Spooky goings on in Psychology!

Given that it is Halloween, it seems only right to discuss some recent psychology experiments relating to potential paranormal phenomenon!

Can ‘psychic’ abilities be demonstrated during controlled experiments?

Can ‘psychics’ sense information others can’t?

Today Merseyside Sceptics Society published the results of a ‘Halloween psychic challenge’. They invited a number of the UK’s top psychics* to attempt to prove their abilities under controlled conditions, although only two psychics accepted the invitation (1, 2). In the test each psychic had to sit in the presence of 5 different female volunteers who were not known to them. These volunteers acted as ‘sitters’ and the psychics had to attempt to perform a ‘reading’ on them, in effect to use their putative psychic powers to obtain information about the sitter’s life and personality. During the reading the psychic  was separated from the sitter by a screen such that the psychic could not actually see the sitter. The psychics were also not allowed to talk to the sitters. These conditions ensured that any information the psychics retrieved was not gathered through processes that could be explained using non-psychic means (e.g. cold reading or semantic inference). The psychics recorded their readings by writing them down.

A copy of the 5 readings made by each psychic (one for each sitter) was given to each sitter and they were asked to rate how well each reading described them, and which reading provided the best description. If the psychic abilities were genuine, then each sitter should rate the reading that was made for them as being most accurate. Of the 10 readings (from the 2 psychics for each of the 5 sitters) only 1 was correctly selected by the sitter as being about them, no more than one would expect by chance. Moreover the average ‘accuracy ratings’ provided by the sitters (for the readings that were actually about them) was low for both psychics (approximately 3.2 out of 10). What of the one reading that a sitter did identify as an accurate description (see 1 for a full transcript of this reading)?  It is noticeable that in this reading the statements (some of which were not accurate) were either very general, or could be inferred from the knowledge that the sitter was a young, adult female (e.g ‘wants children’). The (correct) statement that most impressed the sitter (‘wants to go to South America’) was also pretty general and is probably true of a decent proportion of young woman. It can be safely concluded therefore that even this ‘accurate’ reading happened by chance.

In terms of the experimental design it is important to note that both psychics had, prior to the experiment, agreed to the methodology in the belief that they would be able to demonstrate their psychics powers under such conditions. Likewise both psychics rated their confidence in the readings they gave during the experiment highly, suggesting that they didn’t think that anything which occurred during the experiment might have upset their psychic powers. The study could be criticised for its small sample size, although this is due to many psychics, including some of the better known ones like Derek Acorah and Sally Morgan, apparently refusing to take part. It could therefore be argued that despite the psychics involved in the study failing the test, other ‘better’ psychics might pass. However such an argument remains merely speculative until such psychics agree to take part in controlled studies.

Although these negative results may not be surprising I still think it might be of interest to perform the experiment a different way. The problem with relying on the sitter’s ratings is that they may reflect attitudes of the sitters concerning psychic abilities (although all the sitters were apparently open to the idea of psychic powers being genuine). For example even though the sitters were unaware of which reading was about them, they could theoretically have given a low rating to an accurate reading to ensure that no psychic abilities were demonstrated. A better methodology might be to get each sitter to provide a self description, and then ask the psychic to choose the description that they think fits their reading of the person best. Such a test would also reduce the problems of interpreting the accuracy of the vague, general statements such as ‘wants children’ that psychics are prone to give. Another interesting idea would be to get psychics, along with non-psychics and self-confessed cold readers, to perform both a blind sitting (e.g. using a method similar to that described above) and a sitting where the participant can see and perhaps talk to the sitter. This could provide evidence to suggest whether claimed psychic abilities are really just a manifestation (even unintentionally) of cold-reading. If this were the case one would expect no difference in performance between the three groups in the blind test, but both the cold-readers and the psychics to perform better in the non-blind test (but with no difference between psychics and cold readers in that condition).

Can we see into the future?

The second set of experiments that I wish to discuss are potentially more exciting because there is at least a hint of positive results. Instead of testing the telepathy that psychics claim to possess (i.e. the ability to transfer information without the use of known senses) these studies investigated the phenomenon of ‘retroactive influence’ in a random sample of participants. Retroactive influence is the phenomenon of current performance being influenced by future events. In effect it suggests that people can (at least unconsciously) see into the future!

In a series of 9 well-controlled experiments the Psycholgist Daryl Bem produced results that appear to show that participant’s responses in a variety of tasks were influenced by events that occurred after those responses had been made (3). What is most impressive about these results is that Bem used a succession of different paradigms to produce the same effect, ensuring that the effect was not just due to an artifact in one particular experimental design. In brief this is what his results appear to demonstrate:

  1. Precognitive Detection: Participants had to select one of two positions in which they though an emotive picture would appear on a computer. However the computer randomly decided where to place the picture after the participant has made their selection. Nevertheless participants performance suggested that they were able to predict the upcoming positions of a photo at above chance levels,
  2. Retroactive Priming: In priming, the appearance of one stimulus (the ‘prime’) just before a second stimulus that the participant has to perform a task on, can either improve or worsen reaction time to that task, depending on whether the prime is congruent or incongruent with the second, ‘task’ stimulus. For example the appearance of a positive word prior to a negative image will slow reaction time on a valence classification task for the image (i.e. deciding whether the image is positive or negative) because the valence of the word is incongruent with the valence of the image. Bem’s results suggest that this reaction time effect also occurs when the prime is presented after both the image, and the time when the participant has made their response to it.
  3. Retroactive habituation: People tend to habituate to an image, for example an aversive image that has been seen before is rated as less aversive than one that has not been seen before. Bem demonstrated that this habituation can occur even when the repeated presentation occurs after the rating of the image is made (i.e. given the choice between two images, participants will select as less aversive the image that the computer will later present to them several times).
  4. Retroactive facilitation of recall: When participants had to recall a list of words, they were shown to be better at recalling items that they were later required to perform a separate task on, even though they were unaware of which items on the list they would be re-exposed to.

