Content uploaded by Dirk Devroey
Author content
All content in this area was uploaded by Dirk Devroey
Content may be subject to copyright.
The Placebo Effect: How the Subconscious Fits in
J. L. Mommaerts
Dirk Devroey
Perspectives in Biology and Medicine, Volume 55, Number 1, Winter
2012, pp. 43-58 (Article)
Published by The Johns Hopkins University Press
DOI: 10.1353/pbm.2012.0005
For additional information about this article
Access Provided by your local institution at 02/19/13 9:55PM GMT
http://muse.jhu.edu/journals/pbm/summary/v055/55.1.mommaerts.html
J. L . Mo mm ae rt s an d Di rk D ev ro ey
The Placebo Effect
how the subconscious fits in
Department of General Practice, Faculty of Medicine, Free University of Brussels, Belgium.
Correspondence: Jean-Luc Mommaerts, M.D.,M.Sc., Oude Godstraat 200, 2650 Edegem, Belgium.
E-mail: jean.luc.mommaerts@pandora.be.
Perspectives in Biology and Medicine, volume 55, number 1 (winter 2012):43–58
© 2012 by The Johns Hopkins University Press
ABSTRACT The placebo effect is very well known, being replicated in many sci-
entific studies. At the same time, its exact mechanisms still remain unknown. Quite a
few hypothetical explanations for the placebo effect have been suggested, including
faith, belief, hope, classical conditioning, conscious/subconscious expectation, endor-
phins, and the meaning response.This article argues that all these explanations may boil
down to autosuggestion, in the sense of “communication with the subconscious.”An
important implication of this is that the placebo effect can in principle be used effec-
tively without the placebo itself, through a direct use of autosuggestion.The benefits of
such a strategy are clear: fewer side effects from medications, huge cost savings, no
deception of patients, relief of burden on the physician’s time, and healing in domains
where medication or other therapies are problematic.
Pl ac eb o an d th e Pl ac eb o Ef fe ct
A much-cited definition of placebo is from Shapiro and Shapiro (1997):“any ther-
apy (or that component of any therapy) that is intentionally or knowingly used
for its nonspecific, psychological, or psychophysiological, therapeutic effect, or
that is used for a presumed specific therapeutic effect on a patient, symptom, or
illness but is without specific activity for the condition being treated” (p. 41).
What nonspecific means and how it relates to the psyche has been written about
extensively yet inconclusively. In the end, the term nonspecific doesn’t say any-
thing about the crux of the matter.
Talking about placebo, one first has to distinguish between “placebo effect
43
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 43
proper” and “perceived placebo effect.”The latter is the effect of one or more of
the following: regression to the mean, natural history of disease, investigator
effects (such as the nocebo effect called “white coat hypertension”), and Haw-
thorne effect–like changes in patients’ behavior (patients becoming sensitized to
the problem under investigation, once they are included in a trial). In this arti-
cle, placebo effect denotes the “placebo effect proper,” and placebo anything that
provokes this.This said, the placebo effect may be seen as the primary and non-
discernible effect on a person brought about by using a placebo, which is in itself
inert regarding this very effect or part of it. It is what happens in time between
the contact with the placebo and the physiological/psychological consequences
of the placebo effect.There is no clear dividing line between the effect and the
consequences. However, using the term placebo effect puts emphasis on how an ex-
ternal event (such as the taking of a drug) gets translated into a discernible inter-
nal event (such as feeling better or the lowering of blood pressure). In this, the
external event is not the final cause but a precipitating factor inducing some-
thing within the person that itself leads to the consequences of the placebo
effect.
The word placebo usually makes one think of drugs. But placebos can also be
objects (syringes, a copper bracelet), rituals (physical examination, anamnesis),
places (doctor’s office), relationships (with doctors, self-help groups), thoughts
(about performances of medicine), and other entities such as touch,words, social
interventions, a tattoo, or a surgery scar. In fact, anything that has meaning to
someone can act as a placebo.Thus, with each encounter with the medical sys-
tem, a patient’s placebo response is being shaped (Peck and Coleman 1991).
Even the act of diagnosing a person with a medical condition can have a placebo
effect (Moerman 2002). From the other side, in view of the complexity of the
human body-mind system, probably almost any health problem can be influ-
enced through the placebo effect. Many authors agree that the average placebo
effect of all present-day medications together may be higher than half of the total
action, still compar ing very favorably to pre-1900 medications with placebo
effects of mostly the full range (Shapiro and Shapiro 1997).
