functioning of their kneesthan the placebo group. Moreover, the placebo groupreported a better outcome than the patients who hadundergone the debridement procedure.

It is important to note that the American MedicalAssociation (AMA) has recently issued ethical guidelinesregarding surgical placebo controls (Tenery et al., 2002)“Surgical placebo controls should be used only when no othertrial design will yield the requisite data and should always beaccompanied by a rigorous informed consent process and acareful consideration of the related risks and beneWts”.

3.4. Placebo eVects on immune cell number and functions

Only a few human studies have so far tried to aVectimmune parameters on the cellular level by employingbehavioral conditioning procedures. Based on the knowl-edge that adrenaline administration leads to the immediatemobilization of leukocytes in the periphery, especially ofNK cell numbers with simultaneous augmentation of theirlytic activity (Benschop et al., 1996; Schedlowski et al.,1996), one research group assessed the conditionability ofnatural killer (NK) cell numbers and their lytic activity inhealthy volunteers. Although positive results were reportedafter evoking a taste (CS) — adrenaline (US) association(Buske-Kirschbaum et al., 1992; Buske-Kirschbaum et al.,1994), these eVects could not be replicated by other researchgroups (Kirschbaum et al., 1992). The eYcacy of a condi-tioning protocol was also tested in multiple sclerosispatients, for whom four monthly cyclophosphamideinfusions (US) were contingently paired with the taste ofanise-Xavored syrup (CS) (Giang et al., 1996). Long-termtreatment with cyclophosphamide decreases blood leuko-cyte numbers often leading to leukopenia. Interestingly,after six months of administering the placebo infusionpaired with the drink, 8 out of 10 patients showed a condi-tioned/placebo reduction in peripheral leukocytes numbers.In addition, by pairing s.c. interferon-� injections (US) witha distinctively Xavored drink (CS), it was possible to induce

an elevation of neopterin and quinolinic acid serum levelsafter evoking such an association in healthy volunteers(nD10) (Longo et al., 1999). However, it has been hypothe-sized that more than a single associative learning trial,pairing a distinctive taste (CS) with interferon-� injections(US), would be necessary in order to produce immune con-ditioned eVects (Goebel et al., 2005). This view is supportedby experimental data for healthy male volunteers (nD18)where the immunosuppressive drug cyclosporine A (US)was paired four times with a distinctively Xavored/coloredsolution (CS) (Goebel et al., 2002), inducing taste-immuneassociative learning. The immunopharmacological mecha-nism of cyclosporine A involves its binding to cyclophilins,which leads to intracellular phosphatase calcineurin inhibi-tion, then selectively reducing the expression of certaincytokines (e.g., interleukin-2: IL-2 and interferon-�: IFN-�),which Wnally results in speciWc suppression of T-cell func-tion (Bukrinsky, 2002). After association, the mere re-expo-sure to the drink (CS) induced a conditioned/placeboinhibition of ex vivo cytokine (IL-2 and IFN-�) mRNAexpression and cytokine release, as well as of the prolifera-tive responsiveness of human peripheral blood lympho-cytes, similar to the drug eVect.

In summary, experimental data demonstrating immuno-modulatory placebo eVects, whether based on expectationsor conditioning, do indeed support and encourage the useof individualized behavioral therapy as a supportive strat-egy in pharmacological/surgical approaches. Such therapiesshould be designed to enhance therapeutic placebo eVectsand reduce nocebo eVects associated with the maintreatment.

4. Neurobiology of the immunomodulatory placebo eVect

Only a limited number of studies have as yet analyzedthe neurobiological mechanisms responsible for the pla-cebo eVects on peripheral immune functions. Based ontaste-visceral associative learning and reward paradigms,speciWc and discrete neural networks have been discovered.

4.1. Neurobiology of the immunomodulatory placebo/conditioned eVects

The naturalistic associability of food/drink ingestionwith its possible immune consequences has been experi-mentally appraised in rodents and humans employing theconditioned taste aversion paradigm (Garcia et al., 1955).Conditioned taste aversion/avoidance is a type of associa-tive conditioning in which the subjects learn to associate ataste with delayed malaise (Bermúdez-Rattoni, 2004). Thislearning has been conserved across the animalia kingdom(Kawai et al., 2004; Paradis and Cabanac, 2004; Marellaet al., 2006; Schedlowski, 2006), including humans (Garband Stunkard, 1974), attesting to its highly adaptive valuein food selection strategies. In this regard, reduced ingestivebehavior may be only part of a complex and diverse reper-tory of physiological responses that the individual evokes

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 439

to avoid, reject and/or prepare the organism to counteractthe unconditioned eVects (Niemi et al., in press). A discreteneural network involved in taste-visceral associative learn-ing has already been described, mainly including sensoryand hedonic neural pathways (Sewards and Sewards, 2002;Sewards, 2004). Such a neural circuit consistently includesthe nucleus tractus solitary, the parabrachial nucleus,medial thalamus, amygdala and insular cortex (Yamamotoet al., 1994). In particular, the insular cortex is essential forthe acquisition and retention of this associative learning(Bermudez-Rattoni and McGaugh, 1991; Cubero et al.,1999), and it has been postulated that the insular cortexmay integrate gustatory and visceral stimuli (Sewards andSewards, 2001). More recently, using the neuronal activitymarker c-Fos, it was possible to conWrm the preponderantrole of the insular cortex in conditioned/placebo antibodyproduction (Chen et al., 2004), in agreement with a previousreport (Ramírez-Amaya and Bermúdez-Rattoni, 1999).Regarding other forebrain structures, the amygdala seemsto play an important role during formation of aversiveingestive associations (Reilly and Bornovalova, 2005), andis also relevant for limbic-autonomic interaction (Swansonand Petrovich, 1998). A series of reports has indicated thatthe insular cortex and the amygdala are key structures inconditioned/placebo immunosuppression after evokingtaste-cyclophosphamide association (Ramírez-Amayaet al., 1996; Ramírez-Amaya et al., 1998). Additionally, ithas been proposed that the ventromedial hypothalamicnucleus, widely recognized as a satiety center (Vettor et al.,2002), is intimately associated with sympathetic facilitationin peripheral tissues (Saito et al., 1989), including modula-tion of peripheral immune reactivity (Okamoto et al., 1996).In agreement with previous reports (Ramírez-Amaya et al.,1996; Ramírez-Amaya et al., 1998; Ramírez-Amaya andBermúdez-Rattoni, 1999), we have identiWed the neuralsubstrates involved in behaviorally conditioned/placeboimmunosuppression (CS: saccharin/US: cyclosporine A) inrats (Pacheco-López et al., 2005). The conditioned/placeboeVect on the immune system, reducing splenocyte respon-siveness and cytokine production (IL-2 and IFN-�), wasaVected by brain excitotoxic lesions; this shows that theinsular cortex is essential to acquiring and evoking this con-ditioned/placebo response. In contrast, the amygdala seemsto mediate the input of visceral information necessary atacquisition time, whereas the ventromedial hypothalamicnucleus appears to participate in the output pathway to theimmune system, needed to evoke the behaviorally condi-tioned immune response.

Using a pharmacological approach, the neurochemicalfeatures of the conditioned/placebo eVect enhancing NKcell activity in rodents have been described. Central cate-cholamines seem to be essential, and glutamate - but notGABA - is also required at the recall stage (Hsueh et al.,1999; Kuo et al., 2001). In addition, it has been demon-strated that cholinergic, as well as serotonergic, centralsystems are required at the association and recall stages(Hsueh et al., 2002).

