The Frontal Cortex

Character Memory

So I’m sitting in the movie theater the other day (I went to see The Lives of Others – go see it), and as soon as the first scene begins, the elderly lady sitting next to me says to herself: “Gosh darnit! I’ve already seen this movie! But it sounded so different when I read about it!”

When the movie was over, I struck up a conversation with the woman. It turns out that she read the description of the movie in the lobby and didn’t recognize any of the plot elements. However, as soon as she saw the face of the main character she instantly remembered having seen the movie a few weeks before. At that point, she was able to recall the names and motivations of the various characters, even though the plot of the movie still eluded her.

What fascinates me about this story is that it suggests a subdivision within episodic memory. The woman had no memory of the events that occurred within the movie (or any of the semantic concepts associated with the film), and yet she maintained a fairly astute recollection of all of the characters. She instantly recognized the people involved in the story, but couldn’t recognize the story itself.

The more I thought about it, the more this distinction made sense. Why wouldn’t the brain have a type of memory dedicated just to people? We are highly social animals, and much of our mental life is devoted to thinking about the behavior of others. It seems that we’d need a way to keep track of all of the “characters” in our life, independent of the events that the characters are involved in. I’d also wager that this type of “character-centric” memory is responsible for the fundamental attribution error.

Does anyone know of any neurological patients or experimental studies that have focused on this sort of “character memory”? Is there a case of somebody being unable to remember people, but able to remember everything else?


  1. #1 DavidD
    April 12, 2007

    The perception of familiarity in this woman you met is what’s missing in the Capgras delusion. VS Ramanchandran presented a case of this among other cases on a PBS show 6 years ago. After a head trauma this patient no longer had autonomic responses to seeing familiar faces and felt they were imposters, including himself. Interestingly it turned out the patient recognized his father’s voice normally over the phone, so the damage causing this was selective to vision.

    Wikipedia has a article on Capgras delusion in general.

  2. #2 Steve
    April 12, 2007

    I don’t think we have to posit a fancy new subdivision of episodic memory to explain this situation… plot descriptions are typically general and vague, and reading one might not call to mind a specific film. on the other hand, perceptual (and especially facial) memory is quite robust (people shown 10,000 natural scene images 1 at a time for 100 ms each are about 80% accurate in recognizing those pictures the next day, and are still accurate a week later).

    i also don’t see how this might be responsible for the fundamental attribution error. would you care to elaborate on that? i am especially skeptical because many members of east asian cultures do not show the FAE, and the best current theories take a more cultural perspective and suggest that cultural models (e.g. of the self) mediate this and other attributional behavior.

  3. #3 DavidD
    April 12, 2007

    Something else occurred to me. Wikipedia also has an article on prosopagnosia, where patients deny any conscious recognition of familiar faces, but usually they still have autonomic responses suggesting an unconscious awareness of a familiar face even without any conscious awareness. The Capgras delusion is the opposite, with a conscious familiarity, but lacking enough unconsciously that the subject eventually decides this familiarity they have is not real. The other person is an imposter.

    With this woman in the movies, it’s hard to know what she was experiencing reading about the movie in the lobby. Did she have some conscious familiarity with what she was reading, but it wasn’t complete enough for her to realize she had already seen it? Was it completely novel to her consciously, but underneath her autonomic responses were telling her how much she already liked this? So she decided she’d like it enough to go see it, only to discover she’d already seen it when she saw something more memorable?

    There are probably many examples of how much we need both our conscious and unconscious minds, and even then we don’t necessarily come up with the right answer until after we’ve made a choice we regret.

  4. #4 jonah
    April 12, 2007

    Thanks for your comments. The fundamental attribution error occurs when, in the process of interpreting other people’s behavior, we overestimate the importance of fundamental character traits and underestimate the importance of the situation (event). My somewhat silly conjecture is that a type of memory dedicated to “characters” would remember people solely in terms of their attributes, without considering what events elicited those attributes. So the woman watching the movie remembered all of the characters – and whether they were good or bad, honest or dishonest – but didn’t remember the plot twists that caused them to act in such a way.

  5. #5 Morris Berg
    April 15, 2007

    I read a memory/stress study (in an eye-witness type of scenario perhaps?) a while back that documented a very clear gender-difference along these lines. Women were more likely to recall “details” (characters?) whereas men were more like to recall the “gist” (plot?) under various stresses and after various lengths of time. Though not exactly on point, I read your post and immediately thought of this study. Like the males in this study, however, I can’t remember a damn thing about the particulars.

New comments have been disabled.