For the first time, scientists have created significantly smarter mice by adding a single gene to rodent embryos, showing that a seemingly minor genetic alteration can improve performance on a wide range of learning and memory tasks.

Researchers said their creation of "Doogie" mice -- named after the precocious television character Doogie Howser, M.D. -- could speed development of medicines for various cognitive disorders in people, including age-related memory loss and Alzheimer's disease.

But the finding also brings immediacy to a long-simmering debate about the ethics of making "designer babies" and the appropriateness of offering genetic enhancements that would not simply correct abnormalities but would increase human physical or mental capacities above those currently deemed normal.

The work, which involves a gene that helps the brain recognize patterns of cause and effect, also could help scientists create pets, farm animals or other creatures with unusual capacities to learn and remember, raising novel issues in the animal rights arena.

"We're in an era when breakthroughs in biology and intelligence are outpacing the culture's capacity to deal with the ethics," said Joe Tsien, the Princeton University molecular biologist who led the new effort. "There will be issues of access and who can afford it. Whether the social wealthy class will have the intellectual advantage over poor people, these are real questions coming down the road."

Others, however, warned against interpreting the work to mean that intelligence is a purely genetic trait. Especially in people, they said, intellect is an extremely complex phenomenon, and countless social and environmental experiences influence the way a brain assimilates and organizes information.

"There are lots of kinds of intelligence," said Charles Stevens, a Howard Hughes investigator at the Salk Institute in La Jolla, Calif. "The kinds of intelligence that make people good basketball players may not be the kinds that make a good symphony conductor or a good citizen."

The new research, described in today's issue of the journal Nature, lends support to a prominent theory about how the brain creates memories and then uses those memories to learn from experience. The work suggests that many different kinds of cognitive skills, such as recognizing something new in a room full of familiar objects or remembering where something is stored, rely on a single underlying molecular mechanism in the brain.

That mechanism, called long-term potentiation, entails the strengthening of key connections among the 100 billion neurons in the three-pound human brain. Intelligence is much more than the ability to create neuronal memory pathways with long-term potentiation, said Larry Squire, a professor of psychiatry and neuroscience at the University of California at San Diego and the San Diego Veterans Affairs Medical Center. Indeed, injury or disease can destroy certain memory skills without compromising intelligence. "But in the real world," Squire said, "memory and intelligence are closely linked."

In the new experiments, researchers inserted into mouse embryos extra copies of a gene that enhances long-term potentiation. The gene, called NMDA receptor 2B, or NR2B, is present in all mammals, including people. It directs production of a nerve protein that helps the brain recognize that two things are linked, such as the ringing of a bell and the delivery of food.

The NR2B gene normally becomes less active in middle age in humans and other animals, as memory and learning capacities decline. But Doogie mice have extra copies of the gene, and those copies have been programmed to remain active in old age. The enhancement is permanent and is passed on to offspring.

The researchers tested the mice to assess various kinds of intellectual prowess. In one test, they allowed the mice to explore two objects for a few minutes and later placed the mice in a box with one of those objects and a new object. In general, mice that remember an old object will spend more time with the novel object, and Doogie mice were far more likely to do so than were their non-engineered litter mates one day and three days after their original session -- an indication of better long-term memory.

In a second test, mice received foot shocks whenever a tone was sounded. Up to 10 days later, Doogie mice were significantly more likely to freeze in anticipation of pain when just the tone was sounded, compared with control mice. Doogie mice also were quicker to disassociate the tone and the pain -- or "unlearn" the link -- when taught to do so.

In another test, mice were placed in a tank of milky water and had to learn the location of a submerged platform on which they could rest. Compared with normal mice, Doogie mice were quicker to remember the platform's location once found and were more likely to swim immediately to that part of the tank in subsequent trials.

Scientists emphasized that the genetic change had not turned the mice into geniuses, whatever a rodent genius might be. In the first test, for example, even Doogie mice devoted equal attention to novel objects and old objects a week after becoming familiar with the old objects, indicating that the week-long memory of the old objects had been lost.

"They are quantitatively smarter, but they are not Einsteins," said Tim Bliss, chief of neurophysiology at the National Institute for Medical Research in London.

Nonetheless, the finding that such a minor manipulation might make a measurable difference in even a subset of intelligence raised immediate speculation about its potential medical use -- and abuse.

Human gene therapy so far has been tried only on adults, and soon may be tried in fetuses, but it has never been attempted in embryos and has never been used for anything other than the treatment of disease. But researchers predicted that with NR2B's role in learning now verified, drug companies will quickly discover medicinal compounds that enhance the gene's activity in the brain, which could then be given to adults.

"This is what pharmaceutical companies do best," Tsien said. "NR2 is a very good target."

Such interventions may come with a price, however. The NR2B gene enhances memory by increasing the flow of calcium into neurons, and high calcium levels increase the risk of brain seizures, such as those seen in epilepsy. Calcium also can exacerbate brain damage after a stroke. Although Doogie mice seem healthy, no one knows whether they will develop problems later.

"There's a lot one would want to know about before one took the step of doing this in people," said Squire. "We don't know if this improvement comes for free or not."

The prospect of engineering smart genes into babies also raises societal concerns, said Thomas Murray, president of the Hastings Center, a bioethics think tank in Garrison, N.Y.

"Perfectableism," or the belief that people should strive endlessly to be perfect, is a seemingly high-minded worldview with potentially large psychological and cultural downsides, Murray said. Constantly rising expectations can become a big burden, especially for children whose braininess has been "ordered up" from scratch.

The more imminent prospect of scientists genetically engineering greater intelligence into animals also raises profound questions, Murray said. "I don't know that upregulating this gene in chimps is going to get them to start reciting Shakespeare," he said. "But it might make them smarter, and we may have to revisit the question of what makes us different."