In the 2004 film Eternal Sunshine of the Spotless Mind, a couple has their conscious memories of each other erased. But soon they are drawn together again, without knowing why. A new experiment shows that this surprising dissociation between conscious and unconscious forms of memory is not science fiction: People with a severe form of amnesia continued to experience an emotion powerfully, even after forgetting the experience that generated it.
“Where in the brain this lingering emotional experience is happening we don’t yet know. But we do know that it’s there, and these patients cannot remember why,” says Justin Feinstein, a graduate student at the University of Iowa who was first author of the study.
The study appeared on April 12 in the online version of the Proceedings of the National Academy of Sciences. The senior investigator was neurologist Daniel Tranel, whose laboratory has long studied people with brain damage to gain insights into normal cognitive brain functions.
Feinstein, Tranel, and neuroscientist Melissa Duff found five patients with severe damage to the hippocampus on both sides of the brain. The hippocampus is important for the formation of new declarative memories. All five patients had been diagnosed with “anterograde amnesia:” They are unable to remember events occurring after they had sustained brain damage.
Feinstein and his colleagues had the five patients (plus a control group of five healthy subjects matched for age, gender, and education) watch short film clips that could reliably evoke sadness. The test subjects reported feeling very sad, and raters who weren’t aware of the film clips’ content verified their emotional expressions. Minutes afterwards, memory tests showed that the five with amnesia had lost all or nearly all their declarative memories of the film clips, while the control group had retained theirs. Yet despite the loss of their explicit memories of the film clips, the five continued to feel markedly sad a half-hour later. On a subsequent day with the same test subjects, the experimenters obtained similar results with happiness-evoking film clips.
“It’s a neat study,” says Karim Nader, an expert on memory processes at McGill University in Montreal. And while it’s the first study in humans to dissociate explicit declarative memory from emotion in this way, Nader adds, its results are consistent with the current “two-streams” model of perception and memory: “It’s been argued that there is an unconscious processing pathway and a conscious pathway, both of which can acquire information,” he says.
Hints of this complementary structure in the brain have come largely from other observations in brain-damaged animals or people, including amnesic patients. In a famous report from 1911, cited by Feinstein, a physician invited an amnesic patient to shake his hand, and then pricked the patient with a concealed pin. Minutes later the patient had forgotten the incident, as well as the physician. But when the physician tried to introduce himself again, the patient refused to shake his hand—despite not being able to recall having been stuck by the pin.
In 1995, a group of researchers at the University of Iowa, including Tranel, showed more formally that a patient with bilateral damage to the hippocampus could not make an explicit memory of a stimulus, but did exhibit the appropriate conditioning to the stimulus, as if at some unconscious, nonverbal level the stimulus had been remembered. By contrast, a different patient with damage to the amygdala, known to be important for emotional memory, could remember the facts about the stimulus but couldn’t acquire the appropriate conditioning. From experiments such as these, researchers have come to believe that these conscious and unconscious pathways include the hippocampus and amygdala, respectively, although they haven’t fully mapped these pathways. “And how independent they are or how much they help each other out is still a big issue,” says Nader.
But the study led by Feinstein suggested that in the amnesic patients, the spared, unconscious, emotionally based pathway had compensated for the loss of conscious memory by becoming more sensitive. After these patients had viewed the film clips, their emotional states persisted longer than those of the healthy controls.
Nader notes that this “momentum” of emotion is a feature that is seldom discussed by researchers, but partly accounts for the results in Feinstein’s study. “Conscious declarative memories you can recall and then they go away. Emotional memories tend to linger.” For example, a mouse conditioned to freeze with fear at a special tone will tend to stay frozen for a while after the tone ends; and in people, too, emotional states can persist long after their objective reason for being is past.
Such a momentum may exist, in this apparently more primitive emotional memory system, for very basic evolutionary reasons. “The chances are that a predator you see nearby won’t disappear right away,” Nader says.