Imagine you are invisible. Congratulations, you are now actually less likely to remember what you were doing a few minutes ago, and possibly a lot longer ago than that. At least, this is the basic finding from a 2002 article by Sahakyan & Kelley, who showed that when people are asked to forget something they’d learned, they may actually do this by rapidly changing their internal context in a way that is similar to what happens if, say, you’re asked to imagine that you’re invisible.
In the laboratory, this is usually studied in the “directed forgetting paradigm,” in which subjects learn a series of items (List 1) to a certain performance criterion, and then half of the subjects are told to forget those items. All the subjects then learn a second list of items (List 2), and are all subsequently tested on their recall of both List 2 and List 1 items. Typically, the results show that directed forgetting actually works – subjects who were told to forget List 1 actually are worse at recalling it later, even if offered money to do so. In addition, subjects who were told to forget List 1 are also better at recalling List 2, which may reflect “escape from proactive interference” (i.e., if you forget the first list, it’s less likely to interfere with your memory for items in the second list.)
Some theories suggest that people are actively inhibiting the List 1 items when told to forget them. In contrast, Sahakyan & Kelley hypothesized that people are not actively inhibiting them, but merely changing their “internal context”, such that those items are now less likely to be retrieved. At first, this hypothesis may seem very abstract, but it actually rests on a well-accepted finding in the memory literature known as “encoding specificity.” Briefly, this principle states that you may be more likely to remember a childhood experience if you’re walking around your childhood home; or you may be more likely to remember a studied fact during an exam if you’re in a similar mood as when you studied it before. Encoding specificity has been shown in terms of mood, smells, and the physical environment.
To test this idea, the authors asked subjects to report their strategies during a directed forgetting experiment – sure enough, some reported “thinking of something else, such as the upcoming wedding of a sister.” In a subsequent experiment, they asked all subjects to imagine that they were invisible (and to state what they would do while invisible) after learning List 1 items. The results showed that all subjects showed that directed forgetting and “imagining you’re invisible” have similar effects on recall, both decreasing List 1 recall and increase List 2 recall by similar amounts relative to a control group. A subsequent experiment showed this “internal contextual change” manipulation in directed forgetting seems very similar to previous work involving a external (ie, environmental) context change between the lists.
It would be interesting to know how far into memory the effects of internal context change can actually extend. For example, are you less likely to remember everything after specifying a new context? Are you more likely to remember childhood experiences, perhaps a time when you also imagined you were invisible, or 10 feet tall? These are probably intractable but nonetheless very interesting questions for future research.
A recently published article cites an in-press follow-up to this study, in which they showed the effects of internal context to be even stronger among those with higher working memory spans. Unsworth & Engle have interpreted this to reflect improved context maintenance skills in those with high working memory capacity. I’m still waiting to get my hands on a copy of that paper, but will post an update when I do…
Reworking Working Memory