Memory in Mice, Nothing More Yet

By Kathleen Sheridan

Just six weeks after the provocative "Genetic enhancement of learning and memory in mice" (Tang YP, Shimizu E, Dube GR, Rampon C., Kerchner GA, Zhuo M, Liu G, & Tsien JZ) appeared in the September 2, 1999 issue of Nature, the media have moved on in search of other hot headlines. Since some dust has settled, it may be worth our while to take a calmer look at what the Princeton researchers discovered, and what they did not discover, and how these issues affect ethics and health policy.

According to Jacqueline N. Crawley, Ph.D., Chief of Section on Behavioral Neuropharmacology at the National Institute of Mental Health, nearly all recent advances in behavioral genetics are coming from rigorous and replicated studies with mice. The Princeton groupís remarkably creative project was based on an observation that as mice age, memory function declines Ė a phenomenon not uncommon in human aging. In mice, and in humans, the medial temporal lobe and hippocampus are critical in shifting short-term memories into long-term ones. The researchers enhanced a gene, NR2B (very active in young mammals but less active in adults), which helps build a protein in the hippocampus called NMDA. The NMDA receptor acts as a graded switch for long-term memory formation. The enhanced activation of NMDA in these mice corresponded with their learning faster and remembering longer in a series of behavioral tasks. In reporting their results, the researchers concluded with a now oft-quoted line, "our results suggest that the genetic enhancement of mental and cognitive attributes such as intelligence and memory in mammals is feasible."

In Research Methods 101, we all learned Commandment Number One (Commandments Two through Ten are Replicate, Replicate, Replicate, Ö Replicate). The First Commandment states that when interpreting statistically significant results (one should never interpret nonsignificant results, of course), one must never reach nor stray beyond the bounds of the hypotheses and methodology of the study itself. In other words, never jump to conclusions!

There is some suggestion in media reports that the Princeton researchers knew that line would be ever so quotable and clearly intended it that way. Too bad. The media took the bait and leapt from "memory" to "intelligence" and "IQ," as if they were synonymous terms, unitary constructs, and that they meant the same thing for all mammals, including humans. None of it is quite so simple.

Memory, for instance, is made up of several complex variables. And even though the hippocampal area is central to memory, many areas of the brain work in harmony to create and to retrieve a memory. When I play bridge, for instance, the memory of dealing the cards resides in one group of neurons; adding up my points comes from another; the residual effect over having been set in the last hand I played stems from yet another. Each is used in building a cognitive experience, a memory.

Memory (itself a constellation of several actions), then, is a cognitive process that is part, but only one part, of what behavioral scientists measure when they talk about "intelligence." But intelligence is also a complex construct which includes other such complex constructs as sensation, perception, problem-solving, and abilities to abstract and generalize. Besides, any definition and measurement of intelligence, in animals and humans, is selective, incomplete, and affected not only by genetic information but by agreed-upon sociocultural norms for the efficient expression of intelligence. See what I mean?

The researchers discovered incredibly valuable information about how to genetically enhance one type of hippocampal-mediated memory in laboratory mice who performed simple behavioral tasks. Exciting, pathfinding, generative results! What does this finding tell us about this type of memory in other mammals? In humans? We donít know. Can scientists genetically enhance this type of memory in other mammals? In humans? We donít know.

As ethicists and policy planners, we are obligated to understand and to anticipate the implications of discoveries such as these. Already, the good-bad dichotomy has been brightly cast. These findings are potentially therapeutically useful for older adults, and others with memory impairments, like Alzheimerís sufferers. On the other hand, we are concerned that parents will want to genetically engineer their fetuses to capitalize on such features as "good memory." Where will it all lead? We donít know.

10/21/99