No One Likes A Sure Thing
Parkinson's patients had their dopamine agonists (Sinemet) stopped, and were then told that they would be assigned to one of 4 groups:
You have been randomly assigned, like pulling numbers out of a hat, to Group A. As you read in the consent form, this means that you have a 25% chance, or 1 in 4 chance, of receiving active Sinemet... We took one real Sinemet pill and three placebos and shook them up and withdrew one. This is what we are giving you.
Group B was 50%, C was 75%, and D was 100% chance of getting Sinemet.
In reality, everyone got placebo; all that was different was the patient's expectation of receiving Sinemet.
Those who were told they had a 75% chance of getting Sinemet exhibited a maximal (placebo) response. It was the same amount of clinical improvement, and the same amount of actual dopamine release, as they exhibited back when they were still taking their meds.
This is actual dopamine release (as measured by the decline of a tracer):
and this is clinical improvement:
Interpret this as "less than 50% chance of getting Sinemet meant very little response; more than 50% meant more response."
Perhaps they had a placebo response to the symptoms ("hey, I feel better") which then caused a subsequent release of dopamine, especially in the nucleus accumbens which is the "reward center"? No: dopamine release was entirely dependent on the expectation of the medication.
Think about the words. When you are told it's a 50% chance of it being Sinemet, you are checking with your body to see if it is Sinemet. When you are told it is 75% chance, you are expecting it to be Sinemet. 50% marks a turning point between skepticism and expectation.
Now think of the placebo effect in most other scenarios, e.g. clinical trials: "you're going to get drug or placebo." Imagine, instead, that they told you, "there's a 75% probability of receiving placebo."
II.
So what would happen if you were told that there was a 100% chance of receiving Sinemet, i.e. you were lied to?
Why would 100% certainty result in less dopamine release, and less clinical improvement, than 75%?
Because the placebo response is not simply the body reproducing the effect of the real drug. The placebo response is a reward; and if there is no reward, there is no response.
It doesn't matter what kind of placebo response you want-- pain/endorphin release may not seem dopamine related, but the results would probably be identical. It appears that all placebo responses are substantially mediated by the reward circuits. These are unconscious and immediate. Expecting a response is the same as expecting a cookie, the same pathways are used; and, conceivably, they can be destroyed in the same way.
A more scientific explanation is that the tonic, sustained increases in dopaminergic activity preceding the presentation of a reward are related to the level of uncertainty. 50% probability of reward represents maximal uncertainty, and hence maximum dopaminergic activity. ("Is it going to happen?") It declines like an inverted U from there ("I'm more sure it is/is not going to happen") to zero when you are completely sure.
Another way to say it is: the more that you are unconsciously making a reward prediction (nucleus accumbens), the greater the effect; but certainty bypasses this, leaving no reward contribution to the final clinical effect. In that case, it's all up to the pill.
III.
Feel free to extrapolate.
Digg | 
del.icio.us | 
StumbleUpon | 
Reddit | 
Yahoo! | 
My favourites
August 10, 2010 2:39 PM | Posted by : | Reply
This is why I only show up for half of my appointments with psychiatrists.
Score: 31 (35 votes cast)
August 10, 2010 5:20 PM | Posted by : | Reply
Since the title begs the question, I'll ask the question: does this apply to the expectation of sex, and how? I'm guessing that if sex is the "drug", then it will have its maximum effect when you are most uncertain of getting it?
Score: 8 (12 votes cast)
August 10, 2010 5:36 PM | Posted by : | Reply
@Anonymous 5:20 PM:
If are unable to factually determine whether you have succeeded in obtaining sex, you should seek professional help.
Score: 12 (30 votes cast)
August 10, 2010 6:38 PM | Posted by : | Reply
Okay so here is the study I want to see happen. What if we tried giving people placebos and telling them that they were getting a placebo, but that scientific researched proved that placebo was (25, 50, 100%). And then see if the people KNOWING they were getting a placebo, but having been coached in the hard science of it, recieve improvement.
