The process, one scientist says, is something like what happens when a comic strip character sees something and is immediately reminded of something else.
"I like to think of it like a cartoon character with a thought bubble over his head," explained Geoffrey Schoenbaum, an associate psychological and brain sciences research scientist in the Krieger School of Arts and Sciences at Johns Hopkins. "There's a neural representation of something in the mind that is invoked by the environment, but not yet present in the environment."
This comparison led to a cartoon for the cover of the August 28 issue of the journal Neuron, where the study is published. On the cover, a cartoon rat stands at a fork in the road, consults a map, and thinks, in a thought bubble, of cheese. The cheese isn't present in the rat's surroundings, but the rat knows through past experience that choosing the right path could lead him to it.
Schoenbaum, who becomes an assistant professor of anatomy and neurobiology at the University of Maryland School of Medicine on Sept. 1, directly observed the brain mechanisms involved in such predictive associations by using implanted electrodes to record the activity of individual nerve cells in two regions of the brain, the amygdala and the orbitofrontal cortex.
In earlier studies, the researchers had demonstrated that nerve cells in these two connected brain regions changed their firing patterns to reflect the associations between cues and outcomes during learning.
For the study published in Neuron, they examined how changes in neural activity in amygdala might be supporting changes in the orbitofrontal cortex. To do this, they recorded in orbitofrontal cortex from two groups of rats: a normal group and a group with chemical lesions to their amygdalas. Prior to the experiments, the rats' water sources were taken away for a time to make them thirsty. In repeated trials, scientists would then link odors to the appearance a few moments later of either desirable drinking water, which was laced with sugar, or undesirable drinking water, laced with quinine and unpalatable even to thirsty rats.
"We found that the patterns that normally develop in orbitofrontal cortex when rats are smelling the odor cue, which appear to reflect information about the predicted outcome, failed to appear in rats with amygdala lesions." Schoenbaum said.
As normal rats learned to use the odor cues to predict the type of fluid they would receive, the orbitofrontal cortex activity patterns eventually began appearing much more quickly, starting in response to the odor cue but before the fluid was given. In experimental rats, though, early activation of orbitofrontal cortex patterns never occurred.
"This is the most direct evidence yet for how one brain system, the amygdala, controls the way representations are made in another directly connected system," said Michela Gallagher, chairman of psychological and brain science at Johns Hopkins and a co-author of the Neuron paper. "We have long suspected the existence of this network because the amygdala is necessary for the type of learning process we studied, but we have only recently begun to examine how these interactions occur and shape how the world around us is represented in the brain."
Schoenbaum noted that the rats with lesioned amygdalas still learned to avoid the undesirable drinking water, apparently through other mechanisms that back up those normally present in the connection between the amygdala and the orbitofrontal cortex. He added that scientists can't yet say for certain what the brain cell activity patterns they identified in the orbitofrontal cortex represent -- a mental picture of the water to come, for example, or of the gratification or lack of gratification the water will produce, or something else entirely.
Schoenbaum hopes to apply data from the study to research into brain changes brought on by addiction. He's interested in investigating the possibility that addiction may damage or impede the connection between the amygdala and the orbitofrontal cortex, impairing an addict's ability to adequately assess the consequences of his or her actions. Gallagher will use data from the study to aid her studies of how aging can affect memory functions.
This research was supported by grants from the National Institute of Mental Health and the National Institute of Aging. Other authors on the paper were Barry Setlow and Michael Saddoris.