Doubling Down
Doubling Down
“Belief creates the actual fact.”
—William James
Doubling Down
A fascinating aspect of science is the many measures used to ensure clean results. One such measure is double-blinding, a standard practice used to reduce bias by ensuring that neither the participants nor the researchers know who received which test condition. In some studies, participants may also be kept unaware of the precise hypothesis being tested.
Bear with my crappy example as I explain the rigors of a simple test. Imagine we’re testing if smelling animal dung improves the performance of playing basketball. If you take a group of people, have them smell dung, and then shoot some basketballs, you’ll find their average to be, let’s say, 10 successful shots in some time period. Is that better than if they didn’t smell the dung? We don’t really know. It’s a useless test.
Instead, you have them shoot hoops first, then smell dung, and then test again. Now, after starting with 10, they’ve improved to 13 successful throws after the dung smelling. Astonishing! Dung indeed boosts performance! Except, we’ve failed to account for the fact that they’ve had a little practice now. So how do we account for that? The control group.
You have one group smell dung and one group smell water and if you find that both groups shoot the same average, then clearly dung was no more useful than water. There is one more issue though, and this is where the double-blind part comes into play. Remember before how we discussed how our own emotions and expectations can impact our performance? If the dung-ballers believe that the dung is supposed to enhance their performance… it will!
The participants must be blind to the fact that you’re testing if dung improves basketball performance. All you can tell them is you’re testing average basketball ability in general. Then as they wait in a waiting room, you walk in with dog poo on your shoe and take them to the court while they smell your nastiness. The control group doesn’t have to deal with your foul odor. Now you compare and find the dung group really does make for better performance! Excellent! Except…
You might also remember that we’re talking about double blind studies. We only blinded the participants in this case. If you still know who you expect to perform better, much like in the rat maze situation, you could still somehow influence the results. Instead, you’ll need to tell someone else to go tell the participants to shoot some hoops while you hide a little poo somewhere else. Now the experimenter and the experimentee are both properly blinded to what is being tested and these pesky little emotional transferences are no longer a problem. Turns out, smelling crap makes for crappier performance. How unfortunate.
These effects are truly astonishing. Mere expectation can cause an experiment to behave accordingly. The accepted explanation for this outcome as stated before involves subtle micro-cues like tone, posture, facial expressions, or, presumably in the case of our maze rats, care in handling. But even that feels difficult to wrap the mind around. A micro smile can make someone shoot better hoops?
We are also concluding that if a researcher is handing out pills as part of a medication trial and they know which pills are real and which are placebo, their unconscious behavior can give away the answer to the participant. Even the smallest change in tone, eye contact, or gesture might tip off a participant and subsequently skew the results. Further, somehow picking up on this is itself enough to alter the actual performance of the drug.
Frankly, this borders on the mystical. Humans are notoriously poor at detecting obvious signals let alone micro signals. How many times do we have to mumble, “wow, that’s crazy” and slowly back away before someone gets the hint we’re not interested in their story? How many people can’t pick up they are being flirted with or that they have been put in the friend zone? People miss such obvious cues and we’re told they pick up on micro-cues? How could someone receiving a pill from an unknown scientist detect a near-imperceptible smirk or shift in tone and suddenly know whether they got the real drug?
It seems absurd—and yet, the data confirms the effect is real. Imagine in our maze-rats scenario that they were testing an intelligence drug. The students knowing which group was supposed to be intelligent would have clearly skewed the results and incorrectly validated the useless drug. It would have made the drug appear effective when it was simply their bias impacting the results.
This raises a second, even more astonishing question: why does that knowledge change the outcome in the first place? With the rats, we must conclude that they were deeply affected by the subtle handling techniques of humans so as to depress or excite their maze capabilities. In the case of drug trials, however, even if the experimenter outright says, “This is a sugar pill,” or “This is the real drug,” how could that possibly change a biological result? And yet, it does. This is itself the aptly-named placebo effect—when belief in a treatment’s efficacy actually produces genuine physiological change. There is also a nocebo effect where one’s negative expectation in a treatment will worsen symptoms despite proven efficacy with others.
Simply naming these effects doesn’t make them any less magical. The experimenter’s belief can change the subject’s belief, and the subject’s belief can change their own biology. That’s two layers of belief actively altering reality itself presumably via micro cues which humans notoriously fail at picking up on. Scientists understand this well enough to work around it, but rarely do we stop to question just how profound it really is. Beliefs alter emotions which alter our experienced reality.
In the case of placebo and nocebo effects, the subject’s belief influences their own body’s response. Though shocking, that’s easy enough to accept—it’s their own body, after all. But the experimenter expectancy effect involves one person’s belief impacting another person’s outcome. It suggests a form of influence we can’t truly quantify.
Some might argue these effects don’t necessarily involve emotion, but that claim begins to fall apart under closer scrutiny. After all, what is belief stripped of emotion? Knowledge can be emotionless perhaps, but not a belief. Telling yourself or someone else “you’ve got this” without an emotional intent is not a true belief—it’s merely a sound wave. Belief gains its power through emotion. Unless we're invoking divine intervention or actual magical incantations, statements alone shouldn't change biology—yet when it’s paired with emotional conviction, it does.
Imagine our treadmill runners and their self-talk with robotic voices “you’ve got this… you’ve got this…” versus an emotional, “you’ve got this!”. The power is in the emotion, not the words.
Even if we wish to consider the chemicals that such emotions produce in our bodies as the main source of power, let’s keep it clearly in mind that we cannot simply shoot someone up with dopamine or cortisol and obtain all possible results of every drug that’s ever been tested for weight loss, blood pressure, depression, anxiety, or maze-finding. There is an underlying creative power to the emotional beliefs themselves and not just from chemicals surging through the body. What level of micro cue is going to make the body itself start shedding pounds?
I say all this to impart wonder. Naming the effect does not explain how or why it is so effective and our only explanation seems fairly weak at best. The power of our emotional states alters science itself enough that we have to work around it, and I find that fascinating. To say it yet again, there is power in curating our moods, beliefs, and expectations. If we can keep aligned with a positive can-do attitude with confidence, kindness, and belief, the world shapes itself to meet those expectations—this fact is the cornerstone of good scientific studies.