Mechanisms of the Placebo Effect
Many mechanisms have been described to explain the placebo effect. In this arti-
cle, the five mechanisms most prominent in the medical literature are looked at
in-depth. However,there are also other hypothetical mechanisms, including: opti-
mism/re-moralization, conviction, relationship with the prescriber, magic, diver-
sion of attention, symbolic power, oneness of body and mind, transference, relief
of guilt, sense of control, diminishment of anxiety, fitting in the story, (self-)hyp-
nosis, emotional appeal, and response-appropriate sensation theory (Benson and
Epstein 1975; Brody 2000; Byerly 1976; Evans 1981, 1985; Fisher and Dlin 1956;
Perspectives in Biology and Medicine
J. L. Mommaerts and Dirk Devroey
44
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 44
45
winter 2012 • volume 55, number 1
The Placebo Effect
Frank 1973; Glaser and Kiecolt-Glaser 1994; Hahn 1985; Houston 1938; Johnson
1994; Markus 1983; Miller, Colloca, and Kaptchuk 2009; Moerman 2002; Peck
and Coleman 1991; Sarno 1998; Spiro 1998;Wall 1993).These mechanisms over-
lap to a huge degree, which in itself already points to a probably parsimonious ex-
planation. Indeed, they can all be described as autosuggestion, in the sense of “com-
munication to the subconscious.”When looked upon from a neurophysiological
and neurophilosophical perspective, not hope/faith as such, but true hope/faith
has an effect when reaching the subconscious and becomes autosuggestion
(Churchland 2002; Mommaerts 2000). One cannot purely consciously decide to
have true hope, faith, or expectation, just as one cannot consciously and effec-
tively wish oneself into getting better through a placebo pill or procedure. In this
hypothetical case,the placebo pill would even be senseless. Instead, placebo works
through the subconscious being appealed to—be it, in the case of placebo, in a
covert way. In general, however, the subconscious is painstakingly avoided in
medical literature, explicitly as well as implicitly.
Faith/Belief/Hope
Generally, people take medication because they think it makes them better. At
least, that is their hope. According to many authors, such as W. B. Plotkin (1985)
and H. Spiro (1998), hope or faith constitutes much of the basis for the placebo
response.The placebo itself is the symbol or carrier of hope, which overlaps very
much with faith and belief. Technology, medical authority, and even sympathy
are seen as eliciting the placebo response only indirectly, by virtue of the effect
they have on the patient’s beliefs. It is the belief in the treatment, therefore—
something that lies inside the patient’s mind—that sets off the placebo response
(Evans 2003). Faith/belief/hope has in fact frequently been related to regular
medicine in the past.According to SirWilliam Osler in 1932, it was seen as “the
great stock in trade of the profession . . . the foundation of therapeutics” (Shapiro
and Shapiro 1997, p. 58).
Faith is not to be seen in the purely conceptual sense that “I have faith that
the sun will shine tomorrow.”What is meant is a deep feeling of faith, such as is
the case when someone really believes in God. Only “deep faith” can bring
about healing. Its meaning, its message, has to be a personal conviction, an inner
certainty—what exactly you consciously believe is unimportant for this matter.
By contrast, having superficial “small faith” can even bring about a negative
investment, with an end result the opposite of what deep faith would have
brought. If research does not take this difference into account, then of course the
effect of faith may turn out to be nil.
Putting forward hope as a placebo mechanism may shed some light upon why
the placebo effect plays a bigger role in clinical pain than in experimental pain
(Beecher 1955). In the latter case, hope is not so readily relevant except when
the pain lasts long or is intense.This explains the observation that placebo anal-
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 45
gesic effects are larger for those forms of experimental pain that are more stress-
ful or of longer duration (Jospe 1978).Where hope to be relieved from pain is
stronger, the placebo effect is more active.
The side effects of pure placebos in double-blind studies are often similar to
those of the drugs to which the placebos are being compared (Ross and Olson
1981). It is not clear how these side effects in themselves can be engendered by
hope. Expectation may be a better model. Subjects who expect side effects get
what they expect. At the utmost, one could say that the subject “hopes” to have
the side effects, as these are seen as proof of receiving the active medication.
In conclusion, faith/hope/belief cures (or heals) when it is deep, true, touch-
ing one’s inner self. It achieves the placebo effect when it constitutes a kind of
“communication to the inner self”—that is, when it acts as one kind of auto-
suggestion.
Classical Conditioning
Classical conditioning is the pairing of an unconditioned stimulus (UCS) with
a conditioned stimulus (CS) until the CS elicits the same response as the UCS.
The term is mostly used with the implicit assumption that the subject is not
consciously aware of the CS-UCS association, or at least that the subject’s con-
scious awareness is not directly involved. The CS can be very concise (such as a
particular medication) or very broad (such as the healing environment, includ-
ing the smell when entering a hospital, the doctor’s and nurse’s attitude towards
the patient and others, ideas of “science” and “caring,” the erroneous idea of hav-
ing the right to be perfectly healthy or get the perfect treatment, and so on).
Probably the first to investigate a conditioned placebo effect from drugs was
Pavlov (1927). He reported that effects from morphine injections occurred in
conditioned animals already when he was preparing the injections or when he
placed the dogs in the experimental chamber where they had previously re-
ceived morphine. The dogs reacted in these cases upon their expectation of
being reinjected with morphine. A good example of human conditioning
towards placebo effect is seen in a study by Smith and McDaniel (1983). In this
study, placebo conditioning clearly reduced the immune response to tuberculin,
thereby showing also once again placebo’s substantial power.
The classical conditioning model traditionally bypasses any reference to the
subconscious being more complex than just a set of simple reflexes. Complex
meaning is, however, always important in situations of classical conditioning.
Indeed it is only the deeply meaningful perception of environment and condi-
tions that acts upon one. Even Pavlov’s dogs expected the food when the bell
tolled, in a way that indicated the bell and the food meant something to the
dogs.The same can be seen in human cases. For instance, in an experiment in
which asthmatics were given a placebo broncho-constrictive inhalation and a
placebo drug, pulmonary function reacted in both cases in a placebo-prone
Perspectives in Biology and Medicine
J. L. Mommaerts and Dirk Devroey
46
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 46
47
winter 2012 • volume 55, number 1
The Placebo Effect
manner.The most important predictive factor in this setting was the manner in
which individuals perceived the experimental setting as consequence of the sug-
gestions they received together with the placebo (Butler and Steptoe 1986).