We've got to find a way to use the power of placebo without lying to people. Couldn't that expectation, belief in cure etc be harnessed through belief in the science of placebo, rather than just the belief in a pill?
Score: 1 (3 votes cast)
August 10, 2010 7:02 PM | Posted by : | Reply
Of course clinical trials are designed to show maximum effectiveness. "Pay no attention to the man behind the curtain!"
And the notion of controlling for placebo, in which there was so much faith, was actually. . .a placebo.
Score: -4 (6 votes cast)
August 10, 2010 7:50 PM | Posted by : | Reply
For completeness, they should have run a second iteration with the same odds explanation, but then have given everyone the drug instead of the placebo.
But what do I know? I'm just a hermitian operator...
Score: 4 (4 votes cast)
August 10, 2010 8:20 PM | Posted by : | Reply
What's your answer to pills that don't work as well as our minds -- but are the justification (?) of psychiatric industry and all it costs us?
Looking forward to your comments on Whitaker's latest book.
Score: 2 (2 votes cast)
August 10, 2010 8:45 PM | Posted by : | Reply
This must be the same uncertainty that causes such a huge placebo response to SSRI's and such.
Score: 2 (2 votes cast)
August 10, 2010 9:04 PM | Posted by : | Reply
I think that drugs probably are already harnessing the placebo response. If you improve when given an inert substance because you expect that it could help it seems sensible to me you should also improve when given an active substance you expect to help (due to placebo mechanism as well as 'real' mechanism)
The remaining advantage of placebos is they are way cheaper compared to real drugs.
Haha, although obviously the study this post is about complicates things a bit.
Score: 2 (2 votes cast)
August 10, 2010 10:24 PM | Posted, in reply to , by : | Reply
There are arguments that the placebo mechanism also accounts for the improved response in the non-placebo group in SSRI studies -- they notice the drug's side effects, and so they become more certain that they received the real pill, and they respond physiologically to that strengthened belief.
Score: 2 (2 votes cast)
August 11, 2010 12:32 AM | Posted by : | Reply
I am HIGHLY skeptical of this.
1) They have to compare these results to a group of people who were taking sinemet as usual without any expectations either way. As it is now, we don't know if the results are evidence of the power of a positive expectation, or evidence of the power of a negative expectation. Is dopamine release being suppressed by the idea of taking a placebo (those in the 25% group) or is it that dopamine release is being augmented by the idea of taking a drug?
2) Those bars on the graph mean nothing to me. For all I know, the actual adjustment in dopamine is clinically insignificant. Just because the graph is presented in such a way as to suggest a 100% increase in dopamine release when positive expecting a drug (or a 50% decrease if negatively expecting a drug), this doesn't MEAN that those changes are meaningful clinically at all.
FOr all I know, active sinemet would result in an increase of dopamine 10 times that of group C.
We can't know because all we are comparing here are placebo responses, with a fancy graph intentionally designed to make us think these group differences mean anything.
If I have 3 cents, and wishful thinking that my lotto ticket is a winner makes me have 6 cents, at the end of the day, 6 cents is socially insignificant regardless of the fact I increased my monies 100%. An actual winning lotto ticket is worth waaaay frigging more, in a socially significant way.
Regarding the clinical improvement graph... again, what do these numbers actually mean? Yes, the groups b and c had a 100% improvement, but improvement over what ?
And in the same vein, we see that those who were expecting "no reward" (certainty of getting sinemet) have less of a dopamine response than those who were gambling (75% expectation)... but STILL it is impossible to know how meaningful these changes are.
if group a has 3 cents and group b has 12 cents, they all still have no money at all from a functional perspective.
3) It is interesting these researchers selected DOPAMINE to test this study of placebo effect. It is well known that gambling is a dopamine driven behavior (in the sense that risky stressful things increase dopamine, and increased dopamine similarly leads to exhibited, disinhibited, risky behavior). Would we find similar results on blood pressure and metoprolol? I bet frigging not.
It's a warm and fuzzy idea to think that we can control our bodies with our minds, but no, NOT REALLY we cant. Sure, gambling can raise dopamine (the expectation of rewards, leads to a dopamine increase)... but this doesn't work at all with hypertension, with diabetes, with the majority of health conditions that have little to do with dopamine and other stress neurohormones.
