Change in neurons' responsiveness marks newly formed sensory associations during learning

During our waking hours, our brains are inundated with sensory information that shifts from one moment to the next. Recognizing meaningful associations between different snippets of this information is a basic form of learning that is essential for survival, even for animals with much simpler brains than our own. For learning to occur, these associations must be made and reinforced in some way at the neuronal level, but how this happens is poorly understood. Research reported this week sheds light on this problem by identifying a group of neurons whose activity changes during the learning process in a way that reflects the new association that is formed between two different sensory stimuli.

The findings are reported in Current Biology by André Fiala and colleagues at Bayerische Julius-Maximilians-Universität Würzburg, Germany.

To address the question of how the relevance of a stimulus is represented at the level of neuronal cells, the researchers used the fruit fly Drosophila as a model organism. These tiny animals can be trained to associate an otherwise neutral odor stimulus with a negative experience, such as a small electric shock, and ultimately learn to avoid this odor. Using sensitive imaging methods, the researchers observed the activity of certain neurons within the fly's brain during such training. They found that prior to training, these cells, which release the neurotransmitter dopamine, become especially activated if the negative stimulus--the small shock--occurs, but show only a weak activity in response to odors. Interestingly, after training, those cells respond with a prolonged activity toward the odor associated with the negative stimulus. In the course of training, the odor has acquired a negative relevance, which is reflected in this prolonged activity of dopamine-releasing cells. Interestingly, human brains seem to function in a similar way, with the difference that in our brains, rewarding stimuli (rather than negative stimuli) are activating dopamine-releasing neurons. The new findings suggest that the mechanisms by which environmental stimuli are evaluated by the brain and become predictive of one another appear to be similar in mammals and fruit flies, and suggest that continued study of the fruit fly system may offer valuable insights into the cellular basis for learning.

The researchers included Thomas Riemensperger, Thomas Völler, Patrick Stock, Erich Buchner, and André Fiala of Theodor-Boveri-Institut Lehrstuhl für Genetik und Neurobiologie Biozentrum, in Würzburg Germany. This study was funded by the by the DFG (SFB 554/A2).

Riemensperger et al.: "Punishment Prediction by Dopaminergic Neurons in Drosophila." Publishing in Current Biology Vol 15 No 19, 1953–1960, November 8, 2005. DOI 10.1016/j.cub.2005.09.042 www.current-biology.com

---

---