Suppressing Unwanted Memories

When reminded of something we would prefer not to think about, we often try to exclude the unwanted memory from awareness. Recent research indicates that people control unwanted memories by stopping memory retrieval, using mechanisms similar to those used to stop reflexive motor responses. Controlling unwanted memories is implemented by the lateral prefrontal cortex, which acts to reduce activity in the hippocampus, thereby impairing retention of those memories. Individual differences in the efficacy of these systems may underlie variation in how well people control intrusive memories and adapt in the aftermath of trauma. This research supports the existence of an active forgetting process and establishes a neurocognitive model for inquiry into motivated forgetting. KEYWORDS—executive control; inhibition; forgetting; trauma; memory To most people, forgetting is a human frailty to be overcome. More than we realize, however, forgetting is what we want and need to do. Sometimes we confront reminders of experiences that sadden us—as when, after a death or a broken relationship, objects and places evoke memories of the lost person. Other times, reminders trigger memories that make us angry, anxious, ashamed, or afraid. A face may remind us of an argument we regret; an envelope may bring to mind an unpleasant task we are avoiding; or an image of the World Trade Center in a movie may elicit upsetting memories of September 11. When confronting these reminders, a familiar reaction often occurs: a flash of experience and feeling followed rapidly by an attempt to exclude the unpleasant memory from awareness. At such times, memory is too effective and must be overcome. Suppressing retrieval shuts out the intrusive memories, restoring control over the direction of thought and our emotional well-being. For veterans of Iraq and Afghanistan, victims of Hurricane Katrina, witnesses of terrorism, and countless people experiencing personal traumas, the day-to-day reality of the need to control intrusive memories is unfortunately all too clear. Forgetting is their goal, and remembering, the human frailty. In this article we review recent research on the cognitive and neurobiological mechanisms supporting the suppression of unwanted memories, focusing on three points. First, stopping retrieval engages inhibitory control processes that make it harder to recall the avoided memory later on. Second, stopping retrieval is accomplished by brain areas similar to those that stop motor responses, which achieve control by reducing activity in brain structures fundamental to memory. Third, people who engage these brain systems effectively are also more successful at suppressing memories, suggesting that difficulties in managing intrusive remindings originate from difficulties in executive control. Collectively, these findings specify a model of motivated forgetting that is of practical importance in understanding and aiding those suffering from intrusive memories. THE NEED FOR RESPONSE OVERRIDE Intrusive memories seem to leap to mind in response to reminders, despite attempts to avoid those memories. Indeed, retrieval often occurs involuntarily in response to reminders. A key premise of our research is that controlling unwanted memories builds on mechanisms necessary to stop automatic motor responses. Consider an example of motor stopping. One evening, the first author accidentally knocked a potted plant off his window sill. As his hand darted to catch the falling object, he realized that the plant was a cactus. Mere centimeters from it, he stopped himself from catching the cactus. The plant fell and was ruined, but he was relieved to not be pierced with little needles. This example illustrates the need to override a strong habitual response to a stimulus, which is a basic function of executive control (Fig. 1). Without the capacity to override prepotent responses, we could not adapt behavior to changes in our goals or circumstances. We would be slaves to habit and reflex. How do we keep from being controlled by habitual actions? One possibility is that we inhibit undesired actions. By this view, when we encounter a stimulus, activation spreads from that cue to possible responses. Activation can be thought of as the amount of ‘‘energy’’ a response has, influencing its accessibility; a response will be emitted once it is sufficiently activated. If one wishes to override the response, one may engage inhibitory control, a subtractive mechanism that reduces the response’s activation. Perhaps we control memories in a similar way. Like Address correspondence to Michael C. Anderson, MRC Cognition and Brain Sciences Unit, 15 Chaucer Road, Cambridge, CB2 7EF England; e-mail: [email protected]. CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE Volume 18—Number 4 189 Copyright r 2009 Association for Psychological Science actions, memories can be triggered by activation spreading from reminders that we encounter. Might inhibition be recruited to stop retrieval, allowing us to avoid catching our ‘‘mental cacti’’? STOPPING RETRIEVAL CAUSES FORGETTING To study how people stop retrieval, Anderson and Green (2001) developed a procedure modeled after the go/no-go task, a paradigm designed to investigate motor stopping. In a typical go/ no-go task, participants press a button as quickly as possible whenever they see a letter appear on a screen except when the letter is an X, for which they are to withhold their response. Their ability to withhold the response measures inhibitory control over action (e.g., how well a person avoids catching the cactus). To see whether stopping retrieval also engages inhibitory control, Anderson and Green (2001) adapted this procedure to create the ‘‘think/no-think’’ paradigm. The think/no-think paradigm mimics situations in which we stumble upon a reminder to a memory we prefer not to think about and try to keep it out of mind. Participants study cue– target pairs (e.g., ordeal–roach), and are trained to recall the second word (roach) whenever they encounter the first word (ordeal) as a reminder. Participants are then asked to exert control over retrieval during the think/no-think phase. Most trials require them to recall the response whenever they see the reminder; but for certain reminders, participants are admonished to avoid retrieving the response. It is emphasized that it is insufficient to avoid saying the response—they must prevent the memory from entering awareness. Thus, participants have to stop the cognitive act of retrieval. Can people recruit inhibition to prevent the memory from intruding into consciousness? Since awareness cannot be observed, it is difficult to know whether a person prevents a memory from entering consciousness. Instead, the think/no-think procedure measures the aftereffects of stopping retrieval, based on the idea that inhibition of the unwanted memory might linger, making these memories harder to recall. To assess this behavioral footprint of suppression, a final test is given in which participants again see each reminder and are asked to recall every response they learned earlier. On this test, ‘‘think’’ items are recalled more often than ‘‘no-think’’ items (Fig. 2). This large difference, known as the total control effect, demonstrates how the intention to control retrieval modulates later memory. Importantly, a third set of pairs are studied initially but do not appear during the think/no-think phase, providing a baseline for measuring both a positive control effect and a negative control effect that contribute to the total control effect. The positive control effect reflects enhanced memory for ‘‘think’’ items above baseline recall due to intentional retrieval, confirming that reminders enhance memory when people are inclined to remember. The negative control effect reflects impaired memory for ‘‘no-think’’ items below baseline due to people’s efforts to stop retrieval. When people try to avoid being reminded, reminders not only fail to enhance memory, they trigger inhibitory processes that actually impair memory. The negative control effect is striking and counterintuitive, since repeatedly encountering reminders could remind us of the memory, making it more accessible, not less. The negative control effect occurs even when people are paid for each item they remember, making it unlikely that people are simply withholding responses. In contrast, asking people to Stimulus Prepotent Response Weaker, Contextually Appropriate Response Fig. 1. A typical response-override situation. In such a situation, a stimulus is associated with two responses, one of which is stronger (prepotent) and the other of which is weaker (indicated by the dotted line). Response override occurs whenever one needs either to select the weaker but more contextually appropriate response or to simply stop the prepotent response from occurring. Inhibitory control is thought to achieve response override by suppressing activation of the prepotent response. This basic situation describes many paradigms in research on executive control, including the go/no-go task. 95 100