In addition to classical neurotransmitters, cytokineshave been demonstrated to play an important role withinthe central nervous system (CNS), modulating neuronaland glial function in non-pathological settings such aslearning and memory processes (Balschun et al., 2004;Dantzer, 2004; Tonelli et al., 2005). SpeciWcally, pro-inXam-matory cytokines, such as IL-1, IL-6 & TNF-�, have beenshown to modulate spatial learning tasks, as well as long-term potentiation phenomena (Gibertini, 1996; Schneideret al., 1998; Fiore et al., 2000; Banks et al., 2001; Matsum-oto et al., 2001; Rachal Pugh et al., 2001; Lynch, 2002; Mat-sumoto et al., 2002). In this sense, it is plausible thatcytokines might be an important factor in the associativeprocesses occurring during the behavioral conditioning ofimmune functions. Apart from these neuromodulatoryproperties, pro-inXammatory cytokines seem to play animportant role in the aVerent pathway between the immunesystem and the CNS (Besedovsky and del Rey, 1996; Turn-bull and Rivier, 1999; Dantzer, 2004). Therefore, it can behypothesized that central cytokines act as mediators in thebrain during an “immune-sensing” phase at the acquisitiontime of behaviorally conditioned immunomodulating para-digms. These hypotheses are supported by observationsthat 1) receptors for these pro-inXammatory cytokines areexpressed in the CNS (Szelényi, 2001; Sredni-Kenigsbuch,2002), 2) peripheral immune changes aVect central cytokineproduction and cytokine receptor expression in the brain(Pitossi et al., 1997; Del Rey et al., 2000), and 3) cytokinescan act as unconditioned stimuli to induce conditionedtaste aversion/avoidance (Tazi et al., 1988; Dyck et al.,1990; Janz et al., 1991; Hiramoto et al., 1993).

To our knowledge, apart from these reports, there hasbeen no systematic attempt to elucidate the neural sub-strates underlying placebo immunotherapeutic eVects basedon associative learning.

4.2. Neurobiology of immunomodulatory placebo/expectation eVects

As mentioned above, it has been suggested that reward/dopamine neural circuitry is mainly responsible for theexpectation/placebo eVects on neurological disorders (de laFuente-Fernández, 2004), and that certainty is a critical fac-tor inXuencing behavioral responses and neural circuitryactivated by expectations (Ploghaus et al., 2003). In thisregard, the fact that rewarding activities, (such as socialsupport), reduce chronic stress and its deleterious eVects onimmune functions in humans has been systematically docu-mented (Andersen et al., 2004; Glaser and Kiecolt-Glaser,2005). However, the eVect of the level of certainty on theimmunomodulatory placebo/expectation eVects has notbeen yet addressed.

Regarding the immunomodulatory mechanisms ofreward; it has been proposed that reduced activity of thehypothalamic-pituitary-adrenal axis is the major mecha-nism behind the beneWcial eVect of social support. How-ever, experimental models of reward have shown that this

440 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

neural process can also aVect peripheral immune responsesvia direct sympathetic innervations of peripheral immuneorgans (Sakic and Vlajkovic, 1990; Wenner et al., 2000).Furthermore, it has been demonstrated that natural killercell activity is controlled, in part, by tonic inhibition fromthe splenic nerve (Okamoto et al., 1996) and sympatheticsplenic innervation seems to be under the central control ofthe ventromedial hypothalamus (Katafuchi et al., 1993;

Katafuchi et al., 1994). While electrical stimulation of theventromedial hypothalamus has been found to arouse sym-pathetic activity (Saito et al., 1989), the lateral hypothala-mus seemed to do the opposite (Bernardis and Bellinger,1993). Thus, the lateral hypothalamus may have immu-noenhancing properties (Wrona and Trojniar, 2003) whichin part antagonize the ventromedial hypothalamus, therebyreducing sympathetic tone inhibition. Importantly, such

Fig. 1. Neurobiology of the immunomodulatory placebo eVects. Conscious expectation and unconscious behavioral conditioning processes appear to bethe major neurobiological mechanisms of placebo eVects. These processes are capable of releasing endogenous neurotransmitters that mimic the expectedor conditioned pharmacological eVects, i.e., the placebo eVect. A discrete neural network has been identiWed that modulates peripheral immune functionsand which can possibly be elicited by the psychosocial context of placebo treatments (see text). Major neural eVerent pathways, through which expecta-tions and/or memories could aVect peripheral immune functions, are the neocortical–sympathetic–immune axis, including limbic and hypothalamic relays;the hypothalamus–pituitary–adrenal immune axis, and the brain stem–vagus–cholinergic pathway. Placebo eVects can thus beneWt end organ functioningand the overall health of the individual through the healing power of belief, positive expectations and conditioning processes. ACTH, adrenocorticotropichormone; Ach, acetylcholine; NE, noradrenaline. Note: The main propose of this sketch is to focus on neural substrates of the immunomodulating pla-cebo eVect which, in this case, takes precedence over anatomical accuracy. Hyphened words depict virtual anatomical positions in diVerent depths of the

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 441

hypothalamic regulation of sympathetic activity seems tobe modulated by the insular cortex (Allen et al., 1991;Cechetto and Chen, 1992; Oppenheimer et al., 1992;Butcher and Cechetto, 1998), providing neuroanatomicalsubstrates that might explain the link between placebo/con-ditioned and placebo/expectation responses. On the periph-ery, much progress has been made in elucidatingphysiological and molecular events, which have revealedthe complexity of neuro-immune interactions (Straub,2004).

Recapitulating, a discrete neural network has meanwhilebeen identiWed that modulates peripheral immunefunctions, and which can possibly be elicited by the psycho-social context of placebo treatments to induce its immuno-modulatory eVects. Major neural eVerent pathwaysthrough which expectations and/or memories could aVectperipheral immune functions are depicted in Fig. 1: theneocortical-sympathetic-immune axis (Elenkov et al., 2000;Moshel et al., 2005; Tuohy, 2005), including limbic (Haasand Schauenstein, 1997) and hypothalamic relays (Butcherand Cechetto, 1998; Wrona, 2006); the hypothalamus-pitui-tary-adrenal immune axis (Besedovsky and del Rey, 1991;Besedovsky and del Rey, 1996), and the brain stem-vagus-cholinergic pathway (Tracey, 2002; Pavlov and Tracey,2005). Placebo eVects can, therefore, beneWt end organfunctioning and the overall health of the individual throughthe healing power of belief, positive expectations and con-ditioning processes. Thus, under appropriate conditions,“expectations and associations may heal”.

5. Conclusion and outlook

Recent clinical and experimental data clearly demon-strate that the placebo phenomenon is real and not amyth. The analysis of the diVerent components of pla-cebo eVects, and the mechanisms driving them, is a greatchallenge for future interdisciplinary research. Only if wecan gain a better understanding of the underlyingpsychological, neurobiological, endocrine and immuno-logical mechanisms, can we make use of placebo healingpower for the patient’s beneWt, at the same time beingaware of the pathological conditions under which pla-cebo eVects are modiWed.

To date, research on placebo eVects with regard toimmune-related diseases is scarce, but there are consistentindications that skin and mucosal inXammatory diseasesare strongly modulated by placebo treatments. Currentlyseveral research groups are attempting to elucidate theunderlying mechanism of placebo eVects in variousimmune-related diseases (e.g., psoriasis, asthma, irritablebowel syndrome and wound healing: http://nccam.nih.gov),and this will oVer important insights into the immunomod-ulatory capacities of placebo treatment. It would be prefer-able, however, if studies investigating placebo eVects inimmune functions dissected expectation from classical con-ditioning, in order to estimate the contribution of eachcomponent in a given placebo eVect.

In the near future, placebo treatment may no longer belimited to the control function in clinical drug trials, neces-sary though that is, but may be systematically prescribed toincrease the eYcacy of drug treatment. As initially pro-posed by Ader (1989, 1997), using the placebo response tosupport pharmacological regimens will likely maximizetherapeutic eYcacy, at the same time reducing unwanteddrug side eVects to the beneWt of the patient and, last butnot least, saving costs.

References

Ader, R., 1976. Conditioned adrenocortical steroid elevations in the rat.J. Comp. Physiol. Psychol. 90, 1156–1163.

Ader, R., 1989. Conditioning eVects in pharmacotherapy and the incom-pleteness of the double-blind, crossover design. Integrative Psychiatry6, 165–170.

Ader, R., 1997. The role of conditioning in pharmacotherapy. In: Harring-ton, A. (Ed.), The Placebo EVect: An Interdisciplinary Exploration.Harvard University Press, Cambridge, pp. 138–165.

Ader, R., Cohen, N., 1975. Behaviorally conditioned immunosuppression.Psychosom Med. 37, 333–340.

Ader, R., Cohen, N., 1991. The inXuence of conditioning on immuneresponses. In: Ader, R., Felten, D., Cohen, N. (Eds.), Pschoneuroimmu-nology. Academic Press, New York, pp. 611–646.