And, on top of what I just said, we still haven't even demonstrated that these changes in dopamine release from gambling are significant in a parkinsonian patient group. If you don't make that much dopamine at baseline, how fucking much can you be expected to get a gambling high? Probably not that much. Manic patients maybe the opposite is true, lol.
We should test placebo response in manic patients and measure dopamine changes. I bet that would be fun.
Score: 10 (16 votes cast)
August 11, 2010 2:06 AM | Posted, in reply to , by : | Reply
Well if you take this study at face value, then no it wouldn't work because then you have an expectation that the placebo will work for you.
Score: 1 (1 votes cast)
August 11, 2010 4:32 AM | Posted, in reply to , by : | Reply
I'm pretty sure the entire concept of a placebo is predicated on it being a lie. It's definitional. It's like asking if there's a way to fly without going into the air.
Score: 1 (1 votes cast)
August 11, 2010 8:44 AM | Posted by : | Reply
'Feel free to extrapolate'
This is about relationships, isn't it?
Score: 3 (3 votes cast)
August 11, 2010 10:40 AM | Posted by : | Reply
And the University of British Columbia Hospital IRB was, like, okay with the flagrant deception of group D?
Score: 0 (0 votes cast)
August 11, 2010 10:24 PM | Posted by : | Reply
what did the authors want to be true?
Score: 4 (4 votes cast)
August 11, 2010 11:05 PM | Posted, in reply to , by : | Reply
Doesn't it disturb you a little bit that you have given over intellectual power to an anonymous blogger? That would make me a bit sad.
Score: -3 (3 votes cast)
August 12, 2010 12:32 AM | Posted, in reply to , by : | Reply
Anonymous blogger or ethereal deity? I suppose it's all in the preference of deception, really.
Score: 0 (2 votes cast)
August 12, 2010 9:59 AM | Posted by : | Reply
I dunno, the explenation of homeopothy is utter nonsense, and people believe in that stuff. I'm just saying what if rather than telling people, "This is a magic sugar pill that been diluted 100 times from some substance that would have killed you if any real trace of it were actually in the pill, but the way we diluted it until it wasn't even present in the pill left a mystical force of power that's going to heal you. So here you go"
We tried, "We've been studying the placebo affect and have found that just the act of taking a pill that you believe works can actually heal your body. Your body actually has the power of healing itself if you take this pill and believe in your bodies own ability to heal your body really will heal from (whatever)."
I really can't believe the results would be THAT far from homeopothy?
Per how those results would compare to this study, if everyone was told they were getting a placebo, but that the placebo affect was going to cure them and they had a 25%, 50% 100% chance of being cured based on their past research, I'm not sure how the admission would actually affect the results. Hence why it would be interesting to see the results of such a study.
But really, "Magical force that cures you if you believe in it" (placebo) vs "Magical force that cures you from a sugar pill that's been homeopathically charged"?? How different is that message?
It seems, to me, like there would be a good possibility that people could be coached to use the placebo affect even knowing it was placebo, such as we have seen in some areas of health with biofeedback.
Score: 2 (4 votes cast)
August 12, 2010 12:56 PM | Posted by : | Reply
So I guess double blind trials are not really necessary, since if someone is certain they are getting the active drug, there is no placebo effect exhibited.
Score: 0 (2 votes cast)
August 13, 2010 6:08 PM | Posted by : | Reply
So what happens if you set the expectations the other way? Shake up 10 pills in a bottle, and tell the patient they only have a 10% chance of getting real medicine when in fact all 10 pills are the real thing? Does negative expectation lessen the pill's effectiveness?
Score: 4 (4 votes cast)
August 15, 2010 10:57 PM | Posted by : | Reply
"But really, "Magical force that cures you if you believe in it" (placebo) vs "Magical force that cures you from a sugar pill that's been homeopathically charged"?? How different is that message?"
Placebos are NOT magical. They do not use any "magical forces". The message that they do has NO place in responsible medicine.