Taking the argument even further, both unconditioned and conditioned stim-
uli in the classical conditioning paradigm are themselves in fact meanings. To
Pavlov’s dogs, food means “yummy yummy,” thus saliva flows.Then by association
the bell comes to mean “food will come,” thus “yummy yummy” and saliva
flows. Neither food nor bell make saliva flow by themselves: “yummy yummy”
does. This is therefore the real stimulus or UCS, while “bell–yummy yummy”
is the real CS. If the dog becomes brain-damaged in a specific way, the bell
doesn’t mean anything anymore, and there’s no flow of saliva.
A suggestive message can come from prior exposure and association. Re-
newed exposure then carries the suggestion of probably leading to the same
result. Looked upon in a mechanical way, this fits the conditioning paradigm.
However, it doesn’t work without the deeper layer, which is why many authors
such as Reiss (1980) put expectation before conditioning. Expectancies can also
be formed without direct personal experience or, for example, through observa-
tional learning, verbal information, persuasion, and other symbolic processes
(Peck and Coleman 1991). Price and Fields (1997) contend that although clas-
sical conditioning can change one’s expectation, other types of learning can also
contribute. For example, expectation can reflect knowledge about the therapeu-
tic agent, the circumstances under which it is administered, and the condition to
be treated. Expectation of relief may cause a placebo response without prior
exposure to a therapeutic agent, though such exposure certainly will increase
expectation.
Some authors appear to see conditioning as a kind of “subconscious expecta-
tion,” where the subconscious is seen as a black box. In a specific experiment in
which the subjects weren’t told the real aim of the experiment, Amanzio and
Benedetti (1999) noted that placebo responses occurred “without expectation”
of pain relief. If subjects were previously conditioned with either morphine or
ketorolac, the lack of expectation cues only reduced but did not prevent the
placebo effect. Thus, previously conditioned subjects experienced analgesic
effects even “when not expecting any.” However, the fact is that such subjects do
expect an analgesic effect, not consciously but subconsciously—and apparently,
that’s enough.The distinction generally made between conditioning and expec-
tation results from not taking the subconscious into account in the concept of
expectation. In short, conditioning seems to be only one of the mechanisms by
which the expectancies exert their effect (Peck and Coleman 1991).This means
explanations for the placebo effect may be offered at different process levels but
still prove to be fully compatible to each other.
On the other side, there is a continuum between conditioning and what is no
more than physiological reflexes. R. J. Herrnstein (1962) reported that the effects
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 47
of scopolamine injections in rats also appeared also after saline injections in
“conditioned rats.” Several authors see in this an example of animal condition-
ing. However, no internal mental processing is necessary in this case. Generally
put, if there is no “meaning,” the process should not properly be called “condi-
tioning.” In the case of the “conditioned” rats, what was observed may be a
purely physiological reflex and change, such as that due to volume distention of
the cardiovascular system, without any expectation involved. To call this a
placebo effect is confusing: no conclusions about classical conditioning should
be drawn from mere physiological changes.
Conscious/Subconscious Expectation
Many authors suggest that expectation is a salient determinant of the placebo
effect in general and placebo analgesia in particular (Laska and Sunshine 1973;
White,Tursky and Schwarts 1985). However, there is little or no explicit men-
tion of subconscious expectation.An exception is the single case stating that “ques-
tions about the possibility of unconscious expectancy remain unresolved” (Hoff-
man, Harrington, and Fields 2005, p. 257). Moreover, there is only rarely even an
implicit reference to the subconscious. Most authors use the word expectation as
if only conscious expectation is meant—this is, what one can consciously put
into words and communicate in a formal way. To assess conscious expectation,
one can simply ask the subjects of an experiment about their expectations before
the experiment begins, as where placebo reactors are (thought to be) identified
before the trial by asking the subjects what they expect as outcome of the ther-
apy (Wall 1993).
Most authors agree that expectation can be very powerful.It can even override
pharmacological effects, so that expectancies contrary to the pharmacological
effect of a drug can in some cases prevail over the drug’s effect itself (Kirsch 1985).
Moreover, expectations (and thus the placebogenic effect) are nowhere so high as
with surgery.This should be no surprise,since the meaningful aura around surgery
is quite impressive: surroundings, surgeon’s personality, anesthetic, incision, experi-
ences of friends and other patients, length of the illness, amount of pain, accounts
of surgery in the media (Johnson 1994).As is to be expected, expectations as well
as the placebo effect heighten through confidence in the obtainment of relief.This
is clear, for instance, in specific research about branding, in which branded place-
bos are more effective than unbranded placebos in relieving headaches; branded
active ingredients are more effective than unbranded active ingredients; and a
familiar brand is more effective than an unfamiliar brand. Moreover, the placebo
effect of the branded placebo can even be higher than the pharmacological effect
under study (Branthwaite and Cooper 1981). In an experiment by Amanzio et al.
(2001), analgesics given surreptitiously had fewer effects than the same analgesics
given openly. This shows the effect of expectation without the use of a placebo
arm in the study.The underlying mechanism is the same as in a placebo-controlled
trial. Here too,“no expectation” means “no placebo effect.”