It is important to note that in all these experiments the selection (by computer) of which items would appear after the initial task, was performed independent of the participant’s response, so the results could not be due to the computer somehow using the participant’s responses to define its choice of which stimuli to present.

These findings caused much controversy and discussion within the psychological research community. Recently three independent attempts to replicate the ‘retroactive facilitation of recall’ effect have failed, producing null results despite using almost exactly the same method as Bem’s original study, and identical software (4). These failures of replication have highlighted problems in psychological research around the concepts of replication and the ‘file-drawer problem’ (5). There isn’t space to do justice to these issues here, suffice to say that the jury is still out on Bem’s findings at least partly because we can’t be sure whether other failed attempts to produce these effects remain unpublished, thus making Bem’s positive results appear more impressive that they might actually be. Another potential problem that is yet to be fully addressed is the issue of experimenter bias. Again this is a complex issue, and it appears to particularly be a problem in research into paranormal phenomenon, because positive results consistently tend to come from researchers who believe in said phenomenon, while negative results consistently come from sceptical researchers (see 6 for a discussion).

Retroactive facilitation of recall is currently the only of Bem’s effects that others have attempted to replicate in an open manner (i.e. by registering the attempt with an independent body before data collection, and by publishing the results after). Until more replication is attempted the question as to whether we can unconsciously see into the future must be considered open to debate. Hopefully these topics will be subject to much research in the future allowing us to find out whether these effects are real, or just the consequence of some other factor. It is worth mentioning at this point another paradigm that sometimes produces positive results regarding paranormal abilities. In experiments using a Ganzfeld Field (where participant’s auditory and visual systems are flooded with white noise and uniform light respectively) there is some evidence that those experiencing such stimulus are able to ‘receive’ information from someone sitting in a separate room (see 7 for a review). This appears therefore to be a potential demonstration of telepathy, although the effect is open to the same issues of replication and experimenter bias that surround Bem’s findings. Even ignoring these uncertainties, it should be noted that in these Ganzfeld Field experiments, and in Bem’s study, the size of the effects are very modest. For example in Bem’s precognitive detection paradigm, participants overall performance was at 53% as compared to chance level performance of 50%, while in the Ganzfeld experiments performance (choosing which one of four stimuli were being ‘transmitted’) is at around 32% against a chance performance of 25%. While these differences are found to be statistically significant (in some studies) because of the large number of participants or trials used, they don’t exactly represent impressive performance! Therefore even if such paranormal phenomenon were to be eventually proven as genuine, this wouldn’t mean that the sort of mind reading abilities claimed by psychics are actually possible!

 

*note that in this article the term ‘psychics’ is used merely as a label to define people who claim to have psychic powers, its use does not represent acceptance that such powers actually exist.

References
1) http://www.guardian.co.uk/science/2012/oct/31/halloween-challenge-psychics-scientific-trial
2) http://www.merseysideskeptics.org.uk/
3) Bem, D. J. (2011). Feeling the Future: Experimental Evidence for Anomalous Retroactive Influences on Cognition and Affect. Journal of Personality and Social Psychology, 100(3), 407-425. Link
4) Ritchie, S. J., Wiseman, R., & French, C. C. (2012). Failing the Future: Three Unsuccessful Attempts to Replicate Bem’s ‘Retroactive Facilitation of Recall’ Effect. Plos One, 7(3). Link
5) Ritchie, S. J., Wiseman, R., & French, C. C. (2012). Replication, replication, replication. Psychologist, 25(5), 346-348. Link
6) Schlitz, M., Wiseman, R., Watt, C., & Radin, D. (2006). Of two minds: Skeptic-proponent collaboration within parapsychology. British Journal of Psychology, 97, 313-322. Link
7) Wackermann, J., Putz, P. & Allefeld, C. (2008) Ganzfeld-induced hallucinatory experience, its phenomenology and cerebral electrophysiology. Cortex 44, 1364-1378 Link

Image from ‘Seance on a wet afternoon’ (1964) Dir: Bryan Forbes, Distribution: Rank Organisation, Studio: Allied Film Makers.

Rob Hoskin

Received a PhD from the Neuroscience Department of Sheffield University. Views expressed in blog posts do not necessarily represent the views of the Science Brainwaves organisation. https://twitter.com/Hoskin_R

2 thoughts to “Spooky goings on in Psychology!”

  1. Note: At least 2 of the famous psychics whose names have been used for publicity in this Halloween challenge were never formally invited, and only found out about the challenge through other people talking about it.

    I would have preferred to have seen these famous psychics formally invited, instead of just having their names used in such a way to suggest they were scared to be tested, when this was not the case.

    It lessens the reputation of the society when they pull stunts like this purely in the name of getting publicity.

    The test itself was as expected, but the inclusion of Patricia Putt is very interesting as she has already failed tests similar to this in the past.

    This test in itself has only achieved to give MSS publicity, which itself was minimal anyway.
    More attention was given to Sally Morgan appearing on Come Dine With Me.

    I would prefer to see such blatant publicity stunts put aside and replaced with genuine scientific study and respectable credible testing.

    Skepticism doesn’t need such stunts as it just devalues the hard work of others who unfairly get lumped in just because of the label “skeptics”

Leave a Reply

Your email address will not be published. Required fields are marked *