Pain: Placebo Effects

Fabrizio Benedetti. Encyclopedia of Perception. Editor: E Bruce Goldstein. 2010. Sage Publication.

The placebo effect is the reduction or the disappearance of a symptom when an inert treatment (the placebo) is administered to a subject who is told, and indeed believes and expects, that it is an effective therapy. Most of the knowledge about its mechanisms comes from the field of pain, thus placebo analgesia is currently the most investigated model. However, other systems and apparatuses, such as the motor, immune, and endocrine systems, are emerging as interesting models. Although the placebo effect has so far been considered a nuisance in clinical research when a new treatment has to be tested, it has now become a target of scientific investigation to better understand the physiological and neurobiological mechanisms that link a complex mental activity to different functions of the body. Usually, in clinical research the term placebo effect refers to any improvement in the condition of a group of subjects that has received a placebo treatment. Conversely, the term placebo response refers to the change in an individual caused by a placebo manipulation. However, today these two terms are used interchangeably.

The placebo effect is basically a context effect, whereby the psychosocial context around the patient plays a key role. For example, the therapist’s words, the sight of complex machines, and other sensory inputs that tell the patient that a treatment is being performed, all represent important factors in the occurrence of a placebo response. In the case of pain, this psychosocial context is capable of modulating pain perception. This is the reason why the placebo effect is currently a useful model for understanding the complex psychological modulation of pain. This entry describes the identification and mechanisms of the placebo effect, as well as the nocebo effect (a placebo effect in the opposite direction).

Identification of the Placebo Effect

The investigation of the placebo effect is full of pitfalls and drawbacks because, in order to identify a real psychobiological placebo response, several other phenomena have to be ruled out. For example, most painful conditions show a spontaneous temporal variation that is known as natural history. If subjects take a placebo just before their discomfort starts decreasing, they may believe that the placebo is effective, although that decrease would have occurred anyway. Clearly, this is merely a misinterpretation of the cause-effect relationship. Another example is regression to the mean, a statistical phenomenon whereby individuals, after reporting severe pain at an initial clinical assessment, tend to report lower levels of pain at a second assessment. A further source of confusion is represented by signal detection theory, whereby errors in the detection of ambiguous signals may occur. For example, the subject may believe that a pain reduction has occurred, a false positive, although no real reduction is present. It also happens that a co-intervention actually is responsible for the reduction of a symptom, such as the analgesic effect following the mechanical insertion of a needle to inject an inert solution. All these examples show that, although an improvement may occur after the administration of a placebo, the placebo is not necessarily the cause of the effect that is observed.

Because all these phenomena are sometimes difficult to identify, the mechanisms of the placebo response must be investigated under strictly controlled experimental conditions. For example, in order to rule out spontaneous remission, a group taking the placebo is compared to a group receiving no treatment, the latter giving information on the natural course of the symptom. The difference between the placebo group and the no-treatment group represents the real psychobiological placebo response.


The placebo effect involves both cognitive factors and conditioning mechanisms. For example, the deceptive administration of a placebo treatment can lead the subjects to believe that the treatment is effective, so as to induce positive expectations about the therapeutic outcome. Indeed, several studies show that different verbal instructions lead to different expectations and thus to placebo responses of different magnitude. The context around a therapy may therefore act through more than only expectation and conscious anticipatory processes. In fact, the placebo response is sometimes a conditioned response due to repeated associations between a conditioned stimulus (e.g., shape and color of pills) and an unconditioned stimulus (the active substance inside the pill). In this case, it is the context itself that is the conditioned stimulus. However, even by considering a typical conditioning procedure, it has been shown that a conditioned placebo analgesic response can result from conditioning but is actually mediated by expectation. In other words, conditioning would lead to the expectation that a given event will follow another event, and this occurs on the basis of the information that the conditioned stimulus provides about the unconditioned stimulus.

There is not a single placebo effect but many effects, with different mechanisms at work in different circumstances. Sometimes cognitive mechanisms are more important than conditioning, and vice versa. For example, verbally induced expectations of either analgesia or hyperalgesia completely eliminated the effects of a conditioning procedure in experimentally induced pain. By contrast, verbally induced expectations of either increase or decrease of hormones do not have any effect on their secretion. However, if a pharmacological preconditioning is performed by means of repeated administrations of a hormone-stimulating drug, conditioned hormonal placebo responses can be induced. These findings suggest that placebo responses are mediated by conditioning when unconscious physiological functions, like hormonal secretion, are involved, whereas they are mediated by expectation when conscious physiological processes, like pain, come into play. Thus, the placebo effect seems to be a phenomenon that can be learned either consciously or unconsciously, depending on the system that is involved (e.g., pain or hormone secretion).

