Ep 174: The First Replication of Daryl Bem’s Research on Psychic Phenomenon

MichaelCognition, Intelligence and Language, Critical Thinking, Uncategorized14 Comments

Have psychologists recently found evidence for the existence of psychic ability? Last year, well-known psychologist Daryl Bem published an article called Feeling the Future in which he describes a number of studies, all of which provided support for a kind of phi phenomenon he calls “retroactive influence”. The research appeared in the Journal of Personality and Social Psychology. The article caused a storm of controversy and calls for changes to how research and the peer review process is conducted. I covered some of those recommendations in video episode 165: Psychological Research Under Fire. In this episode I interview the lead author, Stuart Ritchie, of the first published replication of one part of Bem’s work. Listen as Stuart describes what he did and what he found on this very controversial topic.

Resources on the Controversy of Daryl Bem’s Psi Phenomenon Research

Below is a concept map with loads of resources on this topic. You can also click on this link to see this map on Bem’s research.

Concept Map with Details of the Replication of Bem's Feeling the Future study

Click the image above to open the map in a new window

  • Retraction Watch
  • Psi Your Mind?
  • More Bem Fallout
  • False Positive Psychoogy
  • Richard Wiseman on Bem

  • Joseph P. Simmons, Leif D. Nelson, and Uri Simonsohn (2011).
    False-Positive Psychology: Undisclosed Flexibility in Data Collection
    and Analysis Allows Presenting Anything as Significant
    Psychological Science October 2011
Related Posts Plugin for WordPress, Blogger...

14 Comments on “Ep 174: The First Replication of Daryl Bem’s Research on Psychic Phenomenon”

  1. Richard: thanks for this link. I’m definitely going to give this a read and even if Dr. Bem doesn’t consent to do an interview with me (I’m sure he’s had more than enough requests from others) I may do a follow-up episode on this topic myself. Your comment is not the first time I’ve heard that perhaps the Ritchie et. al research needs itself to be given a second look. Appreciate the comment and the link very much.

  2. Michael, as you have not been able to secure Bem for another interview, I would refer interested parties to a podcast where he responds to the accusations made by Ritchie, Wiseman and French.


    Bem’s purported psi findings may make us uncomfortable but the methods employed by the debunking brigade left me shocked and dismayed. It is clear that Ritchie, Wiseman and French have been deceitful to the point of blatant dishonesty. Disgraceful behavior from those who would be seen as representing the broader skeptical Scientific community.

  3. I’m trying to keep an open mind about this. Bem’s a respected researcher. But there’s no denying that his studies had design flaws. It’s true I’m skeptical, but my reaction to his studies was not predetermined. I’ll email Dr. Bem to see if he wants to come on the show to defend his results.

  4. oh dear, oh dear. The first half of the podcast discusses design disasters, methodological flaws and statistical errors in Bem’s work – the implication being that the psi effect he claims to have measured is, *giggle giggle* a most dreadful ‘error’. Then we get to the replication – where, surprise surprise – no psi effect was found. But if, perchance, a psi effect HAD have been demonstrated in the replication? Of course it would have been dismissed due to the design flaws! If ANYTHING was predetermined, it was that Bem’s work was going to be disproved by these ‘skeptics’. And really? Supporting the idea of alcohol fueled discussions, led by ‘skeptics’, at Pubs ? Embarrassingly obvious bias revealed here by Michael, Ritchie et al. Quality Science sadly lacking in this episode, I’m afraid. Perhaps an episode asking Dr. Bem to explain himself would be more worthwhile?

  5. @casper,

    If you read the study, Bem used two types of RNGs to rule out the possibility that the subjects are influencing the computers on a mind-over-matter sort of way. He used true RNG which could, in theory be affected by psychokinesis and pseudo RNGs which can not be affected because once the initial seed is set the sequence that follows is determined. Both RNGs gave the same results.

