Archives for January 2011

Since people often confuse the metaphysics of philosophy with the metaphysics of popular culture (things like healing crystals), it is not surprising that I have been asked what I think about the recent paper on ESP. My quick response is that my general view on ESP has not been changed: I see it as a possibility, but very unlikely. However, it seems worthwhile to consider the research in question (which might actually be an elaborate prank).

Bem’s paper discusses nine experiments he has conducted over the past ten years. The purpose of the research was to test student’s ability to sense random events, most famously the random porn images that students were allegedly able to sense before they appeared on the screen of a computer.

Bem seems to be trying to find some sort of backwards causal connection in that the cause is in the future relative to the effect. Since this is not how causation is supposed to work, this is a rather odd sort of thing.

In one experiment Bem had 100 students take a memory test and then had them practice the words they had used in the test. He claims that “The results show that practicing a set of words after the recall test does, in fact, reach back in time to facilitate the recall of those words.”

In his famous porn experiment, the subjects were asked to pick which virtual curtain concealed an image on the computer. Once the choice was made, a program randomly “placed” an image “behind” one curtain or the other. According to Bem, the subjects were able to pick the curtain that “hid” an image 53% of the time when that image was erotic. They were able to pick non-erotic photos only 50% of the time. He claims that “What I showed was that unselected subjects could sense the erotic photos,” Dr. Bem said, “but my guess is that if you use more talented people, who are better at this, they could find any of the photos.”

While I do find ESP fascinating and the paper interesting, there seem to be some serious problems with the experiments and his inferences.

First, the sample sizes are too small. While I am not a statistician by profession, an experiment requires a sample that is off adequate size. Oversimplifying things a bit, the smaller the sample in an experiment, then the less able it is to discern differences between chance and an actual causal connection. While the subjects presumably were able to pick the erotic image 53% of the time, that does not seem to be statistically significant given the small size of the sample-it could be the result of pure chance rather than sensory powers.

Second, it seems rather important to take into account the notion of regressing to the mean. To use an example, if you give 100 students a true/false test and have them guess the answers, then it is very likely that some of them will guess correctly at a rate higher than 50% and thus seem “lucky.” However, if they are tested again and again, the results will eventually approach the expected 50/50 mark. Interestingly enough, attempts to replicate Bem’s results seem to have failed-which seems like reasonably good evidence that the 53% result was a fluke rather than evidence of ESP.

Third, he makes a rather strong leap to a very definite conclusion based on evidence that fails to warrant such a conclusion. To be specific, he seems to be concluding that his experiments serve as proof for backwards causation. However, even if it is assumed that the results are statistically significant, there seem to be many alternative explanations and these need to be properly considered before accepting backwards causation or special senses that can detect the future.

To use something almost as weird as predicting the future, it could be that the students have the power to sense the states of the computer and are thus able to predict what image it will present. After all, one might argue, sharks can detect minute electrical fields. Maybe humans can do something similar. Interestingly, computers do not actually generate true random numbers and hence it could be argued that people have a way of detecting the states of the computer and what result will probably occur. Of course, that is a rather implausible theory. But no more implausible than backwards causation.

Another odd possibility is that the universe is deterministic (as Hobbes and Spinoza argued). As such, knowing what the future will bring is not a mater of the future causing the past, but a matter of knowing what the past will cause in the future. Philosophers, such as Spinoza, have argued that what appears as chance is actually a matter of ignorance because the universe is deterministic (the future follows of necessity from the past). Spinoza and other thinkers even argued that if a person had the right information (or used some sort of rather unexplained rational intuition) then she could know what will be. So, perhaps Spinoza was right and these students are not getting a vision of the future but intuiting what will arise from current events.

Naturally, it is easy enough to go on and on giving alternatives that are more or less as odd or plausible as ESP. Since many of these are consistent with the alleged evidence, the conclusion Bem draws seems to be rather hasty.