Time to rewrite your autobiography?

Remember that time you went on a hot-air balloon ride at a fair when you were little?? You went with your dad – here's a photo of him. You were worried at first but he reassured you…it made you a much more trusting individual. Quite a momentous day.Except it never happened. As strange as it might seem, plenty of psychological research shows that it is possible to plant these kinds of false autobiographical memories, using various techniques.

Suggestive interviews
One method for creating wholly false memories involves multiple highly suggestive interviews. In these studies, adults (usually undergraduates but sometimes members of the public) are asked to read descriptions of events that they experienced as children. However, unbeknownst to them, one event is fabricated by the experimenter.

For example, Loftus and Pickrell's (1995) participants read a fictional story about getting lost in a shopping mall when they were five, and being rescued by an elderly person who reunited them with their family. (The false events in suggestive interview studies are always moderately significant personal experiences that participants' family members can verify never happened.) Over two or three sessions, the participants are encouraged to recall their unremembered events, sometimes using techniques that purport to aid memory, such as mentally recreating the physical context of the event.

Once the study is over, independent judges are trained to code transcripts of participants' memory reports and to determine who reported false memories of the target event. Some studies have differentiated, in various ways, between participants who reported 'partial' and 'complete' false memories. The gist of the distinction is that participants who provide evidence that they genuinely believe they are remembering the false event ('I remember feeling scared when I couldn't find my mum') and details beyond those that are provided in the false description, are classified as having complete false memories. Participants who merely accept that the event occurred or speculate about it ('Well, it might have happened in the mall down the road from our house') are classified as having partial false memories (See also the recent debate on distinguishing between false memories and false beliefs: e.g. Scoboria et al., 2004; Wade et al., 2007).

So how many participants report partial or compete false memories? We calculated that one third of the 560 participants in 10 published studies were scored as having partial or complete false memories. And in the studies that differentiated between partial and complete, 17 per cent were scored as partial and 17 per cent as complete. The highest rate of complete false memories in an individual study was 26 per cent (Porter et al., 1999), but recent studies have modified the suggestive interview procedure and revealed even more dramatic rates of false recall (for a brief review see Wade & Garry, 2005). We return to this point later.

The content and qualities of these false memories vary widely between participants and studies. Sometimes the false memories are emotional (Porter et al., 1999), sometimes they are held with great confidence (Hyman & Billings, 1998), sometimes they are extremely detailed (though on average they tend to be somewhat less detailed than comparable true memories; see Heaps & Nash, 2001). That is, false memories tend to vary in the same sorts of ways that true memories do (Heaps & Nash, 2001; Laney & Loftus, in press). Considered as a whole, the suggestive interview research shows that false-memory creation is a robust phenomenon.

How is it then, that simply thinking about an event can lead people to create false memories? There is some debate, with two main theories.

According to the 'fuzzy trace' account (see Brainerd & Reyna, 2005), people create two kinds of memory traces: verbatim traces that record surface-level sensory details (smell, colour, sound) and gist traces that record more abstract details (meanings, interpretations, elaborations). Verbatim traces are thought to be susceptible to interference and to fragment quickly over time, whereas gist traces are more robust. According to fuzzy trace proponents, false memories arise when people rely on gist memories alone or when verbatim memories from one source are mistaken as memories from another source (see Brainerd & Reyna, 2005 for a detailed description).

Alternatively, the source monitoring framework (SMF) argues that the process ultimately comes down to an error in source monitoring – sometimes people mistake imagined details for genuine memories (Johnson, 2006; Johnson et al., 1993). According to the SMF, people do not know the source of a memory outright. Instead, we use a variety of cues to infer the source of every thought, image and memory we experience. This process works well, most of the time, because there are noticeable systematic differences between externally perceived (real) memories and internally generated (imagined) memories. Real memories, on average, contain more sensory details (colours, sounds, smell) and conceptual details (spatial, temporal) than imagined events. Real memories also tend to be more coherent, logical and consistent. Together these differences allow us to make accurate judgments about the source of our memories. But when an image or thought has the hallmarks of a real memory, we sometimes confuse imagined events for genuine experiences.