There is now compelling experimental evidence that in some circumstances placebo-induced analgesia is mediated by endogenous opioid systems—opioid systems that originate naturally in the body. In fact, several studies found that placebo analgesia is reduced by the opioid antagonist, naloxone. In addition, some in vivo receptor binding studies, whereby the activity of neurotransmitter receptors is assessed, found the activation of opioid neuro-transmission in some brain regions, like the anterior cingulate cortex, the dorsolateral prefrontal cortex, the insula, and the nucleus accumbens. Moreover, the cholecystokinin (CCK) antagonist, proglumide, has been found to enhance the placebo analgesic effect, which indicates that CCK has an inhibitory role in placebo-induced analgesia. The placebo analgesic response is thus the result of the balance between endogenous opioids and endogenous CCK.

Placebo analgesia is not always mediated by endogenous opioids. In fact, if the placebo response is induced by means of either strong expectation cues or morphine preconditioning, it can be blocked by the opioid antagonist naloxone. Conversely, if the placebo response is induced by means of prior conditioning with a non-opioid drug, like ketoro-lac, it is naloxone insensitive. Today we know that specific placebo analgesic responses can be obtained in different parts of the body, and that these responses are naloxone reversible, which suggests that the placebo-activated endogenous opioid systems have a somatotopic organization. In other words, opioid-mediated placebo responses occur only in those parts of the body where a placebo cream had been applied.

The investigation of placebo analgesia by means of brain imaging techniques found that similar regions of the brain, such as the anterior cingulate cortex, are affected by both a placebo and a narcotic drug, which indicates a related mechanism in placebo-induced and opioid-induced analgesia. In addition, during the anticipatory phase of the placebo analgesic response (i.e., when the subject expects the analgesic effect), an activation of the dorsolateral prefrontal cortex, orbitofrontal cortex, superior parietal cortex, periaqueductal gray, and other frontal regions occurs, suggesting the activation of a cognitive-evaluative network just before the placebo response.

Brain imaging studies have also shown the involvement of dopamine in the nucleus accumbens in placebo analgesia. As the nucleus accumbens is involved in the neuronal circuitry of reward mechanisms, these studies suggest that an important mechanism in placebo responsiveness might be reward. In the case of the placebo effect, the reward is represented by the clinical benefit.

The Nocebo Effect

The nocebo effect, or response, is a placebo effect in the opposite direction, whereby expectation of pain increase may induce a hyperalgesic effect. In this case, anticipatory anxiety plays a fundamental role, as the induction of a nocebo response represents a stressful and anxiogenic procedure. The term nocebo (“I shall harm”) was introduced in contraposition to the term placebo (“I shall please”) by a number of authors in order to distinguish the pleasing from the noxious effects of placebo. It is important to stress that the study of the nocebo effect relates to the negative psychosocial context surrounding the treatment, and its neurobiological investigation is the analysis of the effects of this negative context on the patient’s brain and body. As with the placebo effect, the nocebo effect follows the administration of an inert substance, along with the suggestion that the subject will get worse.

Brain imaging techniques have been crucial to understanding the neurobiology of negative expectations, and most of this research has been performed in the field of pain. Overall, negative expectations may result in the amplification of pain, and several brain regions, like the anterior cingulate cortex, the prefrontal cortex, and the insula, have been found to be activated during the anticipation of pain.

Besides neuroimaging techniques, pharmacological studies give us insights into the biochemistry of the nocebo effect and of negative expectations. A model has recently been proposed whereby the opioidergic and the CCK-ergic systems may be activated by opposite expectations of either analgesia or hyperalgesia, respectively. In other words, verbal suggestions of a positive outcome (pain decrease) activate opioid receptors, while suggestions of a negative outcome (pain increase) activate CCK receptors. The involvement of CCK in nocebo hyperalgesia is likely to be mediated by anxiety, as benzodiazepines have been found to block both nocebo-induced hyperalgesia and the typical anxiety-induced hypothalamus-pituitary-adrenal hyperactivity. Conversely, the CCK antagonist, proglumide, has been found to prevent nocebo hyperalgesia but not the hypothalamus-pituitary-adrenal hyperactivity, which suggests two independent biochemical pathways activated by nocebo suggestions and anxiety.

More recent studies have found that nocebo effects are also associated with a decrease in dopamine and opioid activity in the nucleus accumbens, thus underscoring the role of the reward and motivational circuits in nocebo effects as well. In other words, the activation/deactivation balance of both dopamine and opioids in the nucleus accumbens would account for the modulation of placebo and nocebo responses. Therefore, different neurotransmitters, such as CCK, dopamine, and opioids, are involved in the complex psychosocial modulation of pain perception by placebos and nocebos.