Ader, R., Cohen, N., 2001. Conditioning and Immunity. In: Ader, R., Fel-ten, D., Cohen, N. (Eds.), Psychoneuroimmunology, vol. 2nd. Aca-demic Press, New York, pp. 3–34.

Allen, G.V., Saper, C.B., Hurley, K.M., Cechetto, D.F., 1991. Organizationof visceral and limbic connections in the insular cortex of the rat.J. Comp. Neurol. 311, 1–16.

Amanzio, M., Benedetti, F., 1999. Neuropharmacological dissection ofplacebo analgesia: expectation-activated opioid systems versusconditioning-activated speciWc subsystems. J. Neurosci. 19, 484–494.

Andersen, B.L., Farrar, W.B., Golden-Kreutz, D.M., Glaser, R., Emery,C.F., Crespin, T.R., Shapiro, C.L., Carson, W.E., 2004. Psychological,behavioral, and immune changes after a psychological intervention: aclinical trial. J. Clin. Oncol. 22, 3570–3580.

Andrykowski, M.A., 1988. DeWning anticipatory nausea and vomiting:diVerences among cancer chemotherapy patients who report pretreat-ment nausea. J. Behav. Med. 11, 59–69.

Bailar 3rd, J.C., 2001. The powerful placebo and the Wizard of Oz. N.Engl. J. Med. 344, 1630–1632.

Balschun, D., Wetzel, W., Del Rey, A., Pitossi, F., Schneider, H., Zuschrat-ter, W., Besedovsky, H.O., 2004. Interleukin-6: a cytokine to forget.FASEB J. 18, 1788–1790.

Banks, W.A., Farr, S.A., La Scola, M.E., Morley, J.E., 2001. Intravenoushuman interleukin-1alpha impairs memory processing in mice:dependence on blood–brain barrier transport into posterior division ofthe septum. J. Pharmacol. Exp. Ther. 299, 536–541.

Beecher, H.K., 1955. The powerful placebo. J. Am. Med. Assoc. 159,1602–1606.

Benedetti, F., 2005. The importance of considering the eVects of perceivedgroup assignment in placebo-controlled trials. Eval. Health Prof. 28, 5–6.

Benedetti, F., Arduino, C., Costa, S., Vighetti, S., Tarenzi, L., Rainero, I.,Asteggiano, G., 2006. Loss of expectation-related mechanisms in Alz-heimer’s disease makes analgesic therapies less eVective. Pain 121,133–144.

Benedetti, F., Colloca, L., Torre, E., Lanotte, M., Melcarne, A., Pesare, M.,Bergamasco, B., Lopiano, L., 2004. Placebo-responsive Parkinsonpatients show decreased activity in single neurons of subthalamicnucleus. Nat. Neurosci. 7, 587–588.

Benedetti, F., Mayberg, H.S., Wager, T.D., Stohler, C.S., Zubieta, J.K.,2005. Neurobiological mechanisms of the placebo eVect. J. Neurosci.25, 10390–10402.

442 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

Benedetti, F., Pollo, A., Lopiano, L., Lanotte, M., Vighetti, S., Rainero, I., 2003.Conscious expectation and unconscious conditioning in analgesic, motor,and hormonal placebo/nocebo responses. J. Neurosci. 23, 4315–4323.

Benschop, R.J., Rodriguez-Feuerhahn, M., Schedlowski, M., 1996. Cate-cholamine-induced leukocytosis: early observations, current research,and future directions. Brain Behav. Immun. 10, 77–91.

Benson, H., 1997. The nocebo eVect: history and physiology. Prev. Med.26, 612–615.

Berman, D., Dudai, Y., 2001. Memory extinction, learning anew, andlearning the new: dissociations in the molecular machinery of learningin cortex. Science 291, 2417–2419.

Bermúdez-Rattoni, F., 2004. Molecular mechanisms of taste-recognitionmemory. Nat. Rev. Neurosci. 5, 209–217.

Bermudez-Rattoni, F., McGaugh, J.L., 1991. Insular cortex and amygdalalesions diVerentially aVect acquisition on inhibitory avoidance andconditioned taste aversion. Brain Res. 549, 165–170.

Bernardis, L.L., Bellinger, L.L., 1993. The lateral hypothalamic area revis-ited: neuroanatomy, body weight regulation, neuroendocrinology andmetabolism. Neurosci. Biobehav. Rev. 17, 141–193.

Besedovsky, H.O., del Rey, A., 1991. Feed-back interactions betweenimmunological cells and the hypothalamus–pituitary–adrenal axis.Neth. J. Med. 39, 274–280.

Besedovsky, H.O., del Rey, A., 1996. Immune-neuro-endocrine interac-tions: facts and hypotheses. Endocr. Rev. 17, 64–102.

Boissel, J.P., Philippon, A.M., Gauthier, E., Schbath, J., Destors, J.M., 1986.Time course of long-term placebo therapy eVects in angina pectoris.Eur. Heart J. 7, 1030–1036.

Booth, R.J., Petrie, K.J., Brook, R.J., 1995. Conditioning allergic skinresponses in humans: a controlled trial. Psychosom. Med. 57, 492–495.

Bovbjerg, D., 2003. Conditioning, cancer, and immune regulation. BrainBehav. Immun. 17 (Suppl 1), S58–S61.

Bovbjerg, D.H., Montgomery, G.H., Raptis, G., 2005. Evidence for classi-cally conditioned fatigue responses in patients receiving chemotherapytreatment for breast cancer. J. Behav. Med. 28, 231–237.

Bovbjerg, D.H., Redd, W.H., Maier, L.A., Holland, J.C., Lesko, L.M.,Niedzwiecki, D., Rubin, S.C., Hakes, T.B., 1990. Anticipatory immunesuppression and nausea in women receiving cyclic chemotherapy forovarian cancer. J. Consult. Clin. Psychol. 58, 153–157.

Branthwaite, A., Cooper, P., 1981. Analgesic eVects of branding in treat-ment of headaches. Br. Med. J. (Clin Res Ed) 282, 1576–1578.

Braverman, A.J., 1986. Dose validation and study design criteria in currentcimetidine studies. Clin. Ther. 8 (Suppl A), 49–56.

Bretlau, P., Thomsen, J., Tos, M., Johnsen, N.J., 1989. Placebo eVect in sur-gery for Menière’s disease: nine-year follow-up. Am. J. Otol. 10, 259–261.

Brittain, R., Wiener, N., 1985. Neural and Pavlovian inXuences on immu-nity. Pavlov. J. Biol Sci. 20, 181–194.

Brody, H., 1980. Placebos and the Philosophy of Medicine: Clinical, Con-ceptual, and Ethical Issues, ed.. Chicago University Press, Chicago.

Brody, H., 1985. Placebo eVect: an examination of Grumbaum’s deWnition.In: White, L.T.B., Schwartz, G.E. (Eds.), Placebo: Theory, Research,and Mechanismspp. The Guilford Press, New York, pp. 37–58.

Bukrinsky, M., 2002. Cyclophilins: unexpected messengers in intercellularcommunications. Trends Immunol. 23, 323–325.

Buske-Kirschbaum, A., Kirschbaum, C., Stierle, H., Jabaij, L., Hellham-mer, D., 1994. Conditioned manipulation of natural killer (NK) cells inhumans using a discriminative learning protocol. Biol. Psychol. 38,143–155.

Buske-Kirschbaum, A., Kirschbaum, C., Stierle, H., Lehnert, H., Hellham-mer, D., 1992. Conditioned increase of natural killer cell activity(NKCA) in humans. Psychosom. Med. 54, 123–132.

Butcher, K.S., Cechetto, D.F., 1998. Neurotransmission in the medullamediating insular cortical and lateral hypothalamic sympatheticresponses. Can J. Physiol. Pharmacol. 76, 737–746.

Byerly, H., 1976. Explaining and exploiting placebo eVects. Perspect. Biol.Med. 19, 423–436.

Cechetto, D.F., Chen, S.J., 1992. Hypothalamic and cortical sympa-thetic responses relay in the medulla of the rat. Am. J. Physiol. 263,R544–R552.

Chen, J., Lin, W., Wang, W., Shao, F., Yang, J., Wang, B., Kuang, F.,Duan, X., Ju, G., 2004. Enhancement of antibody production andexpression of c-Fos in the insular cortex in response to a conditionedstimulus after a single-trial learning paradigm. Behav. Brain Res. 154,557–565.

