Conditioned Taste Aversion: Learning about Toxic Foods

Learning and memory are fundamental phenomena of our conscious and unconscious experience. The eventual goal of memory research is to explain how humans learn (e.g., learning a foreign language, or remembering a recent lecture). It is difficult to study the neurobiological methods of such learning in animals. Aside from the obvious difficulties in determining the conscious experience of animals, it is likely that the changes in the brain responsible for remembering discrete events are too subtle to be detected easily. There are billions of brain cells (neurons) and most of them are dedicated to the basic operation and survival of the body. The memory of a lecture is probably encoded by the connections of only a few dozen or a few hundred neurons. Thus, finding the physiological trace of a single episodic memory within the busy circuits of the brain is like trying to pull a single conversation from the hub-bub of a busy train station.

My laboratory approaches this problem by looking at an unsubtle form of learning that is basic to the survival of animal and which engages major circuit modules of the brain. Animals, especially omnivores like rats, are extremely good at learning which foods are safe to eat, and which foods are poisonous and hence to be avoided. Because of its survival value, animals can learn after a single experience that a food is toxic. In the laboratory, I use a model of food learning, conditioned taste version, in which the taste of a palatable solution (e.g., sugar water) is paired with a toxic drug (e.g., an injection of lithium that induces nausea). The sweet taste of sugar is a novel experience for a laboratory rat, and when the rat subsequently gets sick from the drug injection, the rat "blames" the sweet solution. (Of course, food poisoning and subsequent taste aversions are common experiences for humans too!) Thus, after a single pairing of taste and toxin, rats dramatically change their behavioral response, subsequently avoiding and rejecting the taste for months or years.Because conditioned taste aversion learning involves a radical change in behavior (from preference to aversion), and because it involves basic survival circuits of the brain (the lower parts of the brain that process taste sensation and gastrointestinal function), conditioned taste aversion is an ideal model for studying memory formation. Compared to tracking down the cells in the brain that encode a fleeting episodic memory, finding changes in neurons that control essential feeding functions should be relatively easy.

The link between my research and the more general experience of learning is the hypothesis that the same types of cellular changes that mediate taste aversion learning probably underlie all forms of learning and memory, albeit in different specific circuits. Changes in brain circuitry are caused by altered connections between neurons, so that a sensory input is re-routed to generate a different learned response. Taste aversion engages the evolutionarily ancient (and robust) circuits mediating feeding and digestion, while the more subtle learning of a foreign language engages the higher level linguistic circuits.