48
J. L. Mommaerts and Dirk Devroey
Perspectives in Biology and Medicine
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 48
In former times, prior to rigorous studies, a new drug was often greeted with
much enthusiasm and therapeutic effectiveness in the vast majority of patients
(70–90%), due in large part to a strong placebo effect evoked by the expectations
of investigators and clinicians.After better controlled studies,the enthusiasm typ-
ically dipped to a low level, and so did the therapeutic effectiveness on the field.
This has been described as a recurrent pattern (Benson and McCallie 1979). A
well-known adage from the 19th-century French physician Armand Trousseau
notes that: “You should treat as many patients as possible with the new drugs
while they still have the power to heal.”
One’s expectation of relief can be modified without prior exposure/con-
ditioning to the treatment under question. It has therefore been recommended
to deliberately change expectations prior to performing the procedure, through
information, persuasion, or learning from others who have been helped by such
procedures (Peck and Coleman 1991). If you give someone the expectation of
cure, for example, by showing others who apparently get better, then in the same
kind of environment this can lead to a positive placebo effect, even without the
need of any positive phar macological action. This probably accounts for the
many of the successful outcomes of 19th-century and earlier medicine. (By con-
trast, practitioners who use weak non-placebos, or who primarily rely upon
placebos, will weaken the non-placebo component of their therapy.) In princi-
ple, one can rely purely on placebos and still strengthen the placebo effect of
one’s actions, even if the medication fails, because it is a well-documented char-
acteristic of human nature that failure leads more readily to “more of the same”
than to a reconsideration of the working hypothesis (Baron 1990). If the med-
ication fails, the general deduction in practice quite often is that one didn’t use
enough of it. This helps explain why, although 18th- and 19th-centur y blood-
letting almost always failed, it only drove most physicians to more bloodletting,
up to utter extremes and including the death of the first president of the United
States (Shapiro and Shapiro 1997).
The special nature of an intervention may increase expectation.Thus, the fla-
vor of exoticism that surrounds therapies such as acupuncture, the high-tech
magic of treatments like ultrasound or laser therapy, or the unusual nature of a
therapeutic encounter, such as history taking in homoeopathy, may all increase
the placebo effect (Ernst 2001). Expectations of clinicians themselves can also be
effective in heightening placebo effects even without (consciously) involving
patients. In an experiment by Gracely et al. (1985), the effect of a pure placebo
painkiller in the context of soothing the pain of the extraction of wisdom teeth
was substantially heightened simply by telling the clinicians (dentists) that it
might contain an active product versus telling them that it was only a placebo.
The patients’ expectations were heightened by the clinicians’expectations, with-
out consciously conveying anything. It all happened on a subconscious level.
Conceptually very close to expectation is anticipation. The latter seems just a
degree stronger or less doubtful.When studying postoperative patients respond-
The Placebo Effect
winter 2012 • volume 55, number 1 49
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 49
ing to analgesic drugs and to placebos, Lasagna et al. (1954) observed that a pos-
itive placebo response indicated a psychological set predisposing to anticipation
of pain relief, including in the case of morphine and other pharmacologically
active drugs.
Expectation is mostly seen as a kind of “conscious belief.” However, it is clear
that conscious expectation alone cannot have a placebogenic effect. Expectation
does not operate on a conscious level.You “know” that you expect something
like you “know” that your heart beats. But you cannot consciously start or end,
nor even heighten or lower, your expectation as a matter of simply deciding to
do so. Likewise, you cannot consciously heighten or lower the frequency of your
heart beat without taking recourse to subconscious help such as by visualization.
Moreover, the depth of your belief/expectation—whether it deeply touches
you, whether it moves your “heart” or “soul”—is also a factor. If you consciously
have an expectation but at the same time you do not have that expectation deep
inside, then it will hardly have any effect. D. Evans (2003) sees “belief” as under-
lying both the conditioning theory of placebo and the expectancy-based theory.
Expectancies are simply beliefs about the future, and conditioning can be seen as
one way in which such beliefs are acquired.The evidence of direct experience
may be compatible or at variance with other sources of belief, such as the voice
of authority.
Interestingly, expectation also plays a big role in psychotherapy. Most clinical
psychologists are eclectic, rather than belonging to a particular school of psy-
chotherapy. This has led to an emphasis on ingredients common to all thera-
peutic schools, the most central of which is the patient’s expectation of benefit.
This is frequently assumed to be ultimately responsible for the effectiveness of
placebos (Bootzin 1985). In this vein, A. K. Shapiro and L. A. Morris (1978) tend
to equate psychotherapy with the placebo effect. They note that while the
placebo effect is commonly believed to be just a superstitious response to a sugar
pill, it is actually an important ingredient and perhaps the entire basis for the
popularity and effectiveness of most, if not all, methods of psychotherapy.