Cho, H.J., Hotopf, M., Wessely, S., 2005. The placebo response in the treat-ment of chronic fatigue syndrome: a systematic review and meta-analy-sis. Psychosom Med. 67, 301–313.

ChvetzoV, G., Tannock, I.F., 2003. Placebo eVects in oncology. J. Natl.Cancer Inst. 95, 19–29.

Cobb, L.A., Thomas, G.I., Dillard, D.H., Merendino, K.A., Bruce, R.A.,1959. An evaluation of internal-mammary-artery ligation by a double-blind technic. N Engl. J. Med. 260, 1115–1118.

Colloca, L., Benedetti, F., 2005. Placebos and painkillers: is mind as real asmatter? Nat. Rev. Neurosci. 6, 545–552.

Colloca, L., Lopiano, L., Lanotte, M., Benedetti, F., 2004. Overt versuscovert treatment for pain, anxiety, and Parkinson’s disease. LancetNeurol. 3, 679–684.

Coryell, W., Noyes, R., 1988. Placebo response in panic disorder. Am. J.Psychiatry. 145, 1138–1140.

Cubero, I., Thiele, T.E., Bernstein, I.L., 1999. Insular cortex lesions andtaste aversion learning: eVects of conditioning method and timing oflesion. Brain Res. 839, 323–330.

Dadds, M.R., Bovbjerg, D.H., Redd, W.H., Cutmore, T.R., 1997. Imageryin human classical conditioning. Psychol. Bull. 122, 89–103.

Dantzer, R., 2004. Cytokine-induced sickness behaviour: a neuroim-mune response to activation of innate immunity. Eur. J. Pharmacol.500, 399–411.

de Craen, A.J., Kaptchuk, T.J., Tijssen, J.G., Kleijnen, J., 1999a. Placebosand placebo eVects in medicine: historical overview. J. R Soc. Med.92, 511–515.

de Craen, A.J., Moerman, D.E., Heisterkamp, S.H., Tytgat, G.N., Tijssen,J.G., Kleijnen, J., 1999b. Placebo eVect in the treatment of duodenalulcer. Br. J. Clin. Pharmacol. 48, 853–860.

de la Fuente-Fernández, R., 2004. Uncovering the hidden placebo eVect indeep-brain stimulation for Parkinson’s disease. Parkin. Relat. Disord.10, 125–127.

de la Fuente-Fernández, R., Ruth, T.J., Sossi, V., Schulzer, M., Calne, D.B.,Stoessl, A.J., 2001. Expectation and dopamine release: mechanism ofthe placebo eVect in Parkinson’s disease. Science 293, 1164–1166.

de la Fuente-Fernández, R., Schulzer, M., Stoessl, A.J., 2004. Placebomechanisms and reward circuitry: clues from Parkinson’s disease. Biol.Psychiatry 56, 67–71.

de la Fuente-Fernández, R., Stoessl, A.J., 2002. The placebo eVect in Par-kinson’s disease. Trends Neurosci. 25, 302–306.

Dekker, E., Pelser, H., Groen, J., 1957. Conditioning as a cause of asth-matic attacks; a laboratory study. J. Psychosom. Res. 2, 97–108.

Del Rey, A., Randolf, A., Pitossi, F., Rogausch, H., Besedovsky, H.O.,2000. Not all peripheral immune stimuli that activate the HPA axisinduce proinXammatory cytokine gene expression in the hypothala-mus. Ann. N. Y. Acad. Sci. 917, 169–174.

DiLorenzo, T.A., Jacobsen, P.B., Bovbjerg, D.H., Chang, H., Hudis, C.A.,Sklarin, N.T., Norton, L., 1995. Sources of anticipatory emotional dis-tress in women receiving chemotherapy for breast cancer. Ann. Oncol.6, 705–711.

Dimond, E.G., Kittle, C.F., Crockett, J.E., 1960. Comparison of internalmammary artery ligation and sham operation for angina pectoris. Am.J. Cardiol. 5, 483–486.

Dolin, A., Krylov, V., Luk’ianenko, V., Flerov, B., 1960. Recent experi-mental data on conditioned reXex production and inhibition ofimmune and allergic reactions. Zh. Vyssh Nerv. Deiat. Im. I P Pav-lova 10, 832–841.

Dolin, A.O., Krylov, V.N., 1952. Role of the cerebral cortex in immuno-logic reactions of the organism. Zh. Vyssh Nerv Deiat Im. I P Pavlova2, 547–560.

Dyck, D.G., Janz, L., Osachuk, T.A., Falk, J., Labinsky, J., Greenberg,A.H., 1990. The Pavlovian conditioning of IL-1-induced glucocorticoidsecretion. Brain Behav. Immun. 4, 93–104.

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 443

Elenkov, I.J., Wilder, R.L., Chrousos, G.P., Vizi, E.S., 2000. The sympa-thetic nerve–an integrative interface between two supersystems: thebrain and the immune system. Pharmacol. Rev. 52, 595–638.

Ernst, E., Resch, K.L., 1995. Concept of true and perceived placebo eVects.BMJ 311, 551–553.

Evans, D., 2005. Suppression of the acute-phase response as a biologicalmechanism for the placebo eVect. Med. Hypotheses 64, 1–7.

Evans, F., 1974. The placebo response in pain reduction. In: Bonica, J.(Ed.), Advances in Neurology. Raven Press, New York, pp. 289–296.

Evans, F., 1985. Expectancy, therapeutic instructions, and the placeboresponse. In: White, L.T.B., Schwartz, G.E. (Eds.), Placebo: Theory,Research, and Mechanisms. The Guilford Press, New York, pp. 215–228.

Finneson, B., 1969. Diagnosis and management of pain syndromes ed..Saunders, Philadephia.

Fiore, M., Angelucci, F., Alleva, E., Branchi, I., Probert, L., Aloe, L.,2000. Learning performances, brain NGF distribution and NPY lev-els in transgenic mice expressing TNF-alpha. Behav Brain. Res. 112,165–175.

Fiorillo, C.D., Tobler, P.N., Schultz, W., 2003. Discrete coding of rewardprobability and uncertainty by dopamine neurons. Science 299,1898–1902.

Flaten, M.A., Blumenthal, T.D., 1999. CaVeine-associated stimuli elicitconditioned responses: an experimental model of the placebo eVect.Psychopharmacology (Berl.) 145, 105–112.

Flaten, M.A., Simonsen, T., Olsen, H., 1999. Drug-related information gen-erates placebo and nocebo responses that modify the drug response.Psychosom. Med. 61, 250–255.

Flaten, M.A., Simonsen, T., Zahlsen, K., Aamo, T., Sager, G., Olsen, H.,2004. Stimulant and relaxant drugs combined with stimulant and relax-ant information: a study of active placebo. Psychopharmacology(Berl.) 176, 426–434.

Fricchione, G., Stefano, G.B., 2005. Placebo neural systems: nitric oxide,morphine and the dopamine brain reward and motivation circuitries.Med. Sci. Monit. 11, MS54–MS65.

Gallagher, E.J., Viscoli, C.M., Horwitz, R.I., 1993. The relationship oftreatment adherence to the risk of death after myocardial infarction inwomen. Jama. 270, 742–744.

Garb, J.L., Stunkard, A.J., 1974. Taste aversions in man. Am. J. Psychiatry131, 1204–1207.

Garcia, J., Kimeldorf, D.J., Koelling, R.A., 1955. Conditioned aversion tosaccharin resulting from exposure to gamma radiation. Science 122,157–158.

Gauci, M., Husband, A.J., Saxarra, H., King, M.G., 1994. Pavlovian condi-tioning of nasal tryptase release in human subjects with allergic rhinitis.Physiology and Behavior 55, 823–825.

Geers, A.L., Helfer, S.G., Kosbab, K., Weiland, P.E., Landry, S.J., 2005a.Reconsidering the role of personality in placebo eVects: dispositionaloptimism, situational expectations, and the placebo response. J. Psy-chosom. Res. 58, 121–127.

Geers, A.L., Weiland, P.E., Kosbab, K., Landry, S.J., Helfer, S.G., 2005b.Goal activation, expectations, and the placebo eVect. J. Pers. Soc. Psy-chol. 89, 143–159.

