I have already written a few posts about the pros and cons of The placebo effect. The placebo effect can loosely be defined as the effects you get when you think you are receiving medical care, be it psychological counselling or an injection of some substance.
I have also written previously about classical conditioning (see for example Learning described at the cellular level: Finding from our laboratory in Lund). Classical conditioning is a type of learning in which you pair a neutral stimuli such as a tone, a light, a touch, a picture, a thought or something else with another stimulus that has an associated reflex such as a puff of air in the eye (causes you to blink), a punch in the face (also causes you to blink), an electrical shock (causes fear and pain), or a shot of morphine (causes you to enter heaven for a short while).
If you present a neutral stimulus to someone and then, directly afterwards a reflex eliciting stimulus to the same person, and then repeat this procedure many times, then you will eventually notice that merely presenting the neutral stimulus can cause the person to perform the reflexive behavior that was formerly associated only with the second stimulus.
For example, play a tone just before giving an air-puff to the eye of your participant, over and over again, and you will eventually see that the participant blinks when he or she hears the tone. Show someone a picture just before you give that person an injection of morphine, and eventually (after a number of repetitions), that person will feel as if he or she was in heaven just from looking at the picture.
I don't know whether the relationship between the placebo effect and classical conditioning has become obvious to you yet, but if not, let me spell it out. Say that you have a spinal injury causing permanent and intensive pain. In some cases the only available therapy is regular morphine injections. You go to your doctors, enter the hospital, the doctor takes out an injection needle, and sticks it through your arm into a vein and injects a fluid which contains morphine. Notice that all these experiences leading to up the injection of the morphine can work as a neutral stimuli that is consistently associated with an injection of morphine. In other word, it is likely that one of these neutral stimuli such as seeing the injection needle (just like seeing a nice steak will cause an increased saliva production (at least for me)), will start processes in your body similar to the ones you experience when you get the actual morphine injection.
This was a hypothetical example for which I have no reference, however, you can find examples in the scientific literature which quite strongly supports the idea that placebo's sometimes are a form of classical conditioning.
For example it has been shown that people who are allergic to pollen, can get allergic reactions just from seeing a flower, even if it is made of plastic and therefore has no pollen.
In a more extreme example, a boy was getting chemotherapy to treat his cancer. The chemotherapy was paired with a specific taste and smell. Later on, the boy was presented with only the taste and smell and surprisingly (or not), the doctors observed the same effects as during the "real" chemotherapy...
These results have all but convinced me that there is indeed many parallells between the placebo effect and classical conditioning. Further these findings do suggest that you can help patients quite a lot without actually giving them "real" medicine. However, I also think it is wise not to go to far with this (see again: The placebo effect).
At the very least, this is a truly fascinating subjects that I, for one, will continue to follow.
See also my post on Complementary and alternative medicine - Spontaneous recovery from disease