Endorphins
It has been well documented for quite a while that endorphins are an impor-
tant element in how the body-mind complex deals with pain in a placebo set-
ting, although sometimes with mixed support (Grevert and Goldstein 1985;
Levine et al. 1979; Peck and Coleman 1991). Some authors see the modulation
of endorphins in a placebo setting as an “explanation” or “mechanism” of the
placebo effect (Levine, Gordon, and Fields 1978). But what then causes the
change in endorphins? In fact, this change is itself the consequence of the pla-
cebo effect: it is proof that such consequences can be material, but it is not an
explanation.Wall (1993) clarifies this with a nice analogy:
50
J. L. Mommaerts and Dirk Devroey
Perspectives in Biology and Medicine
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 50
If a newspaper headline reads:“Scientists discover the origin of music and
poetry” followed by an article showing that music could not be performed when
curare prevented the effect of acetyl choline released from the motor axons [and
the performer was, thereby, paralyzed], one would not be overwhelmed by the
insight into the nature of music and poetry. Similarly, it is not clear what insight
into the overall placebo phenomenon is provided by showing that some link in
the machinery involves endorphins. (p. 97)
From another viewpoint, one can say that the change in endorphins is part of
the placebo effect, with body and mind acting as one: there is no cause and effect
involved, since there is no time lapse in between, any “movement of mind” being
at the same time a “movement of body.” However, from this viewpoint, endor-
phins still do not explain the placebo effect.They only make it more complex,
as the whereabouts of our endorphins become part and parcel of our subcon-
scious mental processing.
The same is true for the changes seen in the domain of neurohormones and
of the immune system, as well as for regional metabolic brain changes. Changes
in many brain regions are clearly associated with the placebo response, with
increases in some domains, decreases in others (Mayberg et al. 2002). These
changes are broadly the same as those seen with active medication, which may
indicate a common pathway. In the domain of pain, lots of research points to the
same conclusion: placebo conditions can show large reductions in pain and in
brain activation within pain-related regions (Price et al. 2007). In any case,
although such bodily changes are sometimes used in order to explain the placebo
itself, they actually do not explain but merely point to the huge complexity of
mind-body unity.
Meaning Response
Meaning can be something that one looks up in a dictionary, but this is not
what is meant here. Something gets a “deeper meaning” when it touches you
deeply, and in the case of emotions it may prone to change your physiology (as
in blushing or changes in blood pressure).
The meaning response is the set of physiological and psychological effects of
meaning in the treatment of illness. It follows from the interaction with the con-
text in which healing occurs, such as the power of the laser in surgery, or the red
color of a stimulating medication (Moerman 2002).The placebogenic influence
of colors of medication, for instance, can be explained by their “meanings”: red
typically means “up,” “hot,” or “danger,” while blue means “down,” “cool,” or
“quiet” (Moerman and Jonas 2002). Meaning is also involved with the placebo-
genic effect of surgery, the shedding of blood being inevitably meaningful. In
addition to this, surgical procedures usually have compelling rational explana-
tions, which drug treatments often do not have (Moerman and Jonas 2002).
Jerome Frank saw this notion of meaning as what lies behind all psychothera-
The Placebo Effect
winter 2012 • volume 55, number 1 51
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 51
pies. He noted that human beings do not react to facts or events themselves, but
to their meanings. Psychotherapy can thus be seen as the transformation from
negative to positive of meanings that patients attribute to events (Holland and
Guerra 1998).
A placebo provokes a meaning response, like a key in a slot. Before the pla-
cebo turns up, meaning in the sense of the ability to respond to something
meaningful is of course already subconsciously present, and so is the power of
this meaning. The placebo only serves to awaken what is already there. It is like
a push given in the direction of a person’s faculties of self-healing.The “deeper”
the meaning response goes, the better it reaches the subconscious, thus the more
it is a genuine kind of autosuggestion.
Wh er e th e Su bc on sc io us F it s In :
Th e Co nc ep t of A ut os ug ge st io n
A lot of mental processing happens beyond the level of conscious processing. A
good example of this is vision (Churchland and Sejnowski 1992; Marr 1982).
Conscious vision is only a distinct end-product of a tremendous amount of pro-
cessing on a subliminal level. Moreover, this subconscious level is already full of
“meaning”: a lot of decisions have been made before we actually consciously
“see” something (Dennett 1992). In the human mind/brain, this mechanism is
all-pervasive (Zeki 1993).A meaningful lot happens on a subconscious level, and
taking this into account may well be very important for health and well-being.
Autosuggestion is where the subconscious fits in. Although Freud understood
autosuggestion as straightforward suggestions of “getting better/having fewer
symptoms every day” (Grünbaum 1984), in this article autosuggestion is about
“deep meaning”—this is, how one becomes deeply touched by something.
Imagine your dearest pet dies and you are deeply touched.Tears come up, sad-
ness, mourning.Your body reacts to your grief in different ways. How does this
come about? How does one become “deeply touched”? This death awakens
something inside you. It is like a very complex key. Autosuggestion is the enact-
ment of such keys, which can provoke feelings or other alterations in body and
mind.
One cannot consciously decide to have less pain.The effect of consciously
wanting it is nil. It has to reach one’s subconscious, or deeper mind, where it can
be transformed into action. The pattern “I expect this medication to relieve my
pain” means that taking this medication “communicates” to “me” that my pain
will be relieved.The placebo effect is proof that our deeper minds are capable of
many very complex things and of a kind of purpose.
Placebo as Autosuggestion
Autosuggestion in this sense brings together the placebo effect that we see in
almost all regular and alternative medicine’s medications and actions, psy-
52
J. L. Mommaerts and Dirk Devroey
Perspectives in Biology and Medicine
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 52
chotherapy,“miracle cures,” and hypnosis. It is a concept and domain that one
can communicate about, investigate, ameliorate, and make practically available.
Table 1 presents some patterns of mental processing in the form of phrases.