Giang, D.W., Goodman, A.D., SchiVer, R.B., Mattson, D.H., Petrie, M.,Cohen, N., Ader, R., 1996. Conditioning of cyclophosphamide-inducedleukopenia in humans. J. Neuropsychiatry Clin. Neurosci. 8, 194–201.

Gibertini, M., 1996. IL1 beta impairs relational but not procedural rodentlearning in a water maze task. Adv. Exp. Med. Biol. 402, 207–217.

Glanzman, D., 1995. The cellular basis of classical conditioning in Aplysiacalifornica–it’s less simple than you think. Trends Neurosci. 18, 30–36.

Glaser, R., Kiecolt-Glaser, J.K., 2005. Stress-induced immune dysfunction:implications for health. Nat. Rev. Immunol. 5, 243–251.

Gleave, M.E., Elhilali, M., Fradet, Y., Davis, I., Venner, P., Saad, F., Klotz,L.H., Moore, M.J., Paton, V., Bajamonde, A., 1998. Interferon gamma-1b compared with placebo in metastatic renal-cell carcinoma.Canadian Urologic Oncology Group. N. Engl. J. Med. 338, 1265–1271.

Goebel, M.U., Hübell, D., Kou, W., Janssen, O.E., Katsarava, Z., Limm-roth, V., Schedlowski, M., 2005. Behavioral conditioning with inter-feron beta-1a in humans. Physiol. Behav. 84, 807–814.

Goebel, M.U., Trebst, A.E., Steiner, J., Xie, Y.F., Exton, M.S., Frede, S.,Canbay, A.E., Michel, M.C., Heemann, U., Schedlowski, M., 2002.Behavioral conditioning of immunosuppression is possible in humans.FASEB J. 16, 1869–1873.

Graz, B., Wietlisbach, V., Porchet, F., Vader, J.P., 2005. Prognosis or“curabo eVect?”: physician prediction and patient outcome of surgeryfor low back pain and sciatica. Spine 30, 1448–1452. discussion 1453.

Gryll, S.L., Katahn, M., 1978. Situational factors contributing to theplacebos eVect. Psychopharmacology (Berl) 57, 253–261.

Guess, H.K.A., Kusek, J.W., Engel, L.W., 2002. The science of the placeboed.. BMJ Books, London.

Haas, H.S., Schauenstein, K., 1997. Neuroimmunomodulation via limbicstructures–the neuroanatomy of psychoimmunology. Prog. Neurobiol.51, 195–222.

Hahn, R.A., 1997. The nocebo phenomenon: concept, evidence, and impli-cations for public health. Prev. Med. 26, 607–611.

Haour, F., 2005. Mechanisms of the placebo eVect and of conditioning.Neuroimmunomodulation 12, 195–200.

Hashish, I., Hai, H.K., Harvey, W., Feinmann, C., Harris, M., 1988. Reduc-tion of postoperative pain and swelling by ultrasound treatment: a pla-cebo eVect. Pain 33, 303–311.

Hashish, I., Harvey, W., Harris, M., 1986. Anti-inXammatory eVects ofultrasound therapy: evidence for a major placebo eVect. Br. J Rheuma-tol. 25, 77–81.

Herrnstein, R.J., 1962. Placebo eVect in the rat. Science 138, 677–678.Hiramoto, R., Ghanta, V., Solvason, B., Lorden, J., Hsueh, C.M., Rogers, C.,

Demissie, S., Hiramoto, N., 1993. IdentiWcation of speciWc pathways of com-munication between the CNS and NK cell system. Life Sci. 53, 527–540.

Hiramoto, R., Rogers, C., Demissie, S., Hsueh, C., Hiramoto, N., Lorden,J., Ghanta, V., 1997. Psychoneuroendocrine immunology: site of recog-nition, learning and memory in the immune system and the brain. Int.J. Neurosci. 92, 259–285.

Ho, K.H., Hashish, I., Salmon, P., Freeman, R., Harvey, W., 1988. Reduc-tion of post-operative swelling by a placebo eVect. J. Psychosom. Res.32, 197–205.

Horwitz, R.I., Viscoli, C.M., Berkman, L., Donaldson, R.M., Horwitz,S.M., Murray, C.J., RansohoV, D.F., Sindelar, J., 1990. Treatmentadherence and risk of death after a myocardial infarction. Lancet 336,542–545.

Hróbjartsson, A., Gøtzsche, P.C., 2001. Is the placebo powerless? An anal-ysis of clinical trials comparing placebo with no treatment. N. Engl. J.Med. 344, 1594–1602.

Hsueh, C., Chen, S., Lin, R., Chao, H., 2002. Cholinergic and serotonergicactivities are required in triggering conditioned NK cell response. J.Neuroimmunol. 123, 102–111.

Hsueh, C., Kuo, J., Chen, S., Huang, H., Cheng, F., Chung, L., Lin, R.,1999. Involvement of catecholamines in recall of the conditioned NKcell response. J. Neuroimmunol. 94, 172–181.

Hucklebridge, F., 2002. Behavioral conditioning of the immune system.Int. Rev. Neurobiol. 52, 325–351.

Ikemi, Y., Nakagawa, S., 1962. A Psychosomatic Study of ContagiousDermatitis. Kyushu. J. Med. Sci. 13, 335–350.

Ilnyckyj, A., Shanahan, F., Anton, P.A., Cheang, M., Bernstein, C.N., 1997.QuantiWcation of the placebo response in ulcerative colitis. Gastroen-terology 112, 1854–1858.

Jacobsen, P.B., Bovbjerg, D.H., Redd, W.H., 1993. Anticipatory anxiety inwomen receiving chemotherapy for breast cancer. Health Psychol. 12,469–475.

Janz, L.J., Brown, R., Zuo, L., Falk, J., Greenberg, A.H., Dyck, D.G., 1991.Conditioned taste aversion but not adrenal activity develops to ICVadministration of interleukin-1 in rats. Physiol. Behav. 49, 691–694.

Janz, L.J., Green-Johnson, J., Murray, L., Vriend, C.Y., Nance, D.M.,Greenberg, A.H., Dyck, D.G., 1996. Pavlovian conditioning of LPS-induced responses: eVects on corticosterone, splenic NE, and IL-2 pro-duction. Physiol. Behav. 59, 1103–1109.

Jensen, M.P., Karoly, P., 1991. Motivation and expectancy factors insymptom perception: a laboratory study of the placebo eVect.Psychosom Med. 53, 144–152.

444 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

Joyce, D.P., Jackevicius, C., Chapman, K.R., McIvor, R.A., Kesten, S.,2000. The placebo eVect in asthma drug therapy trials: a meta-analysis.J. Asthma. 37, 303–318.

Kandel, E., Abrams, T., Bernier, L., Carew, T., Hawkins, R., Schwartz, J.,1983. Classical conditioning and sensitization share aspects of the samemolecular cascade in Aplysia. Cold Spring Harb. Symp. Quant. Biol. 48(Pt. 2), 821–830.

Kaptchuk, T.J., 1998. Powerful placebo: the dark side of the randomisedcontrolled trial. Lancet 351, 1722–1725.

Kaptchuk, T.J., Stason, W.B., Davis, R.B., Legedza, A.R., Schnyer, R.N.,Kerr, C.E., Stone, D.A., Nam, B.H., Kirsch, I., Goldman, R.H., 2006.Sham device v inert pill: randomised controlled trial of two placebotreatments. BMJ 332, 391–397.

Katafuchi, T., Ichijo, T., Take, S., Hori, T., 1993. Hypothalamicmodulation of splenic natural killer cell activity in rats. J. Physiol. 471,209–221.

Katafuchi, T., Okada, E., Take, S., Hori, T., 1994. The biphasic changes insplenic natural killer cell activity following ventromedial hypothalamiclesions in rats. Brain Res. 652, 164–168.

Kawai, R., Sunada, H., Horikoshi, T., Sakakibara, M., 2004. Conditionedtaste aversion with sucrose and tactile stimuli in the pond snail Lym-naea stagnalis. Neurobiol. Learn Mem. 82, 164–168.

