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

New memories, or simply lying?

  1. Apr 14, 2004 #1


    User Avatar
    Staff Emeritus
    Gold Member

    New memories or lying?

    I read this article today, and I was left wondering whether participants were actually developing a new memory, or if they would be lying to the researcher about a clearly harmless issue.

    Sometime ago, another study showed that people lie much more often than you would expect. IIRC, the study showed also that women tend to lie to make other people feel good and men tend to lie to make themselves look good. Clearly, that study may interfere with drawing conclusions from this one (were subjects trying to make the interviewer feel good? did males think it was "cool" to have done such a feat in third grade?).
  2. jcsd
  3. Apr 14, 2004 #2

    Les Sleeth

    User Avatar
    Gold Member

    I don't think they were lying. I suspect the combination of the misty past, trusting being told that something had actually happened in their past (which they accepted because they couldn't remember enough to deny it), plus some familiarity with the images they were shown encouraged them to "try" to remember. That "try," in the absense of a real memory, made their imagination join in to help. The vagueness of early memories is close enough in mental strength to the quality of imagined images that it wasn't much of a leap to substitute sn imagined for a real event.

    I remember the McMasters (I think that's the name) trial where children in their childcare business claimed incredible abuse. Yet after finding out how investigators had encouraged the children to "remember," most now think it was all imagined. I don't believe the children were purposely prevaricating.
  4. Apr 19, 2004 #3
    I don't think they were lying, either.

    The dynamic at work in this particular case seems to be that the people were put in a position where their memory was cast into doubt. They weren't asked "Did this event happen?", rather, "Don't you remember this event?" They aren't given room to doubt the event so they suddenly are left wondering is why they don't remember it. This is strong unconscious incentive to "repair" their memory, so to speak. They proceed to spontaneously hypnotize themselves into "recalling" details.
    Last edited: Apr 22, 2004
  5. Apr 19, 2004 #4
    False memory syndrome and hindsight bias

    ...Also known as hindsight bias, a syndrome characteristic of field dependents.
  6. Apr 22, 2004 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    My vote actually goes for lying. They said the subjects were undergraduate students, which means probably the same thing it does at most universities, that the undergrads were "coerced" to "volunteer" for psych experiments as part of their credits for a general psychology class. I know that I would consider all the psych experiments done at my undergrad institution completely invalid. I was one of those subjects and upon volunteering for one survey-based study, compared notes with several of my friends from class who signed up for the same study. We ALL lied on several questions that we knew were critical to the study. It was very quickly obvious to us that there was a "trick" to the survey, that they weren't really testing what they said they were testing, so we all made up answers different from our initial instincts. The last question was something like, "do you think this survey was testing anything other than what we stated in the introduction? If yes, explain." Well, we had no intention of spending any longer there than we absolutely needed to for our class credit of participating, so every single one of us answered "No" because that didn't require explanation and ended the survey. Undergraduates are terrible subjects. If several of us were taking the survey together and all did the same thing without discussing it prior to or during the survey, you can bet a LOT more subjects did the same thing. So, I'd suspect the same thing on this experiment. If the subjects got the impression the session would last longer if they said they didn't remember the experience, they might have just quickly created a fictitious experience to please the experimenter and go home.
  7. Apr 22, 2004 #6
    That's good to know, Moonbear. Someone should do a study about it.
  8. Apr 24, 2004 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    As I was writing that reply, I was thinking the same thing. Someone really should find a way to determine how likely it is their test population is being truthful. When a study is measuring a subjective response, it seems all too easy for the way the questions are asked or the environment in which they are asked to influence the outcome. If the experimenter knows the outcome they are expecting, then they could be subtely biased in how they ask the questions of the subjects. How often are psych experiments done double-blind? The person administering the survey should be unaware of why they are asking the questions. But who checks that the questions are phrased correctly to not elicit a specific response? For example, asking, "Don't you remember this experience?" already posits the experience happened and the question is whether you remember it or not. I don't know how one would test the veracity of answers though. Yes, most surveys have built in questions to test the consistency of answers that is supposed to address this, but that really picks up more on the person who is randomly checking boxes, not the one who is lying and able to be consistent about it. But perhaps someone could create a questionnaire with some very obvious flaws or that appears to have an intent different from the purpose explained at the beginning, then ask at the end if any questions appeared flawed or if the subject thought something other than the explained purpose was being tested. See how truthful they are. That would be good in conjunction with a question about why they chose to participate in the study, such as to receive credit for a class, to receive financial compensation, their friend asked them to go along with them, they are fascinated by psychology and wanted to know what it's about, etc. Then one could start to determine if particular classes of subjects were more likely to be unreliable.