--- --- ---
As for the study, what about 0% expectation? Seems incomplete without it. In science all assumptions are the enemy, even the necessary ones.
Score: -2 (2 votes cast)
August 16, 2010 1:14 AM | Posted by : | Reply
in response to anonymous
"Doesn't it disturb you a little bit that you have given over intellectual power to an anonymous blogger? That would make me a bit sad."
i havent given over any power, just wondering why TLP didn't ask this question himself. the results, and the motivation of the investigators, seem to have been accepted without question. perhaps this study is in step with TLPs bias.
i have other questions, but they make a post like this long, and long posts are often not read. why is actual dopamine release not related to clinical improvement? how did they get away with lying to subjects about almost every element of the study? how long had sinemet been held and was this trial one dose or a course of weeks? parkinsons is a disease that damages dopamine pathways, is it reasonable to extrapolate this data to people with normal dopamine pathways?
Score: 2 (2 votes cast)
August 17, 2010 1:16 PM | Posted by : | Reply
"Placebos are NOT magical. They do not use any "magical forces". The message that they do has NO place in responsible medicine."
Oh, you're kidding, placebos are not magical?!! Thanks for informing me!
But yes of course if something occurs in the real word it automatically means it is the stuff of real science, which is the study of things that actually happen.
Meaning placebo affect is, of course, real in every sense of the word.
But, really, we are talking about the body healing itself through a mechanism that we don't fully understand. It sounds just as outlandish as the explanation of homeopathy. And if placebo affect works even though there is no actual medicine being administered, it should make sense that we could identify and harness whatever the actual biological process that causes placebo healing outside of lying to people.
Why does it work to tell someone they are receiving a medicine that will cure them, and not work when people are told they are getting a placebo that will cure them?
Shouldn't the same force that causes people to heal when given a sugar pill work just as well if they were told that sugar pills actually have a success rate that is based in real science?
Which was my point. I'm sorry you missed it while dwelling on semantics.
Score: 1 (1 votes cast)
August 17, 2010 6:25 PM | Posted by : | Reply
Hang on a moment. Isn't dopamine involved in tasks involving the uncertainty of rewards? I bet that this ain't a placebo effect.
http://www2.psych.ubc.ca/~krigolson/ODoherty%202004.pdf
"Human neuroimaging studies of classical conditioning for reward report prediction error signals in prominent target areas of dopamine neurons, namely ventral putamen and OFC"
"... dopamine release has been reported in the striatum during reward prediction using positron emission tomography (PET) ligand measures"
Score: 2 (2 votes cast)
December 19, 2010 1:23 AM | Posted by : | Reply
It's well known that cash makes us autonomous. But how to act when one doesn't have cash? The one way only is to try to get the home loans or short term loan.
Score: 0 (0 votes cast)
July 27, 2011 7:06 AM | Posted by : | Reply
I simply wanted to compose a small remark to be able to express gratitude to you for all the stunning tips and tricks you are giving on this website. My rather long internet research has finally been compensated with reputable information to go over with my classmates and friends. I would repeat that most of us website visitors are extremely endowed to dwell in a fantastic network with very many awesome professionals with good principles. I feel rather fortunate to have encountered your site and look forward to some more exciting moments reading here. Thanks a lot once again for all the details.
Score: 0 (0 votes cast)
June 24, 2013 12:52 AM | Posted by : | Reply
This is a very comprehensive clinical analysis. That is applaudable. However, it is important to note that analyzing awesomeness is something that can be not given much attention on. Luis of research papers.
Score: 0 (0 votes cast)
November 14, 2013 10:12 PM | Posted by : | Reply
There's an interesting article about the connection between dopamine and gambling. Recent medical studies have found that anywhere from 3 to 13 percent of patients on the kind of medication Klinestiver was taking [synthetic dopamine] develop severe gambling addictions or related compulsions.
Score: 0 (0 votes cast)
September 24, 2014 11:27 AM | Posted, in reply to , by : | Reply
You mean go to a prostitute?
Score: 0 (0 votes cast)
Comments