Scientists have known for a long time that the mind plays a role in the process of medicinal treatment. They found that the psychosocial context of a treatment is central in the strength of the placebo effect, but the underlying neurological basis for this psychological phenomenon has only recently gained momentum in the field as a topic worthy of studying. This paper will discuss the experimental procedures and results of two independent experiments, and then examine how effective these experiments are in answering the broad question of “What factors create the placebo effect?”
In the first study, Benedetti et al. (2005) recruited 28 patients diagnosed with early stages of Alzheimer ’s disease (AD) and 16 healthy participants matched for sex and age as controls. Participants went into the lab for two blood tests on two consecutive days. In the open condition, participants were given lidocaine, a local anesthetic, applied at the location of the skin region where the needle was inserted. They were able to see the anesthetic applied, and they were told that the pain should subside in a few minutes. Participants in the hidden condition were not told they were given any lidocaine; the lidocaine was secretly applied on tape that was adhered to the punctured skin. They were not told that the pain was to subside. Participants in both conditions were then asked to rate their pain according to a numerical rating scale (NRS). The researchers recorded the participants’ electrocardiogram (ECG) and electroencephalogram (EEG) while they had their blood drawn. The conditions are switched the second day they go into the lab for a blood test. All participants then came back into the lab one year later and repeated all the procedures.
The results of this study are interesting. The pain ratings and heart rate reductions for both groups were not too different from each other. After 1 year, however, EEGs showed that some AD patients had significant reduction of activity in the prefrontal regions while others showed reduction in the temporal-parietal-occipital regions. Evidently, when the reduced activity was in the prefrontal regions, the AD patients showed no difference in their pain ratings or heart rates regardless of the conditions they were in. The AD patients with reduced activity in the prefrontal lobes were the only group that showed no difference between open and hidden lidocaine at the second test administered one year later. These results support the researchers’ hypothesis that the prefrontal lobes play a crucial role in the placebo effect. Reducing communication between the prefrontal and the rest of the brain may make analgesic treatments less effective because the expectation-related mechanisms are lost.
The design of this experiment is appropriate for the question they are trying to answer. Benedetti et al. were able to use a placebo without administering any injections into the body or needing to deceive their participants. They were able to separate the psychological effects and pharmacodynamic effects caused by the lidocaine and with the ECG, they were able to see if the lidocaine had any physiological effects in the body that the participant may not be aware of. The mismatch they found between their controls and AD patients in terms of the pain and heart rate reductions made it clear that a working prefrontal lobe is important in determining the strength of placebos. Without the expectation mechanism that is in healthy participants, the AD patients was not under the influence of the placebo effect. It seems that the prefrontal lobe’s ability to make predictions plays an important part facilitating the placebo effect.
In another experiment, researchers Scott et al. (2007) used different methods to measure how placebos affect the brain. In their study, they looked at how variable is the placebo effect across a group of individuals. They hypothesized that in healthy subjects, individual variation in placebo-induced nucleus accumbens (NAC) dopamine (DA) activity and in the synaptic activity of this region during reward anticipation are correlated with the variability of the placebo effects. They used functional magnetic imaging (fMRI) and positron emission tomography (PET) to see what areas of the brain are activated while the participants perform their tasks.
For the PET segment of the study, participants were first asked to rate how effective they expect the analgesia to be. First in the non-painful expectation condition, they were injected with sterile isotonic saline and told to rate the pain intensity every 15 seconds for 20 minutes using the visual analog scale (VAS). After that trial was over, they were then injected with 5% hypertonic saline solution which was supposed to induce pain. The target pain rating was maintained between 30-40 VAS intensity units. Participants then entered the placebo condition where they were injected with 1 ml 0.95% isotonic saline intravenously every 4 minutes over 20 minutes. They were asked to estimate the expected analgesia before the placebo and then asked to rate the efficacy of the placebo after the pain challenge. In the fMRI segment, participants were in the scanner while performing the Monetary Incentive Delay (MID) task, which is to respond to target displays while anticipating monetary reward. The MID is known to activate NAC synaptic activity.
The results of this study showed that the anticipation of analgesia was significantly and positively correlated with the average reduction in pain ratings in the PET study (r = 0.48, p < 0.05). They also showed a positive correlation between fMRI BOLD activation during the reward anticipation and analgesic effects of the placebo trials (r = 0.53, p = 0.02). Scott et al. concluded that this latter finding accounted for 28% of the variance in the formation of placebo analgesia.
The design of this latter study was suitable for the question the topic the researchers wanted to address. They were able to administer a placebo and effectively used brain imaging techniques to clearly see which parts of the brain is most activated when the placebos were taking effect. They showed that expectation is an example of reward processing and that individual variations in brain response are associated with the differences they felt during a placebo administration. The researchers did a good job in controlling for individual tolerance for pain, as this would certainly alter their findings if everyone’s pain tolerance level is different.
Overall, the findings of the two studies supplement each other. While implementing experimental procedures using different methods, the studies were able to investigate the areas of the brain involved in the expectation mechanisms that play an important role in the placebo effect. One limitation of the study by Benedetti et al. is that their use of the EEG method is not as precise as the methods used by Scott et al. EEG recordings only review brain activity of the cortical areas of the brain. Hence, Benedetti et al. were not able to pinpoint which area of the prefrontal cortical area is most crucial for the placebo effect to take place. They know that AD patients, because of their weakening cognitive abilities, are less affected by the placebo, but they do not know which specific part of the brain may be causing this. Benedetti et al.’s findings, do give strong implications on what should be done in real therapeutic settings. Because AD patients with prefrontal lobe impairment are less sensitive to the placebo effect and because we know that the placebo effect is beneficial for treatment recovery, this study suggests that analgesic treatments for patients with pathological conditions involving the prefrontal lobes should be increased to compensate for the loss of placebo mechanisms.
Scott et al.’s study is a stronger study in the sense that they are able to give a better picture of what specific part of the brain is involved – namely the dopaminergic neural regions in the nucleus accumbens. They were also better able to see what role expectancy plays in the placebo effect, as they specifically used a task that measured the participants’ expectancies on monetary rewards. Their finding generalizes the ventral basal ganglia as an important mechanism for placebo-associated experiences. The more active that area is (more expectation for the placebo to work), the more effective the placebo will actual be for the subject expecting treatment.
Both studies show that intrinsic neurobiological mechanisms will influence how much of a placebo effect an individual may experience. These results greatly contribute to the current literature on the placebo effect and have potential to drive revisions in many therapeutic methods so that treatment can help patients to recover quickly. Some other questions that can be further addressed include: what is the involvement of both dopamine and opioid in placebo effects, how important is the placebo effect in treatment, and does the disruption of place-related mechanisms also affect treatments other than analgesics?
Benedetti, F., Arduino, C., Costa, S., Vighetti, S., Tarenzi, L., Rainero, I., & Asteggiano, G. (2006). Loss of expectation-related mechanisms in Alzheimer’s disease makes analgesic therapies less effective. Pain: 121, 133-144. doi:10.1016/j.pain.2005.12.016.
Scott, D.J., Stohler, C.S., Egnatuk, C. M., Wang, H., Koeppe, R. A., & Zubieta, J. (2007). Individual differences in reward responding explain placebo-induced expectations and effects. Neuron, 55, 325-336. doi: 10.1016/j.neuron.2007.06.028.