The phrases are examples of how it is possible to put “explanations” of placebo
in the form of overlying patterns of autosuggestion.The patterns are not actu-
ally present in the subconscious as these concrete phrases, but the patterns can
be seen as conglomerates of meaning.
Placebo-Proneness Versus Suggestibility
The placebo effect generally shows poor correlation with standardized tests of
(hypnotic) suggestibility, there being little correlation between tests of sug-
gestibility in a laboratory situation and the placebo effect in a clinical setting
(Shapiro and Shapiro 1997). However, suggestibility tests are standardized tests,
whereas suggestibility itself (and the placebo effect) is influenced by so many ob-
jective and subjective factors (context-sensitive) that it can hardly be standard-
ized. Another critique is that these tests are mainly hypnosis-oriented, and they
painfully lack sufficiently standardized guidelines for administration and inter-
pretation (Shumaker 1991). It should therefore come as no surprise that differ-
ent tests show little correlation with each other.They are no good measure for
real-life suggestibility (Barber, Spanos, and Chaves 1974).
After a long search, it has been concluded that no personality of “placebo-
responder” exists, and that virtually anyone can respond significantly to placebos
given the right circumstances (Peck and Coleman 1991).A person who has lit-
tle placebo response to a tablet may react much more so to a saline injection.
Furthermore, placebo response may be culturally colored. In general, injections
engender higher expectations in the United States than in Europe and have
The Placebo Effect
winter 2012 • volume 55, number 1 53
TAB LE 1 “E XPL ANATI ONS ”OF PLA CEB O I N THE FO RM OF OV ERLY ING
PATTE RNS OF AU TOS UGG EST ION
Hope/Faith
“I believe that this will cure me. It will definitely cure me.”
“I hope so, I long to see it become true.That’s what I want.”
“At other occasions, drugs have helped me. So I believe they will help me now.”
Expectation
“It will become true. I will get better again.”
“This device will make me feel better, as it did before.”
“This new drug is the latest development in medical science. It will surely be
better than anything else.”
Meaning response
“This means a lot to me. There just has to be something in it that is able to relieve
my suffering. I am sure it will.”
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 53
higher effectiveness, such as for relief of migraine (de Craen et al. 2000).
Logically, then, there should be no fixed percentage of placebo effect. In fact,
depending on characteristics of the clinical situation, described placebo effects
can vary from 0 to 100% (Richardson 1994).
The same is true for suggestion in general. No characteristics of “suggestible
persons” have yet really stood the test of time. Suggestibility depends mainly on
the person-context fit. Even the physiological change brought about by (auto)
suggestion/placebo varies when the subjective result is the same, as in the case
of placebo analgesia.The expectation of pain relief normally acts via endorphin
pathways, but this can be changed into other pathways by prior experience (or
“conditioning”) with an NSAID painkiller (Amanzio and Benedetti 1999; Ben-
edetti, Arduino and Amanzio 1999). With such a diversity and context-prone-
ness, it comes as no surprise that there is no consistent relationship between
placebo response and suggestibility tests (Evans 1981).
Is P la ce bo J us ti fi ed D ec ep ti on ?
This section is definitely not about empathy (caring for the patient, providing
genuine encouragement, respecting uniqueness), which can also heighten hope,
expectations, or meaningfulness (Barrett et al. 2006), but not through deception.
Instead, this section is about “placebo proper.” Cabot (1978) observed that “It
never occurred to me until I had given a great many ‘placebos’ that, if they are
to be really effective, they must deceive the patient. . . . It is only when through
the placebo one deceives the patient that any effect is produced. It is only when
we act like quacks that our placebos work” (p. 189).
Many doctors dislike the idea of placebo and like to discuss it even less.
However, it is important to engage with this practice that is still very much
among us, because some real dangers stem from the deceptiveness of placebos.
First, the use of placebos can create a loss of confidence (in doctors, colleagues,
medicine) if patients find out. Second, active placebos (pharmacological agents
used mainly or solely for their placebogenic purposes) bring huge costs and
many side effects.Third, the use of placebos heightens an inappropriate depen-
dence on medicine in general, thereby lowering the power to self-heal. Ad-
ditionally, the use of pure placebos engenders litigation suits for malpractice
unless the physician has been forthright beforehand. The notion of “informed
consent” thus actually precludes placebo: strictly speaking, prescribing a placebo
is impossible without breaking the law. Furthermore, placebos do not bring real
cures, they only alleviate symptoms. With this in mind, it is very thought-pro-
voking that present-day evidence-based medication, with a few exceptions such
as antibiotics and chemotherapeutics, acts purely symptomatically. Finally, in any
case, to really go deeper, deception will never do. It may well be that “only truth
can truly cure.”The placebo effect unrightfully mystifies faith healing and many
alternative medicines with probably no further basis than this, and it impedes the
54
J. L. Mommaerts and Dirk Devroey
Perspectives in Biology and Medicine
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 54
further development of medical science itself, thereby directly hampering tre-
mendous cost savings and at the same time a better, deeper and more encom-
passing health for all.
Despite these dangers, different authors make the case that we should con-
sciously take as much advantage as possible of the placebo effect because it is so
powerful. Placebo may be a deception, but what if deception brings health or
even saves your life? Moreover, dropping the use of the placebo effect from reg-
ular medicine (if it could be done) would give to other, less well-meaning par-
ties the exclusive advantage of using it as they have always done. That is not
something to look forward to. Indeed, one of the reasons suggested for why allo-
pathic physicians sometimes fail is that they know that some people will not
respond to their therapies, and they convey this doubt to their patients. An
acupuncturist may have an advantage simply because he or she seems certain that
a therapy will work, thereby convincing the patient that it cannot fail.