    I was also thinking about the example they were giving...putting Slime in a teacher's desk. This is something the experimenter assumes did not happen because the subjects' parents said it hadn't. Quite frankly, I remember playing with Green Slime as a kid and that stuff wound up EVERYWHERE! I'm sure my parents were not fully aware of all the pranks we played with that stuff, so it's also possible the event actually did happen, or something similar enough to trigger a hazy memory of it. I don't think anyone ever put it IN a teacher's desk (though the teacher did after confiscating it), but on the desk, on the chair, on someone else's chair, all over our own desks as we "sneezed", it's a very plausible story that probably did happen with enough frequency in my generation that it was more likely something the parents just never found out about than that it's a completely false memory.
  9. Apr 25, 2004 #8
    To frame the responses as 'lying' vs 'telling the truth' helps us overlook the grey area (no offence intended, Zoobyshoe) inbetween of confabulation and demand characteristics i.e. participants fill in the gaps in their memories with what seems to be the most likely to be true. This isn't lying, its more a natural desire to do what the experiment requires, and naturally not wanting to look like someone with memory problems. It sounds a bit like the experimenter was fairly leading in the questions and perhaps domineering too, so you get a Milgram-type obedience too.

    To quote from the article: "It's one thing to convince someone they remember a relatively harmless childhood prank, and quite another to dredge up memories of traumatic events". It depends upon motivation. For some people, there may be a strong motivation to explain the problems they are experiencing in their lives, and if alien abduction etc is the best available explanation (i.e. 'Its not your fault you feel this way, its those pesky grays'), that's the one we will tend to subscribe too.

    According to Spanos (1999), false memories are easily induced, especially when an authority figure tells us that memory recovery is possible, that we have repressed material, and administers procedures to retrieve the alleged memories.

    As far as students filling out their forms in a deliberately misleading way - affectionately known as the 'screw you' effect - that's an occupational hazard. Many studies in social psychology will need to employ a certain amount of deception, although current APA and BPS codes of ethics frown upon this and state that any deception is revealed in the debriefing session immediately following completion of the study.

    At the end of the day self-report isn't very reliable most of the time, and memory is a process of reconstruction rather than replication. But we get by, most of the time. Just don't ask me how many units I drink per week.
  10. Apr 25, 2004 #9
    Yes, I think the question can only be properly answered, and the proper terms applied, by finding out to what extent people actually come to believe the confabulations in this case. The choice between lying and new memories is probably too black or white.
  11. Apr 25, 2004 #10
    Good point. And how do we find out how much they believe what they say? I can't think of another method but self-report, which means we are back to square one. Unless... we do something ridiculously expensive like PET scanning the poor participants and seeing if there are differences between the 'Slime-convinced' group and a control group who report on something that they did at school, without prompting. Come to think of it, did the study that started this conversation have a control group? Call me old-fashioned, but I'm always a bit wary of comparing people's behaviour with an imagined standard (which includes being wary of Milgram's studies of obedience).

    I have just checked, and the only difference in the groups was that one wasn't shown a school photo i.e. no comparison group.
  12. Apr 25, 2004 #11


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Okay, I finally got around to actually reading the full article. A few things strike me as fishy. First, at the end, it seems all but three of the subjects knew which story was the false one when finally told one was false and asked if they knew which one. Interesting that there are then exactly 3 quotes from subjects expressing disbelief that the false memory really was false.

    The study seems to have a number of design flaws. Controls, folks, controls! So, either the subjects were shown a class photo or not shown any photo. A group shown a totally unrelated photo should have been included to determine how dependent this "trick" is on having a photo related to the event. Afterall, that the researcher had a class photo adds to the illusion that they do know what they are talking about.

    Another control that was lacking was to have a group of subjects that was told either 1) all true stories, or 2) all false stories. Does the number of true stories influence the likelihood they will believe the third story must be true also? It would also be interesting to know how the subjects would respond if the fictitious story were more outrageous. As I mentioned in my above post, Slime in a teacher's desk is a pretty plausible event and they really could be remembering something related enough that with hazy recollection it would seem more real. If, instead they were asked about the day someone rode a horse through the school, would they be more likely to admit they had no memory of something so outrageous?

    The other thing that stands out is the order of questioning. The students were all questioned about the two real events first, then a week later asked about the fictitious event. This not only adds credibility to the stories by setting them up with two that really did happen, but also gives them time to call their parents in between and confirm those stories. Also, if their parents remembered those stories, then they probably have been told enough times that they would be fresher in memory or the memory was better reinforced (and who's to say how much of those events were partially "embellished" with repeated telling), so again add more credibility. It would have been far better controlled to randomize the order in which the false story was presented. If it was presented as the first story, would the students have been more willing to admit they really just didn't remember? Is it because they didn't remember two events, then called their parents during the week in between and confirmed they really happened, that led them to feel more pressured to recall yet another event? Perhaps they just didn't want to admit they couldn't remember any of these events. Seeing a class photo may have just helped them to be more efficient at fabricating the story by permitting them to recall some of the kids in the class to add names to their stories and make them more realistic.

    It also sounds like the researcher really was quite persistent in persuading them that the event did occur and pressured them to say they did remember it. It doesn't actually sound like it's relevant to trying to get a kid to talk about a past event they don't remember well so much as the innocent person arrested and interrogated until they give a false confession, or a witness embellishing their story upon repeated questioning until it suits the prosecution's case better.
  13. Apr 26, 2004 #12
    I agree. Well said, Moonbear.

    "Truth is the first victim in social psychology"?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: New memories, or simply lying?
  1. Lying (Replies: 23)

  2. Memories (Replies: 2)

  3. No memory (Replies: 27)