Kirsch, I., 1997. Specifying NonspeciWcs: Psychological Mechanisms ofPlacebo EVects. In: Harrington, A. (Ed.), The Placebo EVect: An Inter-disciplinary Exploration. Harvard University Press, Cambridge MA,pp. 166–186.

Kirsch, I.S.G., 1998. Listening to Prozac but hearing placebo: A meta-analysis of antidepressant medication. Prevention & Treatment 1.

Kirschbaum, C., Jabaaij, L., Buske-Kirschbaum, A., Hennig, J., Blom, M.,Dorst, K., Bauch, J., DiPauli, R., Schmitz, G., Ballieux, R., 1992.Conditioning of drug-induced immunomodulation in human volunteers:a European collaborative study. Br. J. Clin. Psychol. 31 (Pt 4), 459–472.

Kuo, J., Chen, S., Huang, H., Yang, C., Tsai, P., Hsueh, C., 2001. Theinvolvement of glutamate in recall of the conditioned NK cell response.J. Neuroimmunol. 118, 245–255.

La Mantia, L., Eoli, M., Salmaggi, A., Milanese, C., 1996. Does a placebo-eVect exist in clinical trials on multiple sclerosis? Review of the litera-ture. Ital J. Neurol. Sci. 17, 135–139.

Lanotte, M., Lopiano, L., Torre, E., Bergamasco, B., Colloca, L., Benedetti,F., 2005. Expectation enhances autonomic responses to stimulation ofthe human subthalamic limbic region. Brain Behav. Immun. 19, 500–509.

Lasagna, L., Mosteller, F., Von Felsinger, J.M., Beecher, H.K., 1954. Astudy of the placebo response. Am. J. Med. 16, 770–779.

Leigh, R., MacQueen, G., Tougas, G., Hargreave, F.E., Bienenstock, J.,2003. Change in forced expiratory volume in 1 second after sham bron-choconstrictor in suggestible but not suggestion-resistant asthmaticsubjects: a pilot study. Psychosom. Med. 65, 791–795.

Lekander, M., Fürst, C.J., Rotstein, S., Blomgren, H., Fredrikson, M., 1995.Anticipatory immune changes in women treated with chemotherapyfor ovarian cancer. Int. J. Behav. Med. 2, 1–12.

Leuchter, A.F., Cook, I.A., Witte, E.A., Morgan, M., Abrams, M., 2002.Changes in brain function of depressed subjects during treatment withplacebo. Am. J. Psychiatry 159, 122–129.

Levine, J.D., Gordon, N.C., 1984. InXuence of the method of drug adminis-tration on analgesic response. Nature 312, 755–756.

Longo, D.L., DuVey, P.L., Kopp, W.C., Heyes, M.P., Alvord, W.G., Sharf-man, W.H., Schmidt, P.J., Rubinow, D.R., Rosenstein, D.L., 1999. Con-ditioned immune response to interferon-[gamma] in humans. Clin.Immunol. 90, 173–181.

Lorenz, J., Hauck, M., Paur, R.C., Nakamura, Y., Zimmermann, R.,Bromm, B., Engel, A.K., 2005. Cortical correlates of false expectationsduring pain intensity judgments–a possible manifestation of placebo/nocebo cognitions. Brain Behav. Immun. 19, 283–295.

Luk’ianenko, V., 1961. The problem of conditioned reXex regulation ofimmunobiologic reactions. USP Sovrem. Biol. 51, 170–187.

Luparello, T., Lyons, H.A., Bleecker, E.R., McFadden, E.R., 1968. InXu-ences of suggestion on airway reactivity in asthmatic subjects. Psycho-som. Med. 30, 819–825.

Luparello, T.J., Leist, N., Lourie, C.H., Sweet, P., 1970. The interaction ofpsychologic stimuli and pharmacologic agents on airway reactivity inasthmatic subjects. Psychosom. Med. 32, 509–513.

Lynch, M.A., 2002. Interleukin-1 beta exerts a myriad of eVects in thebrain and in particular in the hippocampus: analysis of some of theseactions. Vitam. Horm. 64, 185–219.

MacKenzie, J., 1886. The production of the so-called rose eVect by meansof an artiWcial rose, with remarks and historical notes. Am. J. Med. Sci.91, 45–57.

Marella, S., Fischler, W., Kong, P., Asgarian, S., Rueckert, E., Scott, K.,2006. Imaging taste responses in the Xy brain reveals a functional mapof taste category and behavior. Neuron 49, 285–295.

Markovic, B., Dimitrijevic, M., Jankovic, B., 1993. Immunomodulation byconditioning: recent developments. Int. J. Neurosci. 71, 231–249.

Matsumoto, Y., Watanabe, S., Suh, Y.H., Yamamoto, T., 2002. EVects ofintrahippocampal CT105, a carboxyl terminal fragment of beta-amy-loid precursor protein, alone/with inXammatory cytokines on workingmemory in rats. J. Neurochem. 82, 234–239.

Matsumoto, Y., Yoshida, M., Watanabe, S., Yamamoto, T., 2001. Involve-ment of cholinergic and glutamatergic functions in working memoryimpairment induced by interleukin-1beta in rats. Eur. J. Pharmacol.430, 283–288.

Matteson, S., Roscoe, J., Hickok, J., Morrow, G.R., 2002. The role ofbehavioral conditioning in the development of nausea. Am. J. Obstet.Gynecol. 186, S239–S243.

McMillan, F.D., 1999. The placebo eVect in animals. J. Am. Vet Med.Assoc. 215, 992–999.

McNair, D.M., Gardos, G., Haskell, D.S., Fisher, S., 1979. Placeboresponse, placebo eVect, and two attributes. Psychopharmacology(Berl) 63, 245–250.

McRae, C., Cherin, E., Yamazaki, T.G., Diem, G., Vo, A.H., Russell, D.,Ellgring, J.H., Fahn, S., Greene, P., Dillon, S., WinWeld, H., Bjugstad,K.B., Freed, C.R., 2004. EVects of perceived treatment on quality of lifeand medical outcomes in a double-blind placebo surgery trial. Arch.Gen. Psychiatry 61, 412–420.

Melzack PD, R.W., 1988. The challenge of pain. Harmondsworth,London.

Menzel, R., Muller, U., 1996. Learning and memory in honeybees: frombehavior to neural substrates. Annu. Rev. Neurosci. 19, 379–404.

Metalnikov, S., Chorine, V., 1926. Role des réXexes conditionnels dansl’immunité. Annales de L’Institut Pasteur 11, 893–900.

Miller, F.G., Wendler, D., Swartzman, L.C., 2005. Deception in research onthe placebo eVect. PLoS Med. 2, e262.

Moerman, D.E., 2000. Cultural variations in the placebo eVect: ulcers, anx-iety, and blood pressure. Med. Anthropol. Q 14, 51–72.

Moertel, C.G., Taylor, W.F., Roth, A., Tyce, F.A., 1976. Who responds tosugar pills? Mayo. Clin. Proc. 51, 96–100.

Montgomery, G.H., Kirsch, I., 1997. Classical conditioning and the pla-cebo eVect. Pain 72, 107–113.

Morrow, G.R., Lindke, J., Black, P.M., 1991. Anticipatory nausea develop-ment in cancer patients: replication and extension of a learning model.Br. J. Psychol. 82 (Pt 1), 61–72.

Moseley, J.B., O’Malley, K., Petersen, N.J., Menke, T.J., Brody, B.A., Kuy-kendall, D.H., Hollingsworth, J.C., Ashton, C.M., Wray, N.P., 2002. Acontrolled trial of arthroscopic surgery for osteoarthritis of the knee.N. Engl. J. Med. 347, 81–88.

Moseley, J.B., Wray, N.P., Kuykendall, D., Willis, K., Landon, G., 1996.Arthroscopic treatment of osteoarthritis of the knee: a prospective,randomized, placebo-controlled trial. Results of a pilot study. Am. J.Sports. Med. 24, 28–34.

Moshel, Y.A., Durkin, H.G., Amassian, V.E., 2005. Lateralized neocorticalcontrol of T lymphocyte export from the thymus I. Increased exportafter left cortical stimulation in behaviorally active rats, mediated bysympathetic pathways in the upper spinal cord. J. Neuroimmunol. 158,3–13.