In the suggestive interview technique, the act of imagining a counterfactual childhood event presumably causes participants to generate perceptually rich and coherent false details about the suggested event. Over time, these images and thoughts become increasingly like real memories, and participants become confused about what is imagined and what is real. Support for this premise comes from studies that show that the more times people imagine performing an action, the more likely they are to incorrectly claim that they actually did perform it (Goff & Roediger, 1998).

Suggestive interviews and props
Recently, one team of false-memory researchers extended the suggestive interview technique to include props, specifically childhood photographs (Lindsay et al., 2004). Trauma-oriented psychotherapists sometimes encourage clients to peruse photographs to help cue long-forgotten memories of upsetting experiences. Lindsay and colleagues predicted that personal photographs, when combined with suggestive interview techniques, might foster the creation of false memories. The researchers were right. They asked some adults to try to remember three childhood events; as usual, one event was false, describing how the participant was reprimanded at school for sneaking Slime (the gooey green toy) into their teacher's desk. All of the participants heard descriptions of the Slime event, but half also received a class-group photo from the relevant school year. Of those subjects who did not receive the class photo, 23 per cent formed false memories of the Slime event. But with the aid of a photograph, the false-memory rate soared to 65 per cent.

The 'Slime event' differs dramatically from the types of events clients recall in therapy. Yet these findings warrant concern about the riskiness of encouraging people to review photographs during attempts to cue suspected but forgotten memories of childhood trauma. Indeed, there are good reasons to believe that the mechanisms responsible for false memories in the laboratory may also contribute to other types of false memories (Wade et al., 2007).

A series of related studies have used digitally altered photographs to implant false childhood memories (Garry & Wade, 2005; Wade et al., 2002). They showed some adults three real childhood photos, and one fake photo depicting the participant taking a hot-air balloon ride as a child (Figure 1). Hot-air ballooning was chosen because it is a moderately significant personal experience that typically requires parental consent; thus, family members could verify it never happened. Participants considered their
set of childhood photos and tried to remember the depicted events in three sessions over a maximum of two weeks, and by the end of the study, 50 per cent remembered something about the hot-air balloon ride. The false-memory reports were often rich with detail. One participant said:

I'm still pretty certain it occurred when I was in Year 6 at the local school – basically for $10 or something you could go up in a hot-air balloon and go up about 20-odd metres. It would have been a Saturday and I think we went with, yeah, parents and, no it wasn't, not my grandmother. I'm not certain who any of the other people are there. I'm pretty certain that Mum is down on the ground taking a photo.

In addition, participants were surprised during the debriefing to learn that one of their photos was a fake.

These studies show that a doctored photograph, without any accompanying verbal description, is enough to lead people to report entirely false memories of significant childhood events (Garry & Gerrie, 2005 provide a comprehensive review of how photographs affect memory). But a new study by Nash and Wade (2007) shows how powerful false evidence can really be.

Significant advances in digital technology mean that fake video evidence can be highly compelling and almost impossible to refute. Indeed, in the last five years, computer scientists having begun developing complex mathematical and computational techniques (called digital forensics) to detect alterations in digital media (Dreifus, 2007). This led Nash and Wade (2007) to examine whether tampered video evidence could lead people to believe they committed an act they never did: cheat on a psychological experiment.

The experimenters filmed participants carrying out a computerised gambling task. The task required participants to answer a series of general knowledge multiple-choice questions and to bet (fake) money on their answers every time they responded to a question. When they answered a question correctly a green tick appeared on the monitor and they were instructed to collect money from the 'bank', whereas when they answered a question incorrectly a red cross appeared on the monitoring and they were instructed to return money to the bank. Two hours later participants returned to the lab and the experimenter accused them of taking money once (Experiment 1) or three times (Experiment 2) in the first session when they should have returned it. In addition, the participants were told that their data could no longer be used and that they would have to forfeit payment for taking part in the study. Half of the participants were told that incriminating video evidence existed, and half were exposed to a doctored video that depicted them committing the act (a green tick was digitally replaced with a red cross so that the participant was apparently taking money from the bank when an error message appeared: see Figure 2).

In two studies, almost all of participants who viewed the fake video falsely confessed to the act. Over 70 per cent told a research confederate that they had made the mistake, indicating that they genuinely believed the false act occurred.

