Precognitive Porn

New York Times Article 

Dr. Daryl J. Bem, a well-known professor of psychology at Cornell, has managed to prove that people have ESP, at least in regards to where nude pictures are stashed.  This of course begs the question of why my parents had so much trouble finding my “stash” while I was a teenager, but that is beside the point.  Actually I am a bit skeptical of the findings in this study.  Big surprise I know, but the only figures I can find are listed as part of the New York Times article which only showed 53 percent accuracy, barely enough to be
statically significant since the people in the study
were picking from one of two hidden pictures.

            Dr. Bern seemed unfazed by this and according to the New York Times article says, “What I showed was that unselected subjects could sense the erotic photos, but my guess is that if you use more talented people, who are better at this, they could find any of the photos.”  I find this a curious statement to make given the underwhelming results in his study and wonder where his certainty comes from.  Especially since the results have been unrepeatable in at least three studies since his initial one completed.

Of course the interesting part of this story is not so much the study as it is the resulting response to it by scientists, and of course the response to that response.  Many scientists were quite offended that such a study even made it into the journal in the first place since there is no known mechanism for ESP and the issue has already been sufficiently debunked.  

I actually might surprise some of you by saying I don't really have a problem with the journal publishing the study.  I am open to correction on this if I am wrong, but it does not seem to me that it is the job of scientific journals to decide what qualifies as a study with correct findings, that is the job of the peer review process that begins after the study is published.  The job of journal is generally just to make sure that the research was done properly, which, according to the editors, it was.  It is, perhaps, strange that a well known journal would publish a study on an idea as thoroughly discredited as ESP, but being published in a journal does not make the findings true.

The curious thing I have found is some of the rather anti-science responses to criticism of the article, even from people who should know better.  One I noticed in particular was an article from Dr. Arri Eisen, a biologist who wrote an article about this for www.religiondispatches.org.  Dr. Eisen writes about the decline effect in scientific studies.  The decline effect is essentially an observation that many published studies start out with positive effects that are observed which cannot be found when the experiment is repeated.  Pseudo-scientists have often excused the lack of repeatability in their findings by claiming certain things like ESP have a decline in efficacy when tested repeatedly.  However, Dr. Eisen points out that the decline effect is observed all the time in science, and then goes on to argue as if this shows a basic weakness in the scientific method.  Then he builds from that an argument that many scientists reject claims such as ESP or the effects of Prayer based on a bias instead of a failure to produce evidence.  After all, they argue, if we can observe the decline effect in both real science and pseudo-science then we only reject the pseudo-science because of a bias not a lack of evidence.

However, in my opinion, Dr. Eisen seems to miss some basic points, which seems odd to me given his background in science.  First, the decline effect is not a weakness of the scientific method, but rather a strength.  It is good science at work.  Consider this, a scientist observes a fact, then proceeds to come with a hypothesis to explain that fact.  The scientist will then proceed to design an experiment to test his hypothesis.  There are only two possibilities, either the experiment falsifies the hypothesis or it does not.  If it falsifies it, the scientist is not likely to publish anything, there is no point is having the peer review process examine a hypothesis that has already been proven false.  If it does not fail then the scientist may publish his findings so that others in his field may examine his findings.  This by no means suggests at such an early stage that the hypothesis is correct.  Indeed the whole point of peer review is so that scientists may further attempt to create experiments that may falsify it.  The principal of empirical falsification and the process of peer review are pretty much set up to guarantee that once a study is published it can only decline; it is only those that do not decline which become accepted scientific theories.  

Pseudo-science claims are rejected for much the same reason that many hypothesis' in real science are rejected, lack of repeatability.  What makes the proponents of a claim pseudo-scientists is the continued insistence by the proponents that the hypothesis is true despite repeated failures to demonstrate this using the scientific method.  

I am not saying that scientists never have any bias, or that their biases never cause failures in the scientific process.  However, this is not a failure of the scientific method as Dr. Eisen seems to think, but a failure of people to properly apply the scientific method, and any revisions therein should be directed at making the process better at weeding out said biases.  Moreover, it does not seem to me that the bias preventing a proper examination of ESP claims is coming from scientists who refuse to believe, but from ESP proponents who refuse to admit that there is simply no significant evidence to suggest ESP works.