Niemi, M.B., Pacheco-López, G., Kou, W., Härting, M., del Rey, A., Bes-edovky, H.O., Schedlowski, M., in press. Murine taste-immune associa-tive learning. Brain Behav. Immun.

G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446 445

Oh, V.M., 1991. Magic or medicine? Clinical pharmacological basis of pla-cebo medication. Ann. Acad. Med. Singapore 20, 31–37.

Okamoto, S., Ibaraki, K., Hayashi, S., Saito, M., 1996. Ventromedial hypo-thalamus suppresses splenic lymphocyte activity through sympatheticinnervation. Brain Res. 739, 308–313.

Oppenheimer, S.M., Saleh, T., Cechetto, D.F., 1992. Lateral hypothalamicarea neurotransmission and neuromodulation of the speciWc cardiaceVects of insular cortex stimulation. Brain Res. 581, 133–142.

Pacheco-López, G., Niemi, M.B., Kou, W., Härting, M., Del Rey, A., Bese-dovsky, H.O., Schedlowski, M., 2004. Behavioural endocrine immune-conditioned response is induced by taste and superantigen pairing.Neuroscience 129, 555–562.

Pacheco-López, G., Niemi, M.B., Kou, W., Härting, M., Fandrey, J.,Schedlowski, M., 2005. Neural substrates for behaviorally conditionedimmunosuppression in the rat. J. Neurosci. 25, 2330–2337.

Pacheco-López, G.N., Engler, Niemi, M.B., Schedlowski, M., 2006.Behaviorally conditioned enhancement of immune responses. In: Ader, R.(Ed.), Psychoneuroimmunology, vol. 1. Academic Press, London, p. 591.

Papakostas, Y.G., Daras, M.D., 2001. Placebos, placebo eVect, and theresponse to the healing situation: the evolution of a concept. Epilepsia42, 1614–1625.

Paradis, S., Cabanac, M., 2004. Flavor aversion learning induced by lith-ium chloride in reptiles but not in amphibians. Behav. Processes 67,11–18.

Patel, S.M., Stason, W.B., Legedza, A., Ock, S.M., Kaptchuk, T.J., Conboy,L., Canenguez, K., Park, J.K., Kelly, E., Jacobson, E., Kerr, C.E.,Lembo, A.J., 2005. The placebo eVect in irritable bowel syndrome tri-als: a meta-analysis. Neurogastroenterol. Motil. 17, 332–340.

Pavlov, I., 1927. Conditioned ReXexes: An Investigation of the Physiologi-cal Activity of the Cerebral Cortex ed.. Oxford University Press, Cam-bridge.

Pavlov, V., Tracey, K., 2005. The cholinergic anti-inXammatory pathway.Brain Behav. Immun., 493–499.

Pihl, R.O., Altman, J., 1971. An experimental analysis of the placebo eVect.J. Clin. Pharmacol. New Drugs 11, 91–95.

Pitossi, F., del Rey, A., Kabiersch, A., Besedovsky, H., 1997. Induction ofcytokine transcripts in the central nervous system and pituitary follow-ing peripheral administration of endotoxin to mice. J. Neurosci. Res.48, 287–298.

Pizzo, P.A., Robichaud, K.J., Edwards, B.K., Schumaker, C., Kramer, B.S.,Johnson, A., 1983. Oral antibiotic prophylaxis in patients with cancer:a double-blind randomized placebo-controlled trial. J. Pediatr. 102,125–133.

Ploghaus, A., Becerra, L., Borras, C., Borsook, D., 2003. Neural circuitryunderlying pain modulation: expectation, hypnosis, placebo. TrendsCogn. Sci. 7, 197–200.

Plato; Charmides, 155-6: I replied that it was a kind of leaf, which requiredto be accompanied by a charm, and if a person would repeat the charmat the same time that he used the cure, he would be made whole; butthat without the charm the leaf would be of no avail.

Price, D.D., Milling, L.S., Kirsch, I., DuV, A., Montgomery, G.H., Nich-olls, S.S., 1999. An analysis of factors that contribute to the magni-tude of placebo analgesia in an experimental paradigm. Pain 83,147–156.

Rachal Pugh, C., Fleshner, M., Watkins, L.R., Maier, S.F., Rudy, J.W.,2001. The immune system and memory consolidation: a role for thecytokine IL-1beta. Neurosci. Biobehav. Rev. 25, 29–41.

Ramírez-Amaya, V., Alvarez-Borda, B., Bermúdez-Rattoni, F., 1998.DiVerential eVects of NMDA-induced lesions into the insular cortexand amygdala on the acquisition and evocation of conditioned immu-nosuppression. Brain Behav. Immun. 12, 149–160.

Ramírez-Amaya, V., Alvarez-Borda, B., Ormsby, C., Martínez, R., Pérez-Montfort, R., Bermúdez-Rattoni, F., 1996. Insular cortex lesionsimpair the acquisition of conditioned immunosuppression. BrainBehav. Immun. 10, 103–114.

Ramírez-Amaya, V., Bermúdez-Rattoni, F., 1999. Conditioned enhance-ment of antibody production is disrupted by insular cortex and amyg-dala but not hippocampal lesions. Brain Behav. Immun. 13, 46–60.

Redd, W.H., Dadds, M.R., Futterman, A.D., Taylor, K.L., Bovbjerg, D.H.,1993. Nausea induced by mental images of chemotherapy. Cancer 72,629–636.

Rees, J.R., Wade, T.J., Levy, D.A., Colford Jr., J.M., Hilton, J.F., 2005.Changes in beliefs identify unblinding in randomized controlled trials:a method to meet CONSORT guidelines. Contemp. Clin. Trials 26,25–37.

Reilly, S., Bornovalova, M.A., 2005. Conditioned taste aversion and amyg-dala lesions in the rat: a critical review. Neurosci. Biobehav. Rev. 29,1067–1088.

Rhudy, J.L., Meagher, M.W., 2000. Fear and anxiety: divergent eVects onhuman pain thresholds. Pain 84, 65–75.

Ross, S.S.S.B., 1963. Further support for the placebo eVect in the rat. Psy-chol. Reports 13, 461–462.

Sabbioni, M.E., Bovbjerg, D.H., Mathew, S., Sikes, C., Lasley, B., Stokes,P.E., 1997. Classically conditioned changes in plasma cortisol levelsinduced by dexamethasone in healthy men. FASEB J. 11, 1291–1296.

Saito, M., Minokoshi, Y., Shimazu, T., 1989. Accelerated norepinephrineturnover in peripheral tissues after ventromedial hypothalamicstimulation in rats. Brain Res. 481, 298–303.

Sakic, B., Vlajkovic, S., 1990. Self-stimulation behavior: consequencesupon immunity? Brain Behav. Immun. 4, 255–264.

SandroV, R., 1980. The potent placebo. Rn 43, 34–37. also 88–96.Sarinopoulos, I., Dixon, G.E., Short, S.J., Davidson, R.J., Nitschke, J.B.,

2006. Brain mechanisms of expectation associated with insula andamygdala response to aversive taste: implications for placebo. BrainBehav. Immun. 20, 120–132.

Schapira, K., McClelland, H.A., GriYths, N.R., Newell, D.J., 1970. Studyon the eVects of tablet colour in the treatment of anxiety states. Br.Med. J. 1, 446–449.

Schedlowski, M., 2006. Insecta immune-cognitive interactions. BrainBehav. Immun. 20, 133–134.

Schedlowski, M., Hosch, W., Oberbeck, R., Benschop, R.J., Jacobs, R.,Raab, H.R., Schmidt, R.E., 1996. Catecholamines modulate humanNK cell circulation and function via spleen-independent beta 2-adren-ergic mechanisms. J. Immunol. 156, 93–99.

Schneider, H., Pitossi, F., Balschun, D., Wagner, A., del Rey, A., Besedov-sky, H.O., 1998. A neuromodulatory role of interleukin-1beta in thehippocampus. Proc. Natl. Acad. Sci. USA 95, 7778–7783.

Schultz, W., 1998. Predictive reward signal of dopamine neurons. J. Neuro-physiol. 80, 1–27.

Sewards, T.V., 2004. Dual separate pathways for sensory and hedonicaspects of taste. Brain Res. Bull. 62, 271–283.

