The memory-transfer episode

It’s March 1960, and James V. McConnell, an assistant professor of psychology at the University of Michigan, is convinced that planarians — common flatworms — hold the key to unraveling the mystery of memory. He has decided to condition them to scrunch when a bright light is flashed. Then, he plans to chop them into pieces, feed them to their cannibalistic brothers and see whether the learned behavior is transferred from the trained victim to the naïve recipient. His eventual goal is to demonstrate that the engram — the physical representation of memory — is encoded in the structure of unique forms of RNA much as inherited traits are encoded in one’s DNA.

The story of “McCannibal and his Mau Mau” hypothesis has become part of the folklore of psychology. Often used in textbooks as a humorous hook to grab students’ attention in chapters devoted to learning and memory, two things are typically included: references to “memory pills” or “professor burgers” and the alleged fact that no one was ever able to truly replicate the findings. Those who did report positive results, the story goes, were poor scientists who either conducted sloppy experiments that lacked proper controls or simply deceived themselves.

But folklore tends to caricature people and events and is lousy history. Although, in the long run, the work did not stand up to the exacting scrutiny of those working in the area of memory research, McConnell’s planarian studies spawned a 15-year episode that tells us much about the workings of science when it is confronted — as it always is — with claims that depart in significant ways from prevailing views. Equivocal results are typical in such episodes and to jump to the conclusion that those who championed a losing cause must be poor scientists is hazardous at best. In fact, by the time the dust had settled roughly 200 independent research teams — many in the upper tiers of science — conducted memory transfer experiments, using dozens of learning paradigms and 23 types of subjects including, in addition to the flatworm and standard lab rat, octopuses, praying mantes, baby chicks, kittens and honey bees. Government agencies granted more than $1 million (in 1960s dollars) to conduct such experiments, and 247 research reports appeared in print. Clearly, something was going on here; there were enough encouraging results to beckon others to try their hand.

The early bird

It started innocently enough. In 1953, McConnell, a graduate student at the University of Texas, collaborated with Robert Thompson to show that planarians could be classically conditioned.

Thompson received his degree and went to Louisiana State University to work with rats, while McConnell, upon his arrival at Michigan, stuck with worms. He knew that by cutting a planarian across the middle into head and tail sections, each part would regenerate its missing half. But, he wondered, if you conditioned a planarian, which half of the bisected beast would retain the conditioned response? Working with two students in the newly formed Planarian Research Group, McConnell found, to his astonishment and delight, that the regenerated tails showed as much retention — and in some cases more — than the regenerated heads.

After nearly a year of McConnell’s wrangling with referees, the paper appeared in the Journal of Comparative and Physiological Psychology. In his next experiments, McConnell and the PRG showed that each regenerated part of trained worms cut in several pieces retained the initial training and, more important, a planarian that, after several regenerations, contained none of the structure of the originally trained animal also retained the memory.

These results led McConnell to think more seriously about the chemical nature of memory. To test this notion, he needed to find a way to transfer the putative molecules from a trained to an untrained animal. But how? They tried to graft the head of a trained worm onto the tail of a naïve worm — but the head kept falling off.

Next, they tried grinding up trained worms and injecting them into naïve recipients, but that didn’t work, either. The hypodermic needles were too big — getting one inside a flatworm was like trying to impale a prune with a javelin — and if, by chance, the needle was positioned well enough to inject the planarian-puree, it either oozed out or caused the worm to explode.

The answer came in March 1960 when fellow worm runner Jay Boyd Best wrote McConnell about the cannibalistic tendencies of a particular planarian species. McConnell and the PRG ran pilot studies in April and obtained positive results. Each of the next four replications — each run blind to guard against experimenter bias — also produced promising results.

Catching the public eye

For many, these results were hard to swallow. That McConnell first reported these results in the Worm Runners Digest, a journal/magazine he edited that published a mixture of straight science and spoof, did not help his case. Of more importance, the planarian work was not easily replicable. The beasts were difficult to train, and various experimenters — most notably a team working under the patronage of Nobel laureate Mac Calvin at Berkeley — reported their failure to do so.

Theoretical concerns made the work even less palatable. The conventional view held that memory consisted of electrical impulses traveling along specific neural pathways. But the spectacular success of Watson and Crick led some to wonder: If genetic information is stored in nucleic acids and proteins, why not acquired information, as well? Although many neurophysiologists thought this analogy nothing more than a bad pun, a number of molecular biologists, thinking the time ripe to apply their tools and analytic approach to the study of memory processes, began to discuss seriously whether RNA played a pivotal role in memory processes. Expectations ran high, and work proceeded along a number of collateral paths. The smart bet, however, was that if RNA or any other biochemical agents played any role, it was merely to fortify and grease the wheels of neural processing.

McConnell wagered on the long shot. Soon after the cannibalism experiments, he successfully injected naïve worms with RNA taken from those trained to negotiate a maze and reported that the training had transferred. He interpreted these findings as providing evidence that specific memories are encoded in the nervous system in the form of unique structural variants of RNA.

The cannibalism studies, both startling and vivid in their imagery, and McConnell, never one to shy away from the media, caught the public eye. At a time when scientists remained sequestered in their labs, McConnell appeared with his cannibalistic worms on television (i.e., “The Way Out Men,” “Mr. Wizard” and “The Steve Allen Show”), while articles profiling his work appeared in Time, Newsweek, Life, Esquire and Fortune. Eminently quotable, McConnell referred to his work as confirming the Mau Mau hypothesis, and the “McCannibal” moniker didn’t bother him one bit. He made grand pronouncements about the future of “memory pills” and “memory injections,” promising more than he and others working in the area could actually deliver.

None of this endeared McConnell to his critics.

Still, McConnell believed that eventually the data would win out, and many eminent psychologists, Donald Hebb, Harry Harlow, Karl Pribram, and Gordon Bower among them, fully supported his efforts, even though they did not share his interpretation of his results. In fact, up until 1965, McConnell was, as he put it, “riding high.” He was invited to share a platform with top-flight molecular biologists and electrophysiologists at conferences at the University of California, Los Angeles, in 1962 and Princeton in 1963. During the period from 1959 through 1964, he received more than $150,000 (in 1960s dollars) from the Atomic Energy Commission and the National Institute of Mental Health designated specifically for the planarian work. He was offered a fellowship to spend a year at the newly created Center for Advanced Study in the Behavioral Sciences in Palo Alto, Calif., in 1960, and he received a prestigious five-year Research Career Development Award from the National Institutes of Health, 1963–68. He received accelerated promotion to full professor at Michigan in 1963.

Everything changed when, in late 1965, four independent labs reported successful memory-transfer experiments using rats (and in one case, cross-species transfer between rats and mice). Two of these reports appeared in the high-impact journals Science and Nature.

No one could argue that rats cannot learn. Within a few months, more than 50 labs, including teams at Berkeley, Harvard, MIT and Yale, conducted transfer experiments. McConnell, after failed attempts using salamanders and mynah birds, also turned to rats.

And then things got really interesting.

Larry Stern is a professor of sociology and chair of the social and behavioral sciences at Collin College in Plano, Texas. Over the past several years, he has pored over the 42 linear feet of McConnell’s papers housed at the Archives of the History of American Psychology in Akron.

Kathy Milar, PhD, of Earlham College, is the historical editor of “Time Capsule.”