This is the first study to demonstrate the dangers of modern digital-manipulation technology when encouraging people to remember recent autobiographical experiences. This study also mimics real-world situations that may lead to false memories. For instance, police interrogators can and do lie about the existence of evidence during interrogations and this oft-used tactic is legal in the United States (Kassin & Gudjonsson, 2004).

False feedback
Another, even simpler procedure has been used recently to implant false memories. In the false feedback paradigm, participants are first given a series of questionnaires on a particular topic, such as childhood experiences with food (Bernstein et al., 2005; Laney, Morris et al., in press). Participants are then told that their data will be analysed by a sophisticated computer program, which will produce a profile of their results. On a subsequent visit to the laboratory, participants are given 'feedback profiles' which say that, for example, they once got sick after eating hard-boiled eggs as a child. But these profiles are not actually created by a sophisticated computer program. Instead, there is just one version of the profile per experimental group. After they read their profiles, participants complete a second series of questionnaires. The data typically show that this very simple manipulation – 'the computer says you had this experience' – can produce false memories in approximately half of manipulated participants. In addition, these participants also demonstrate false-memory consequences: to continue with the egg example, participants often claim to like hard-boiled eggs less, and to want to eat them less in the future.

In a more complicated twist on this false-feedback theme, another group of researchers gave participants false feedback that was designed to mimic information that they might receive in certain kinds of therapy. Mazzoni et al. (1999) assessed participants' confidence that they had been bullied as children, along with other childhood events, using a paper-and-pencil questionnaire. Some participants who were initially confident that they had not been bullied were then asked to participate in an additional study (actually the second phase of the same study) in which a recent dream was interpreted for them by a clinical psychologist. Participants reported a wide range of dreams to the study's two clinical psychologists, but the content of the dreams was irrelevant to what happened next. All participants were told (regardless of the content of their reported dreams) that their dreams were actually evidence of a repressed memory of being bullied at a young age. After the dream interpretation phase, participants' confidence that they had been bullied as children was again assessed. Compared to control participants (whose dreams were not interpreted), experimental participants became more confident that they had been bullied, and those whose confidence increased were also likely to produce concrete (false) memory reports of being bullied.

Social influence
We have shown how direct suggestive methods can play havoc with memory. But a less direct method, and a relatively new area of research in the false autobiographical memory domain, is the role of social influence in creating false memories.

Discussion is an integral part of our lives. We reminisce with family members about shared experiences and attempt to agree on what 'really' happened (Ross, 1997). When we witness surprising or spectacular events we talk to other witnesses and consider whether our versions of events match up. These discussions in the name of accuracy testing can have serious implications for eyewitness memory (see Gabbert et al., 2003).

But can discussion like this influence autobiographical memory? French et al. (2006) showed that it can. The experimenters used an internet-based version of the suggestive interview technique to examine how discussion influenced both genuine and false memories for childhood events. Pairs of adult siblings attempted to recall some real childhood events plus a fake childhood event, which was a hot-air balloon ride. At first, the siblings recalled the events independently, but then they discussed their memories with each other (using the online learning environment Blackboard) before reporting, independently once again, what they could remember. The results showed that participants incorporated snippets from each other's real and false memories into their own recollections; and, although almost 25 per cent of participants reported false memories, the false-memory rate dropped dramatically after participants discussed their memories. One explanation for this drop could be that sceptical participants – those who doubted the authenticity of the false event – pressed their siblings to carefully consider the true source of their memories. As a result, participants became more conservative when accepting and reporting images and thoughts as genuine experiences.

Discussion does not only affect autobiographical memories for middle- and late-childhood experiences, it also affects memories for early childhood. Peterson et al. (in press) asked adult participants to describe their earliest memories; but beforehand, half were exposed to confederates who described their own earliest memories, including their second birthday or first few steps. The participants who were exposed to confederates' very early memories reported memories that were, on average, one year younger than the memories reported by the control participants. Together, the French et al. (2006)
and Peterson et al. results show that simply listening to others share their
own memories might be enough to transform autobiographical memory.

