The meeting of the Memory Disorders Research Society in Chicago, held 9-11 October, opened with a session on two neurodegenerative syndromes: primary progressive aphasia (PPA) and semantic dementia (SD). Each of these syndromes bears some resemblance to the progression of Alzheimer’s disease, but with the key difference that memory is not initially impaired in these patients.
Marsel Mesulam of Northwestern University, Evanston, Illinois, began with a characterization of the progression of PPA. In this disease, patients initially have difficulties in basic language functions such as finding names for objects or comprehending speech. These deficits are associated with brain damage occurring in the left hemisphere in perisylvian regions that looks at the cellular level somewhat similar to the damage associated with Alzheimer’s disease (Mesulam, 2001; Sobrido et al., 2003).
John Hodges of the University of Cambridge presented a discussion of SD. Patients with this syndrome initially have difficulty recognizing objects or remembering the meaning of common words. Brain damage in SD occurs in the areas associated with basic knowledge of objects and facts about the world in the ventral temporal lobe, particularly anterior regions. Hodges and Mesulam discussed the question of symptom overlap, namely that common deficits can make it hard to distinguish PPA from SD in some early cases. Both agreed that it will be crucial to understand the specific cellular mechanisms and genetic factors in these diseases. Comparisons to Alzheimer’s disease may yield important insights into the progression of each of these neurodegenerative syndromes.
The second session of the meeting concerned the role of prefrontal cortex in the retrieval of memories from long-term storage. Memory storage occurs in the temporal lobe, but it is clear that the prefrontal cortex is critical for both enhancing memory storage and helping retrieve memories.
Michael Rugg of University of California, Irvine, described a series of recent functional neuroimaging studies that examined brain activity during successful retrieval, that is, when one tries to remember something and succeeds. Rugg identified a region on the right side of dorsolateral prefrontal cortex where activity is associated with successful retrieval and additional separate areas that appear to be active when retrieval isn’t quite certain.
Daniel Schacter of Harvard University discussed the role of the prefrontal cortex in false memory. In this case, people believe they are retrieving memory successfully, but for some reason the retrieved memory is inaccurate. Schacter reported neuroimaging studies using paradigms designed to create the sense of a false memory for unusual facts. For example, participants might be shown a statement such as, “It takes four hours to hard-boil an ostrich egg,” and told that it is false (it takes six hours). However, later participants may claim that they believe the statement is true, probably reflecting a sense that they have seen it before but can’t remember specifically that it was said to be wrong. Neuroimaging identified differences between good and false memory. Brain activity associated with good memory retrieval tended to be absent when this type of false memory occurred.
Michael Petrides , McGill University, Montreal, rounded off this session by reviewing a series of studies comparing the roles of two regions of prefrontal cortex in memory retrieval. The mid-dorsal region (towards the top of the brain) was found to be more critical for maintaining a set of information over time to guide responses. The mid-ventral region (closer to the bottom of the prefrontal cortex above the eyes) was found to be more critical for memory retrieval.
The prefrontal cortex is an intricately complex region that hosts brain networks supporting memory and virtually all other cognitive functions. It is larger and much more highly developed in humans than in any other animal. The presentations in this session reflected the cutting edge of neuroimaging studies into the basic questions of the role that the frontal cortex plays in one of the most basic of memory functions: the retrieval of stored memories so that we can think about and use them.
Saturday’s session was devoted to the role of sleep in memory reactivation and consolidation. Bruce McNaughton , University of Arizona, Tucson, argued that ensembles of hippocampal activity don’t encode anything. Rather, he suggested that they are random keys to information stored in cortex. In his model, sharp wave sleep (SWS) activity reactivates information in the hippocampus, while hippocampal complex ripples are periods during which memory is being reactivated.
Carlyle Smith of Trent University, Peterborough, Ontario, presented research on rapid eye movement (REM) sleep and memory. He found that subjects with higher IQ had greater post-training REMs and greater REM density during sleep; however, these differences only occurred after training on motor and cognitive skills. Smith’s group also found that alcohol reduces the number and intensity of REM periods, impairing storage of incrementally learned cognitive and motor skills, but not of explicitly learned declarative memories.
Robert Stickgold , Harvard Medical School, described improvements in cognitive-motor skill tasks after sleep, suggesting that sleep is not only important for the retention and consolidation of past learning, but also may be critical for enhancing previous learning Walker et al., 2003). Stickgold presented a broader framework for the role of sleep in memory, in which SWS stabilizes memories while REM sleep enhances learning. In a study of learning in schizophrenic patients, Stickgold noted that the patients showed fine initial learning on their tasks, but no sleep-related improvements as compared to controls. Similar results were found in cocaine addicts. Overall, these talks (and others by Ken Paller of Northwestern, Siderta Ribeiro of Duke University, and Lynn Nadel of the University of Arizona) provided an exciting view of the emerging appreciation of the role of sleep in learning and memory.
Mark Gluck is at Rutgers University, Newark, New Jersey, Paul Reber co-organized the conference and is at Northwestern University, Evanston, Illinois.
- Sobrido MJ, Abu-Khalil A, Weintraub S, Johnson N, Quinn B, Cummings JL, Mesulam MM, Geschwind DH. Possible association of the tau H1/H1 genotype with primary progressive aphasia. Neurology. 2003 Mar 11;60(5):862-4. PubMed.
- Walker MP, Brakefield T, Hobson JA, Stickgold R. Dissociable stages of human memory consolidation and reconsolidation. Nature. 2003 Oct 9;425(6958):616-20. PubMed.
No Available Further Reading