In this summer’s blockbuster Inception, Dominic Cobb (Leonardo DiCaprio) and his team of thieves delve into the minds of unwitting dreamers to extract valuable information. In one encounter, Cobb is approached by powerful businessman, Saito (Ken Watanabe), with a formidable proposal: perform inception, the planting of an idea, to convince business rival Robert Fischer (Cilian Murphy) to break up the Fischer empire.
Cobb’s point man, Arthur (Joseph Gordon-Levitt), insists that inception is impossible. “True inspiration is impossible to fake,” he says, because “The dreamer can always remember the genesis of the idea.”
While the methods Cobb and his team use are elegant, extraordinary, and, unfortunately, pure science fiction, the concept of inception is very real. In fact, inception is possible for exactly the opposite reason Arthur insists that it is impossible: the mind often has a bad habit of forgetting, or “misremembering”, what bits of information originated where. Two concepts in particular are key in understanding how this occurs and the implications of such memory errors: memory misattribution and cognitive fluency.
Our minds are constantly forming connections between different pieces of information. If you’ve ever found yourself thinking of some random topic and suddenly wonder “How did I get here?”, then trace your though processes back to a completely different topic, you have first hand experience of the vast network.
It’s because of this vast network that our minds are able to do such amazing things. We can access a single piece of information from unimaginably huge database in milliseconds. We solve problems in new and different ways. We create metaphors and allegories. However, constantly changing connections can lead to a flaw called memory misattribution.
Misattribution occurs when memories of the original event are combined with information gathered after the fact. After the 1986 Challenger disaster, Neisser and Harsch interviewed witnesses to the event after a few weeks, then again after a year. The correlation between the two recollections were dismal, despite participants insisting their memories were as vivid as ever. The older memories were riddled with misattributions as the witnesses unconsciously incorporated facts they had misinterpreted or never even witnessed until shown on the news months after the disaster.
Errors in older memories are perhaps forgivable, since the older the memory, the more likely misattribution will occur. Even so, there are cases where misattribution can occur immediately after the event, which can be a cause for concern when looking at eyewitness testimony.
Elizabeth Loftus’ famous “Reconstruction of Automobile Destruction” demonstrates just how malleable memories of an event can be. In her study, participants were shown a video of a car accident, then immediately after were asked to answer questions about the accident. The question of interest was: “How fast was the car going when they hit each other?” However the verb used changed participants’ answers. With the question “How fast were the cars going when they smashed into each other?” participants reported much higher speeds than if bumped or hit were used in the question. In a followup study, participants returned one week later and, without viewing the video again asked if there was any broken glass at the scene. Participants who had been asked the question “How fast were the cars going when they smashed into each other?” were more likely to say “yes” even though there was no broken glass in the video. From these experiments and others, it is obvious why psychologists often take such an interest in police interview techniques and accuracy of eyewitness testimony.
So memories are manipulable. This is clear (if not somewhat distressing), but surely it's just small details that can change (such as the speed of the car or the presence of broken glass). We know we saw a car accident, that such an even actually happened to us, right?
Research suggests that, in fact, entirely new memories can be planted. It all comes back to misattribution. The principle is simple: imagine a fictional scenario enough times and the memory of the fantasy can be misattributed as an actual memory.
Much recent research into memory distortion has been spurred by a series of criminal cases with one thing in common: otherwise normal people emerging from psychotherapy to accuse their parents, relative, neighbors, teachers, or other acquaintances of sexual abuse. Many of the methods in psychotherapy include uncovering “repressed” memories with techniques such as age regression, guided visualization, and hypnosis. These methods can be extremely beneficial in confronting childhood trauma, but in understanding memory attribution and suggestibility one can see how even the most well meaning psychotherapist can permanently alter their patient’s memory with the wrong leading questions.
Loftus investigated the phenomena of implanted memories with her “lost in the mall” technique. This experiment (which you can try at home if you like to mess with your sibling’s head) involves giving the participants narratives of childhood events, supposedly supplied by family members. Little did they know, one description was false, namely an event where the participant at age 5 or 6 was lost in the mall for an extended period of time before finally being reunited with their family. After several days of trying to recall this even, 25% of participants gave a detailed account of the otherwise fictitious event.
All this may seem disconcerting, but it is the price we pay for the flexible and adaptable minds that allow us to be human. Still, if you are suddenly worried about grand conspiracies and implanted memories, ease your mind. While 25% is significant, it is by no means a majority. Debate among researchers still rages on about how easy planting memories is. Not only are plausible memories easier to manipulate than implausible memories, but it is often determined by a complex interaction of social, situational, and subjective variables. However, the next time you find yourself in an argument over how something happened or who said what, just consider that perhaps your memory isn’t as infallible as you originally thought.