Conclusions
If memory is the diary we carry about, then it is likely to include truths, half-truths, gaps and falsities. We have shown that memory can be inexact and unreliable – various suggestive techniques can lead to wholly false memories for personal experiences. We hasten to add, however, that substantial errors in memory, and completely false memories, also occur in real life, far outside of the laboratory (see Schacter, 2001). In fact, it is very likely that everyone reading this article has a collection of false autobiographical memories that they are completely unaware of.

The research on false memories has raised some significant questions. For instance, how do participants feel about being deceived in false-memory studies? Together, we have conducted over 25 false-memory studies with over 3200 participants, and between us only one participant has expressed significant concern about the use of deception. We carefully debrief our participants at the end of their participation, both to ensure that the deception used in our studies has minimal long-term effects and to allow participation in our studies to be a learning process for participants. The vast majority of participants in false-memory experiments enjoy the experience and learning about the fallible nature of human memory.

How do we know that participants genuinely believe the false events happened? There are several reasons why, we believe, participants are not confabulating details to please the experimenters. First, we have already mentioned that participants are often surprised to learn, during debriefing, that their (newly acquired) memories are likely to be false. Second, in research that parallels that on detecting deception, there is evidence that third parties are quite poor at distinguishing between true and false-memory reports (Laney, 2006).

Third, a new study has been specifically designed to both assess the role of social demand in false-memory research and minimise its effects (Laney, Kaasa et al., in press). Participants in false-memory studies are always given some kind of cover story designed to hide the true nature of the study. But it is possible that participants may be able to see past these cover stories and determine the true nature of the study. If they do so, then they may yield to the demand of the situation and produce evidence of false memories in the absence of any real change in memory. In this study, the participants were told that the researchers were studying 'food preferences and personality', when they were really trying to implant false memories of loving asparagus the first time it was tried. But beyond that cover story, these participants were also led to believe that the study had a different focus, the American obesity epidemic. This other, implied purpose has been dubbed the 'red herring'. Although almost half of the participants in the study bought into the 'red herring' explanation – and just 8 per cent figured out that the study was attempting to implant false memories – the study still produced false memories in 40 per cent of manipulated participants, a proportion equivalent to previous similar studies. In addition, the few participants who did figure out that they were participating in a study of false memory were no more likely than other participants to produce false memories. The researchers concluded that social demand could not be the primary explanation for the observed false memories.

Another question is whether false memories have behavioural repercussions. Do they cause us to change our behaviours outside of the lab? In addition to the food preference study mentioned earlier (Bernstein et al., 2005), other research has shown that getting people to believe that they had a specific experience with the character Pluto at Disneyland can have behavioural consequences, including reduced willingness to buy a Pluto souvenir (Berkowitz et al., in press).
Finally, one of the ultimate purposes
of conducting false-memory studies is to determine whether there is some characteristic that differentiates real from false memories. If such a characteristic could be found, then psychologists might be able to look at a particular memory and determine whether that memory is true or false. Thus far, there is no such characteristic (but see Okado & Stark, 2005, for some promising future directions). Like true memories, false memories can be held with great confidence, can be detailed, can be vivid, can have behavioural consequences, and can even be emotionally rich. But the fact that a particular memory is confidently held, detailed, vivid, consequential or emotional, or even all of these, cannot guarantee that the memory is real.

Resources

Loftus, E.F. (2003). Our changeable memories: Legal and practical implications. Nature Reviews: Neuroscience, 4, 231-234.
Schacter, D.L. (2001). The seven sins of memory: How the mind forgets and remembers. New York: Houghton Mifflin.
Elizabeth Loftus's homepage: http://socialecology.uci.edu/faculty/eloftus