Sewards, T.V., Sewards, M., 2002. Separate, parallel sensory and hedonicpathways in the mammalian somatosensory system. Brain Res. Bull.58, 243–260.

Sewards, T.V., Sewards, M.A., 2001. Cortical association areas in the gus-tatory system. Neurosci. Biobehav. Rev. 25, 395–407.

Shapiro, A., 1960. A contribution to a history of the placebo eVect. Behav.Sci. 5, 109–135.

Shapiro, A., 1971. Placebo eVects in medicine, psychotherapy, and psycho-analysis. In: Bergin, A.G.S.L. (Ed.), Handbook of Psychotherapy andBehavior Change: Empirical Analysis. Wiley, New York, pp. 439–473.

Simon, R., 2000. Are placebo-controlled clinical trials ethical or neededwhen alternative treatment exists? Ann. Intern. Med. 133, 474–475.

Skinner, B., 1953. Science and Human Behavior ed.. MacMillan, London.Smith, G.R., McDaniel, S.M., 1983. Psychologically mediated eVect on the

delayed hypersensitivity reaction to tuberculin in humans. Psychosom.Med. 45, 65–70.

Solomon, S., 2002. A review of mechanisms of response to pain therapy:why voodoo works. Headache 42, 656–662.

Spiegel, H., 1997. Nocebo: the power of suggestibility. Prev. Med. 26,616–621.

Sredni-Kenigsbuch, D., 2002. TH1/TH2 cytokines in the central nervoussystem. Int. J. Neurosci. 112, 665–703.

Stockhorst, U., Spennes-Saleh, S., Körholz, D., Göbel, U., Schneider, M.E.,Steingrüber, H.J., Klosterhalfen, S., 2000. Anticipatory symptoms andanticipatory immune responses in pediatric cancer patients receiving

446 G. Pacheco-López et al. / Brain, Behavior, and Immunity 20 (2006) 430–446

chemotherapy: features of a classically conditioned response? BrainBehav. Immun. 14, 198–218.

Straub, R.H., 2004. Complexity of the bi-directional neuroimmunejunction in the spleen. Trends Pharmacol. Sci. 25, 640–646.

Su, C., Lichtenstein, G.R., Krok, K., Brensinger, C.M., Lewis, J.D., 2004. Ameta-analysis of the placebo rates of remission and response in clinicaltrials of active Crohn’s disease. Gastroenterology 126, 1257–1269.

Swanson, L.W., Petrovich, G.D., 1998. What is the amygdala? TrendsNeurosci. 21, 323–331.

Szelényi, J., 2001. Cytokines and the central nervous system. Brain Res.Bull. 54, 329–338.

Talbot, M. (2000). The Placebo Prescription. New York Times Magazine,New York, January 9th: www.nytimes.com/library/magazine/home/20000109mag-talbot7.html.

Tazi, A., Dantzer, R., Crestani, F., Le Moal, M., 1988. Interleukin-1induces conditioned taste aversion in rats: a possible explanation for itspituitary-adrenal stimulating activity. Brain Res. 473, 369–371.

Tenery, R., Rakatansky, H., Riddick, F.A., Goldrich, M.S., Morse, L.J.,O’Bannon, J.M., Ray, P., Smalley, S., Weiss, M., Kao, A., Morin, K.,Maixner, A., Seiden, S., 2002. Surgical “placebo” controls. Ann. Surg.235, 303–307.

Thomsen, J., Bretlau, P., Tos, M., Johnsen, N.J., 1981a. Ménière’s disease:endolymphatic sac decompression compared with sham (placebo)decompression. Ann. N. Y. Acad. Sci. 374, 820–830.

Thomsen, J., Bretlau, P., Tos, M., Johnsen, N.J., 1981b. Placebo eVect insurgery for Ménière’s disease. A double-blind, placebo-controlledstudy on endolymphatic sac shunt surgery. Arch. Otolaryngol. 107,271–277.

Thomsen, J., Bretlau, P., Tos, M., Johnsen, N.J., 1983. Ménière’s disease:a 3-year follow-up of patients in a double-blind placebo-controlledstudy on endolymphatic sac shunt surgery. Adv. Otorhinolaryngol.30, 350–354.

Todd, B., 1987. The placebo eVect. Real or imaginary? Geriatr. Nurs. 8,154–156.

Tonelli, L.H., Postolache, T.T., Sternberg, E.M., 2005. InXammatory genesand neural activity: involvement of immune genes in synaptic functionand behavior. Front. Biosci. 10, 675–680.

Tracey, K., 2002. The inXammatory reXex. Nature 420, 853–859.Traut, E.F., Passarelli, E.W., 1957. Placebos in the treatment of rheuma-

toid arthritis and other rheumatic conditions. Ann. Rheum. Dis. 16,18–22.

Tully, T., Boynton, S., Brandes, C., Dura, J., Mihalek, R., Preat, T., Villella,A., 1990. Genetic dissection of memory formation in Drosophilamelanogaster. Cold Spring Harb. Symp. Quant. Biol. 55, 203–211.

Tuohy, V.K., 2005. The neocortical–immune axis. J. Neuroimmunol. 158, 1–2.

Turnbull, A.V., Rivier, C.L., 1999. Regulation of the hypothalamic–pitui-tary–adrenal axis by cytokines: actions and mechanisms of action.Physiol. Rev. 79, 1–71.

Turnbull, J., 1962. Asthma conceived as a learned response. J. Psychosom.Res. 6, 59–70.

Uhlenhuth, E.H., Rickels, K., Fisher, S., Park, L.C., Lipman, R.S., Mock, J.,1966. Drug, doctor’s verbal attitude and clinic setting in the symptom-atic response to pharmacotherapy. Psychopharmacologia 9, 392–418.

Vettor, R., Fabris, R., Pagano, C., Federspil, G., 2002. Neuroendocrineregulation of eating behavior. J. Endocrinol. Invest. 25, 836–854.

Voudouris, N.J., Peck, C.L., Coleman, G., 1990. The role of conditioningand verbal expectancy in the placebo response. Pain 43, 121–128.

Wager, T.D., Matre, D., Casey, K.L., 2006. Placebo eVects in laser-evokedpain potentials. Brain Behav. Immun. 20, 219–230.

Wager, T.D., Nitschke, J.B., 2005. Placebo eVects in the brain: linkingmental and physiological processes. Brain Behav. Immun. 19, 281–282.

Wampold, B.E., Minami, T., Tierney, S.C., Baskin, T.W., Bhati, K.S., 2005.The placebo is powerful: estimating placebo eVects in medicine and psy-chotherapy from randomized clinical trials. J. Clin. Psychol. 61, 835–854.

Wenner, M., Kawamura, N., Ishikawa, T., Matsuda, Y., 2000. Rewardlinked to increased natural killer cell activity in rats. Neuroimmuno-modulation 7, 1–5.

Wickramasekera, I., 1985. A conditioned response model of the placeboeVect: predictions from the model. In: White, L.T.B., Schwartz, G.E.(Eds.), In Placebo: Theory, Research, and Mechanisms. The GuilfordPress, New York, pp. 255–287.

Wolf, S., 1950. EVects of suggestion and conditioning on the action ofchemical agents in human subjects; the pharmacology of placebos.J. Clin. Invest. 29, 100–109.

Wolf, S., 1959. The pharmacology of placebos. Pharmacol. Rev. 11, 689–704.Wrolstad, J., 2002. Depression patients show physical placebo eVect.

News Factor Network, January10th: www.placebo.ucla.edu/news/PDF/newsfactor020110.pdf.

Wrona, D., 2006. Neural-immune interactions: an integrative view of thebidirectional relationship between the brain and immune systems.J. Neuroimmunol. 172, 38–58.

Wrona, D., Trojniar, W., 2003. Chronic electrical stimulation of the lateralhypothalamus increases natural killer cell cytotoxicity in rats. J. Neuro-immunol. 141, 20–29.

Yamamoto, T., Shimura, T., Sako, N., Yasoshima, Y., Sakai, N., 1994.Neural substrates for conditioned taste aversion in the rat. Behav.Brain Res. 65, 123–137.

Zubieta, J.K., Yau, W.Y., Scott, D.J., Stohler, C.S., 2006. Belief or Need?Accounting for individual variations in the neurochemistry of theplacebo eVect. Brain Behav. Immun. 20, 15–26.