  6. Hello Michael!
    Thank you for an interesting episode. (Please excuse my English, I’m from Estonia).

    Do you have any plans to make a “Part 2” on this topic? Because so many questions arise based on how the study was conducted, and how it could be fine-tuned in future experiments.

    First, as it has been already mentioned, could be a flaw in RNG.
    Second, the causality. Why would we think that a human influences a computer in the future? What if we reverse the stages and “influence” the past?
    Step one: A computer picks the words (not shown to the human subject)
    Step two: A human subject tries to memorize the big list.
    Step three: A human recollects the words.

    Because the choice has been made already (by RNG), will it be any corellation?

    Thirdly, do the results differ between subjects in trial? Age, gender, number of tries, their awareness (blind test or not), stressed or calm? Quite interesting to know more about it.

    And last but not least. Today, crowdsourcing is the bottom line. Does it sound reasonable if they put up a webpage with online test, and anybody can participate? Or simply make their program downloadable, with a detailed note on how to pass the test at home. I’m sure that anyone would like to try it and would not fake the results.
    They will get much more precise statistics then.

    Anyway, thank you for reading and good luck!

  7. Hi, Michael–

    I listened to your podcast with interest. I can appreciate your concern when psychologists delve into the paranormal without bending over backwards, methodologically. The science of psychology has come so far, but sadly, there are still some that liken psychology to a “religion” that you “believe in” rather than a science that is rigorously evolving using the strictest and most objective, scientifically-sound standards. Strikingly, I started reading a book a few days ago entitled Fringe-ology (by Steve Volk), a book that attempts to discuss paranormal activity through an objective, journalistic eye. Volk has a chapter devoted to the issue of psi phenomenon, and he discusses the research methodology used by parapsychologists in their attempts to learn more about this subject. (So far, the author appears to try very hard to be objective about both sides of the argument– he has an open mind, but he’s cautious, and he appears fair in his analysis.) Haven’t gotten far enough in the chapter to comment on the research methodology he discusses, but according to the author, typically, the parapsychologists who study psi phenomenon adhere to even stricter standards than most psychologists adhere to, because the bar is so much higher for them, given the subject matter. Will revisit this blog after I’ve had a chance to check out the studies that he discusses in this chapter.


  8. Christina: thank you for those tips! I really appreciate my audience from the UK so I went back into the audio and edited those pronunciations to the correct ones you provided.

    Again thanks!


  9. Michael some tips on the correct pronunciation of the British places mentioned in this podcast! Edinburgh in Scotland is pronounced
    Ed – in – burr – uh and Hertfordshire which is county outside London is pronounced Hart-ferd-sheer.

  10. Toby, Bem used a combination of “true” and “pseudo” RNGs to try to rule out psychokinesis as a counterexplanation. Pseudo RNGs are considered deterministic because once the “seed” is selected, the other numbers can not be changed. Bem discussed this quite early in the article. Not sure I’m buying the argument, but I think studies that are designed to test precognition rather than pk tend to have greater effect size.

    Michael, there is a review of psychokinesis in Psychological Bulletin (I think it was around 2006). The effect size was extremely small though, much smaller than those reported in studies that are designed to investigate precognition, clairvoyance, and telepathy. If you search for ganzfeld and psychological bulletin, you’ll find a number of reviews.

  11. des: I haven’t seen any research to support psychokinesis, but if you have anything I can read I’d be happy to take a look. I’m also not familiar with the Global Consciousness Project but again I’m happy to take a look at the report. I’ll do a Google search on this but if you have any specific links please pass them along.

    I’m open-minded about the topic, but I just haven’t seen any solid supportive research on it. If you have some – please share.

  12. You’re making what is likely a type II error here Michael. These replication are irrelevant when compared to a 10 year study and in fact they were not true replication. As for random number generators, they would be affected through psychokinesis. Not saying they were, but that would be the route. Of course you think that’s impossible, but then how do you explain the very sound Global Consciousness Project finding published recently?

  13. If anything – I’d think it more likely participants were influencing the random number generator than getting a glimpse of the future.

Leave a Reply

Your email address will not be published.