References
Berkowitz, S.R., Laney, C., Morris, E.K. et al. (in press). Pluto behaving badly: False beliefs and their consequences. American Journal of Psychology.
Bernstein, D.M., Laney, C., Morris, E.K. & Loftus, E.F. (2005). False memories about food can lead to food avoidance. Social Cognition, 23, 10–33.
Brainerd, C.J. & Reyna, V.F. (2005). The science of false memory. New York: Oxford University Press.
Dreifus, C. (2007, 2 October). Proving that seeing shouldn't always be believing. New York Times: Science.
French, L., Sutherland, R. & Garry, M. (2006). Discussion affects memory for true and false childhood events. Applied Cognitive Psychology, 20, 671–680.
Gabbert, F., Memon, A. & Allan, K. (2003). Memory conformity. Applied Cognitive Psychology, 17, 533–543.
 Garry, M. & Gerrie, M.P. (2005). When photographs create false memories. Current Directions in Psychological Science, 14, 321–324.
Garry, M. & Wade, K.A. (2005). Actually, a picture is worth less than 45 words: Narratives produce more false memories than photographs do. Psychonomic Bulletin & Review, 12, 359–366.
Goff, L.M. & Roediger, H.L., III. (1998). Imagination inflation for action events: Repeated imaginings lead to illusory recollections. Memory and Cognition, 26, 20–33.
Heaps, C. & Nash, M. (2001). Comparing recollective experience in true and false autobiographical memories. Journal of Experimental Psychology: Learning, Memory, and Cognition, 27, 920–930.
Hyman, I.E., Jr & Billings, F.J. (1998). Individual differences and the creation of false childhood memories. Memory, 6, 1–20.
Johnson, M.K. (2006). Memory and reality. American Psychologist, 61, 760–771.
Johnson, M.K., Hashtroudi, S. & Lindsay, D.S. (1993). Source monitoring. Psychological Bulletin, 114, 3–28.
Kassin, S.M. & Gudjonsson, G.H. (2004). The psychology of confessions. Psychological Science in the Public Interest, 5, 33–67.
Laney, C. (2006). Emotional content of true and false memories (Doctoral dissertation, University of California, Irvine, 2006). Dissertation Abstracts International, 67 (12-B), 7429.
Laney, C. & Loftus, E.F. (in press). Emotional content of true and false memories. Memory.
Laney, C., Kaasa, S.O., Morris, E.K. et al. (in press). The red herring technique: A methodological response to the problem of demand characteristics. Psychological Research.
Laney, C., Morris, E.K., Bernstein, D.M. et al (in press). Asparagus, a love story: Healthier eating could be just a false memory away. Experimental Psychology.
Lindsay, D.S., Hagen, L., Read, J.D. et al. (2004). True photographs and false memories. Psychological Science, 15, 149–154.
Loftus, E.F. & Pickrell, J.E. (1995). The formation of false memories. Psychiatric Annals, 25, 720–725.
Mazzoni, G.A.L., Lombardo, P., Malvagia, S. & Loftus, E.F. (1999). Dream interpretation and false beliefs. Professional Psychology: Research and Practice, 30, 45–50.
Nash, R.A. & Wade, K.A. (2007). Innocent but proven guilty: Fake video evidence and false confessions. Manuscript submitted.
Okado, Y. & Stark, C.E.L. (2005). Neural activity during encoding predicts false memories created by misinformation. Learning and Memory, 12, 3–11.
Peterson, T. Kaasa, S.O. & Loftus, E.F. (in press). Me too! Social modelling influences on early autobiographical memories. Applied Cognitive Psychology.
Porter, S., Yuille, J.C. & Lehman, D.R. (1999). The nature of real, implanted, and fabricated memories for emotional childhood events. Law and Human Behavior, 23, 517–537.
Ross, M. (1997). Validating memories. In N.L. Stein, P.A. Ornstein, B. Tversky, & C. Brainerd (Eds.) Memory for everyday and emotional events (pp.49–82). Hillsdale, NJ: Erlbaum.
Scoboria, A., Mazzoni, G., Kirsch, I. & Relyea, M. (2004). Plausibility and belief in autobiographical memory. Applied Cognitive Psychology, 18, 791–807.
Schacter, D.L. (2001). The seven sins of memory: How the mind forgets and remembers. NY: Houghton Mifflin.
Wade, K.A. & Garry, M. (2005). Strategies for verifying false autobiographical memories. American Journal of Psychology, 118, 587–602.
Wade, K.A., Garry, M., Read, J.D. & Lindsay, S. (2002). A picture is worth a thousand lies: Using false photographs to create false childhood memories. Psychonomic Bulletin & Review, 9, 597–603.
Wade, K.A., Sharman, S.J., Garry, M. et al. (2007). False claims about false memory research. Consciousness and Cognition, 16, 18–28.