Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Experiment shows the weight of confirmation bias

  1. Aug 15, 2005 #1

    SGT

    User Avatar

    A recent experiment shows how people can be fooled and see patterns where there is none.
    The subjects were to play a computer game and score points. Unbeknownst to them, the points were attributed randomly.
    A group scored 33% of the time and the other 66%. After the experiment, the participants were asked to suggest winning strategies for new players. All of them, even those that scored poorly, were able to make suggestions.
    That is what happens in most paranormal experiences. People think they see causality between phenomena, even if there is nothing.
    This is not a proof that paranormal phenomena don't exist, but shows how much we must take care in order to not confound correlation with causality.
     
  2. jcsd
  3. Aug 15, 2005 #2
    Im afraid this statement is not backed up by the results of that experiment.
     
  4. Aug 15, 2005 #3

    SGT

    User Avatar

    Could you elaborate on that?
     
  5. Aug 15, 2005 #4

    Pengwuino

    User Avatar
    Gold Member

    What does making suggestions have to do with paranormal activity?
     
  6. Aug 15, 2005 #5
    "Making suggestions" meant the people in the study were convinced they had percieved a winning strategy. In fact, no such strategy existed. The game was rigged to be random.
    One example of this I can think of is when a person sees what they assume to be a ghost in a house, and then goes looking for someone in the house' history who is dead and may fit the "apparition". If they find someone who seems to fit, it confirms their belief they've seen a ghost.
     
  7. Aug 15, 2005 #6

    Pengwuino

    User Avatar
    Gold Member

    Well how is that paranormal? Just sounds like impressionable humans being fooled with.
     
  8. Aug 15, 2005 #7
    How is what paranormal?
     
  9. Aug 15, 2005 #8

    Pengwuino

    User Avatar
    Gold Member

    grrr, message too short >:(
     
  10. Aug 15, 2005 #9
    I also fail to see a how people developing a winning strategy{either useful or not} has to do with the paranormal. I see nothing to suggest that it does.
     
  11. Aug 15, 2005 #10
    confirmation bias
    Address:http://skepdic.com/confirmbias.html

    You don't see confirmation bias at work in any explorations of the paranormal?
     
    Last edited: Aug 15, 2005
  12. Aug 16, 2005 #11

    Pengwuino

    User Avatar
    Gold Member

    Oh so we're talking about how we percieve possible paranormal activity. I thought you were trying to say confirmation bias is somehow related to actual paranormal activity.
     
  13. Aug 16, 2005 #12
    This is correct.

    I didn't find SGTs transition from one to the other to be nearly as obscure as other people did:

    Perhaps I'm more used to his writing style, though.
     
  14. Aug 16, 2005 #13

    hypnagogue

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I'd agree that confirmation bias probably plays a role in many reports of paranormal phenomena. However, I'm not sure the referenced study is the best means of demonstrating confirmation bias. In a way, the experiment seems rigged (beningly) to creating confirmation bias. It depends on how exactly the subject tasks were conducted and what questions they were asked, but it looks like there might be a priming effect here. If a subject participates in a study like this where he is given a task to do and then asked to derive relavent rules, he might expect that such rules already exist, and thus be more likely to look for and produce them.
     
  15. Aug 16, 2005 #14
    It seems to me, though, that just such priming takes place outside the lab all the time in the form of paranormal lore. People are primed to explore a paranormal explanation for a certain class of strange experiences, by stories that are passed around, movies, and books.
     
  16. Aug 16, 2005 #15

    hypnagogue

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Quite true.
     
  17. Aug 16, 2005 #16

    SGT

    User Avatar

    I am reading your post only now. Zoobieshoe has beaten me in the explanations. He has understood my post and the objective of the experiment.
    The experiment has nothing to do with paranormality. It only intended to show that people are prone to the Post Hoc fallacy: if an event follows some other, people tend to attribute a relationship of cause and effect between them. In the experiment people scored after taking some action, so they concluded that the action caused the scoring.
    In the same way, you have a dream that you remember in the morning. In one day, one week, or a month, something happens that seems to confirm your dream. So, you think you had a prophetic dream.
    I apologize if my original post was not clear enough. English is only my third language, so I don't ever expose my reasoning in a clear manner.
     
  18. Aug 17, 2005 #17

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Yes, and in many cases the interpretations are reasonble considering the experiences claimed, but there is no doubt: Most people don't make good scientists.

    I know of a guy who prays to Thor. I met another who thought he was a chicken. I met another guy who thought every plane seen was a UFO, but these are hardly fair examples of the average person. Heck, I know of one guy who fell for the "I am from Venus" line. :biggrin:
     
  19. Aug 17, 2005 #18

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    However the context for each of the two situations is completely different. Most people don't believe everything that they read or hear, but I think most would be led to particular expectations in a clinical setting.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Experiment shows the weight of confirmation bias
  1. This experience (Replies: 8)

Loading...