Extrasensory perception (ESP) fails the test
Noah Gray
Monday, 14 January 2008 20:13 UTC
You probably didn’t need ESP to see this one coming: using state-of-the-art technology, neural evidence for the existence of telepathy, clairvoyance or precognition is lacking. Harvard researchers used fMRI in an attempt to detect any changes in BOLD signals when participants looked at different images, including ones that were being “sent” to them by a relative, friend, or partner in another room. The participant in the scanner should have recognized some of the images as “familiar”, if they had previously “received” the image from the participant in the other room. No changes in the fMRI response to any of the images meant no ESP.
Although interesting, there are plenty of problems with this study, the most glaring being that participants were randomly selected as opposed to using individuals with established paranormal talents. But independent of that, and despite the nice design of the experiment, the absence of evidence is not the same as the evidence of absence, as stated by Daryl Bem, a psychology professor at Cornell University in an article covering this study.
*Discussion Points
1. Should scientists waste time and resources testing the claims of pseudoscience?
2. Is fMRI the best technique to determine the existence of ESP?
3. How does this evidence compare to the strength of evidence in favor of ESP?*
-
Replies
Jump to resultsResults
-
You are confused. “Evidence of absence” is simply an impossible scientific criteria.
What I think you mean to say is that the study failed to verify an hypothesis for which there was no empirical evidence in the first place.
As to the relevance of the methods used in this study the hypothesis (which I have not seen clearly stated) seems to me to be little more than digging around in a dark room in Mexico with a tooth pick and the expectation that you will find an elephant.
-
Let me correct the wording of part of my prior response (and edit function would be useful). It should read:
As to the relevance of the methods used in this study of the hypothesis, that I have not seen clearly stated, they seem to me to be little more than digging around in a dark room in Mexico with a tooth pick and the expectation that you will find an elephant.
-
What if the hypothesis was: “ESP does not exist.” Then, this study has provided evidence in support of this hypothesis, or rather, evidence for the absence of ESP, not an absence of evidence that ESP does not exist.
From a different perspective, if the same experiments were conducted on individuals claiming to be clairvoyant, with the hypothesis that they actually were, then this study would again have provided evidence of absence of ESP, if the subjects performed at chance level on the task.
I think that is enough semantics discussion for one thread, and I don’t intend to return to this issue. I hope that we can discuss something else regarding this study.
To start off in that direction, here goes: As I mentioned in the post, I think that the study has a clever design, and based on everything we have available to us today for use on human subjects, these are about the most relevant techniques around. Are there alternatives that you would suggest that could potentially provide more meaningful data?
-
“ESP does not exist” is no kind of hypothesis. It assumes a priori that there is some phenomena to examine. It is an assertion no better than the assertion “ESP exists.” Both statements are without meaning from a scientific point of view.
The study itself is vacuous. It experimented with no phenomena with which the study was supposedly concerned and observed no evidence worth reporting.
-
I think getting to the nitty-gritty of the epistemology here is actually quite important, because it challenges us to consider whether these phenomena are in fact testable at all, and if they are, how.
Steven highlights the problem of trying to address these sorts of phenomena experimentally. The formation of a sturdy, testable hypothesis is, in many cases, simply not practical.
The inductive “absence of evidence” case is much weaker than it is for, say, measuring the effect of a drug on a physiological response. If hypertension patients treated with the test drug have BPs no different than placebo, we would conclude that the drug is ineffective. There are, of course, other possibilities (by some freak occurrence those patients just happen to be unresponsive; or the drug batch was dodgy; or the investigators got the tubes mixed up), but adequate controls can be put in place to really narrow the limits of alternative explanations until the most like explanation for the negative data is that the drug simply isn’t effective. Yes, it’s inductive, but induction can be strong enough to be practical, if not philosophically ideal (I think Steven mentioned this).
With ESP, such controls are not feasible, and so the inductive force of the “absence of evidence” is considerably weaker, and not at all forceful enough to allow a conclusion of “evidence of absence”.
Noah said,
“From a different perspective, if the same experiments were conducted on individuals claiming to be clairvoyant, with the hypothesis that they actually were, then this study would again have provided evidence of absence of ESP, if the subjects performed at chance level on the task.”This hypothesis is problematic because it rests on the assumption that the self-proclaimed clairvoyants are telling the truth. If the experiment fails to show evidence that they are clairvoyant, the possibility that they are lying immediately renders the initial hypothesis invalid. This is a paradox that plagues sociobiology and psychology (I think that there’s a name for it, but I forget).
Results
-