So far we have discussed how memories can easily be altered by external influences, both intentional and unintentional. It is no wonder why Cobb and his team go after Fischer’s memories of his father as the basis of their plan. But there is one more factor to understand when it comes to the planting of ideas. For surely that moment of “true inspiration” that Arthur insists is impossible to fake is quite different than a memory.
Fluency, as defined by Wikipedia, is “the property of a person or of a system that delivers information quickly and with expertise.” We talk about reading fluency as the ease with which we comprehend the strings of letters and words on a page, or language fluency as the ease with which we comprehend the sounds we’re hearing or can smoothly articulate the appropriate sounds in return.
Similarly, there is cognitive fluency: the speed of recognizing a stimulus you have seen before. Fluency is often discussed in conjunction with “priming”—essentially that neurons are “warmed up” to respond more easily to a specific stimulus the next time they are activated. The more they fire, the more they are primed. For example, if you’re so engrossed in a novel that you’re tuning out everything else around you, you will probably not notice someone trying to talk to you until they say your name. Why? Because we are primed to hear our names, so it is much harder to “tune out” such a strong neural signal. In a similar scenario, if you’re reading a thrilling article about Ebola, and the TV in the background suddenly mentions Ebola, you are probably more likely to take notice than if it mentions, say, politics.
Fluency is particularly interesting because the effects often hang around even after the original context is forgotten. For example, take this clichéd office comedy scenario: the curmudgeonly boss is looking for ways to save his company/increase profit/crush the competitor/etc. His underling proposes a brilliant idea, but the curmudgeonly boss waves it off. Thirty seconds later, the boss exclaims “Eureka!” and proposes the underling’s idea as his own. This may be an office comedy cliché, but it is a very good example of “planting” an idea using fluency. The boss in this scenario is so much above his lowly underlings he forgets about their existence. Thus, when the underling proposes an idea, he ignores him. Yet the idea is primed, giving it a stronger “signal” than any of his other ideas. So when the boss thinks of the idea, it seems special some how. “Why does this idea stand out to me?” he asks himself (unconsciously). If he remembered his underling, the answer would be “Because my underling just said it.” But the original context is forgotten, so instead the boss thinks “This idea stands out for some reason. It must be a good idea!”
Experiments studying fluency employ techniques similar to the scenario described above, such as Jacoby et al’s False Fame study. In this study, participants are asked to read a list of made-up names. They return a day or two later (after the factual memory has faded) and are asked to rate how famous each name is from a new list. Some names are actual celebrities, some are made up, and some were on the previous list. The names from the first list were rated as more famous than the names participants had never seen before. Because the participants had seen the names before, the detectors were primed causing an increase of fluency. As a result, these names struck the participant as being distinctive in some way; however since the memory of the initial list had faded, they interpreted this change in fluency with their current situation: “I’m answering questions about fame, and since this persons name rings a bell, I must have seen it somewhere before so they must be famous.” In the same way that fake names can become famous, calligraphy designs can become beautiful, and sentences can become true (which says a lot about the impact of propaganda).
The Psychology of "Inception"
Now that we understand the mechanisms of misattribution and fluency, Cobb’s plan, as complex as it is, suddenly makes a whole lot of sense (which is saying a lot for a science fiction movie). While Arthur was wrong in his assertion that inception is impossible, he was right that they could not simply plant an idea by itself. Had they planted a big neon sign in Fischer’s dream that said “I will split up my father's empire,” they certainly would have primed the idea. However, making fluency an effective tool of manipulation not only requires the subject to forget the original context, but also requires a new context so the subject interprets the fluency in the right way (e.g. the False Fame study).
Cobb and company solve this problem by approaching Fischer’s relationship with his father. Their method of manipulating Fisher’s memories with misattribution is elegant. They prime the idea that his father had something planned for Fischer, that there was something more to the old man that he never knew. The maze they put Fischer through only strengthens this priming until the final confrontation with his father, which serves as the context for interpreting the fluency of the idea “I will become my own man and split up my father’s empire.” By the time Fisher wakes up, the seed is planted, and through misattribution the memory of the dream is permanently integrated with memories of his father.
The technology might be science fiction, but Cobb and company’s other methods have a strong basis in reality. Could Cobb and his team have achieved the same results by simply sending Fischer to a corrupt psychologist? It’s possible, but certainly wouldn’t have made as good of a movie.
References and Further Reading
Daniel Schacter, The Seven Sins of Memory
Brown and Nix (1996), "Turning lies into truths: Referential validation of falsehoods".
Harsch and Neisser (1989), "Substantial and irreversible errors in flashbulb memories of the Challenger explosion".
Jacoby, Woloshyn, and Kelley (1989), "Becoming famous without being recognized: Unconscious influences of memory produced by dividing attention".
Loftus and Palmer (1974), "Reconstruction of Automobile Destruction: An Example of the Interaction Between Language and Memory".
Loftus and Pickrell (1995), "The formation of false memories".
CaribeM on October 21, 2012:
Excellent and interesting article.