Vertosick (2000) warns: “If our patients can’t imagine a good result, they
won’t experience a good result” (p. 269).We should therefore strive to make use
of the placebo effect in a completely open manner.This means that we have to
get rid of the notion of placebo as deception. The placebo effect is just one
example of autosuggestion. So with autosuggestion we may have something that
we can use without the need of a placebo—no deception is required.
Co nc lu si on
The different explanations of the placebo effect eventually boil down to the
concept of autosuggestion (communication to the subconscious). This suggests
that the ubiquitous placebo effect is in fact a covert kind of autosuggestion. If this
is the case, we may have to rethink much of modern regular medicine. Autosug-
gestion as overt communication to the subconscious should probably be in the
center of medicine,even if it is not completely understood at present.This means
that we may have to deepen our science. At stake are health and well-being
of all.
Re fe re nc es
Amanzio, M., and F. Benedetti. 1999. Neuropharmacological dissection of placebo anal-
gesia: Expectation-activated opioid systems versus conditioning-activated specific
subsystems. J Neurosci 19(1):484–94.
Amanzio, M., et al. 2001. Response variability to analgesics: A role for non-specific acti-
vation of endogenous opioids. Pain 90:205–15.
Barber,T. X., N. P. Spanos, and J. F. Chaves. 1974. Hypnosis: Imagination and human poten-
tialities. NewYork: Pergamon Press.
Baron, J. 1990. Thinking and deciding. Cambridge: Cambridge Univ. Press.
Barrett, B., et al. 2006. Placebo, meaning, and health. Perspect Biol Med 49(2):178–98.
Beecher, H. 1955.The powerful placebo. JAMA 159:1602–6.
The Placebo Effect
winter 2012 • volume 55, number 1 55
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 55
Benedetti, F., C. Arduino, and M. Amanzio. 1999. Somatotopic activation of opioid sys-
tems by target-directed expectations of analgesia. J Neurosci 19(9):3639–48.
Benson, H., and M. D. Epstein. 1975.The placebo effect: A neglected asset in the care of
patients. JAMA 232:1225–27.
Benson, H., and D. P. McCallie Jr. 1979. Angina pectoris and the placebo effect. N Engl
J Med 300 :1424–29.
Bootzin, R. R. 1985. The role of expectancy in behaviour change. In Placebo: Theory,
research and mechanisms, ed. L. White, B. Tursky, and G. E. Schwarts, 196–210. New
York: Guilford Press.
Branthwaite,A., and P. Cooper. 1981. Analgesic effects of branding in treatment of head-
aches. BMJ 282(6276):1576–78.
Brody, H. 2000. The placebo response. New York: HarperCollins.
Butler, C., and A. Steptoe. 1986. Placebo responses: An experimental study of psycho-
physiological processes in asthmatic volunteers. Br J Clin Psychol 25:173–83.
Byerly, H. 1976. Explaining and exploiting placebo effects. Perspect Biol Med 19:423–36.
Cabot, R. C. 1978.The use of truth and falsehood in medicine. Conn Med 42:189–94.
Churchland, P. S. 2002. Brain-wise: Studies in neurophilosophy. Cambridge: MIT Press.
Churchland, P. S., and T. J. Sejnowski. 1992. The computational brain. Cambridge: MIT Press.
de Craen A. J., et al. 2000. Placebo effect in the acute treatment of migraine: Subcutan-
eous placebos are better than oral placebos. J Neurology 247(3):183–88.
Dennett, D. C. 1992. Consciousness explained. London: Penguin.
Ernst, E. 2001.Towards a scientific understanding of placebo effects. In Understanding the
placebo effect in complementary medicine, ed. D. Peters, 17–30. Edinburgh: Churchill Liv-
ingstone.
Evans, D. 2003. Placebo. London: HarperCollins.
Evans, F. J. 1981.The placebo’s response in pain control. Psychopharmacol Bul 17:72–76.
Evans, F. J. 1985. Expectancy, therapeutic instructions, and the placebo response. In
Placebo:Theor y, research and mechanisms, ed. L.White, B.Tursky, and G. E. Schwarts, 215–
28. New York: Guilford Press.
Fisher, H. K., and B. M. Dlin. 1956. The dynamics of placebo therapy: A clinical study.
Am J Med Sci 232:504–12.
Frank, J. D. 1973. Persuasion and healing. Baltimore: Johns Hopkins Univ. Press.
Glaser, R., and J. Kiecolt-Glaser. 1994. Handbook of human stress and immunity. New York:
Academic.
Gracely, R. H., et al. 1985. Clinicians’ expectations influence placebo analgesia. Lancet 84
(19):43.
Grevert, P., and A. Goldstein. 1985. Placebo analgesia, naloxone, and the role of endoge-
nous opioids. In Placebo: Theory, research and mechanisms, ed. L. White, B. Tursky, and
G. E. Schwarts, 332–50. New York: Guilford Press.
Grünbaum, A. 1984. The foundations of psychoanalysis: A philosophical critique. Berkeley:
Univ. of California Press.
Hahn, R.A. 1985.A sociocultural model of illness and healing. In Placebo:Theory, research
and mechanisms, ed. L.White, B.Tursky, and G. E. Schwarts, 167–95. New York: Guil-
ford Press.
Herrnstein, R. J. 1962. Placebo effect in the rat. Science 138:677–78.
Hoffman, G. A., A. Harrington, and H. L. Fields. 2005. Pain and the placebo: What we
have learned. Perspect Biol Med 48(2):248–65.
56
J. L. Mommaerts and Dirk Devroey
Perspectives in Biology and Medicine
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 56
Holland, H. Z., and P. Guerra. 1998. A conversation with J. Frank. Counseli ng Today. http:
//www.counseling.org/ctonline/archives/ct0898/frank.htm.
Houston,W. R. 1938.The doctor himself as a therapeutic agent. Ann Intern Med 11:1416–
25.
Johnson, A. G. 1994. Surgery as placebo. Lancet 344:1140–42.
Jospe, M. 1978. The placebo effect in healing. Lexington, MA: Lexington Books.
Kirsch, I. 1985. Response expectancy as a determinant of experience and behaviour. Am
Psychol 40 :1189–1202.
Lasagna, L., et al. 1954. A study of the placebo response. Am J Med 16:770–79.
Laska, E., and A. Sunshine. 1973. Anticipation of analgesia a placebo effect. Headache 13:
1–11.
Levine, J. D., N. C. Gordon, and H. L. Fields. 1978. The mechanism of placebo analgesia.
Lancet 2:654-657.
Levine, J. D., et al. 1979. Role of pain in placebo analgesia. Proc Nat. Acad Sci USA 76:
3528–31.
Markus, H. 1983. Self-knowledge: An expanded view. J Personality 51:543–65.
Marr, D. 1982. Vision. NewYork: W. H. Freeman.
Mayberg, H. S., et al. 2002.The functional neuroanatomy of the placebo effect. Am J Psy-
chiatry 159:728–37.
Miller, F. G., L. Colloca, and T. J. Kaptchuk. 2009. The placebo effect: Illness and inter-
personal healing. Perspect Biol Med 52(4):518–39.
Moerman, D. 2002. Meaning, medicine and the “placebo effect.” Cambridge: Cambridge
Univ. Press.
Moerman, D. E., and W. B. Jonas. 2002. Deconstructing the placebo effect and finding
the meaning response. Ann Intern Med 136(6):471–76.
Mommaerts, J.-L. 2000. Genees jezelf. [HealYourself]. Lannoo: Tielt.
Pavlov, I. P. 1927. Conditioned reflexes. London: Routledge and Kegan Paul.
Peck, C., and G. Coleman. 1991. Implications of placebo theory for clinical research and
practice in pain management. Theor Med 12:247–70.
Plotkin,W. B. 1985. A psychological approach to placebo:The role of faith in therapy and
treatment. In Placebo:Theory, research and mechanisms, ed. L.White, B.Tursky, and G. E.
Schwarts, 237–54. New York: Guilford Press.
Price, D. D., and H. L. Fields. 1997.The contribution of desire and expectation to placebo
analgesia: Implications for new research strategies. In The placebo effect: An interdiscipli-
nary exploration, ed. A. Harrington, 117–37. London: Harvard Univ. Press.
Price, D. D., et al. 2007. Placebo analgesia is accompanied by large reductions in pain-
related brain activity in irritable bowel syndrome patients. Pain 127(1–2):63–72.
Reiss, S. 1980. Pavlovian conditioning and human fear: An expectancy model. Behav Ther
11:380–96.
Richardson, P. H. 1994. Placebo effects in pain management. Pain Rev 1:15–32.
Ross, M., and J. M. Olson. 1981.An expectancy-attribution model of the effects of place-
bos. Psychol Rev 88:408–37.
Sarno, J. E. 1998. The mindbody prescription: Healing the body, healing the pain. New York:
Warner Books.
Shapiro, A. K., and L. A. Morris. 1978. The placebo effect in medical and psychological
therapies. In Handbook of psychotherapy and behaviour change: An empirical analysis, 2nd
ed., ed. S. L. Garfield and A. E. Bergin, 367–410. New York: Wiley.
The Placebo Effect
winter 2012 • volume 55, number 1 57
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 57
Shapiro, A. K., and E. Shapiro. 1997. The powerful placebo. Baltimore: Johns Hopkins Univ.
Press.
Shumaker, J. F. 1991. Introduction. In Human suggestibility, 1–33. New York: Routle dge.
Smith, G. R., and P. McDaniel. 1983. Psychologically mediated effect on the delayed
hypersensitivity reaction to tuberculin in humans. Psychosom Med 45:65–70.
Spiro, H. 1998. The power of hope: A doctor’s perspective. London:Yale Univ. Press.
Vertosick, F.T. 2000. Why we hurt. Orlando: Harcourt.
Wall, P. D. 1993. Pain and the placebo response.Ciba Foundation Symposium 174:187–211.
White, L., B.Tursky B., and G. E. Schwarts. 1985. Placebo:Theory, research and mechanisms.
New York: Guilford Press.
Zeki, S. 1993. A vision of the brain. Oxford: Blackwell.
58
J. L. Mommaerts and Dirk Devroey
Perspectives in Biology and Medicine
04_PBM55.1mommaerts 43–58:03_51.3thagard 335– 1/22/12 7:46 AM Page 58
