(L to R) Sebastian Haesler, Naoshige Uchida and Jeremiah Cohen
In 1954, Olds and Milner discovered that direct, electrical stimulation of particular brain areas was powerfully rewarding: rats would compulsively press a lever to obtain stimulation, ignoring naturally-rewarding stimuli such as food, water or mates. More recent work has shown that drugs of abuse hijack these reward systems, which results in compulsive drug taking in addicts.
Dopamine is a key neurotransmitter for reward in the brain. In the 1990s, Schultz and colleagues discovered that dopaminergic neurons in the midbrain responded to reward when the reward was unexpected, whereas their response to reward was diminished when a sensory cue predicted the reward. That is, dopaminergic neurons fire when the outcome is better than expected, while they cease firing when the outcome is worse than expected. This suggests that the brain’s reward system works in an “efficient” way: it changes its firing only when the brain fails to predict outcomes correctly, that is, when its knowledge needs to be adjusted. Such a signal is called “reward prediction error” and can be defined as:
Reward prediction error = received reward – expected reward
Although these findings changed the way we think about the reward system, how such calculations are performed in the brain is unknown. To understand this calculation, we aimed to record the signals different types of neurons produced.
One of the challenges from the outset was how to determine what types of neuron we observed during experiments. In classical neurophysiological studies, the experimenter places a microelectrode in an animal’s brain and compares neuronal activity with the animal’s behavior. Our work is a new application of molecular and genetic techniques developed in the last decade. We “tagged” dopaminergic neurons (and, in separate experiments, inhibitory GABAergic neurons) with a light-activated ion channel originally found in algae. We then implanted several small electrodes and a fiberoptic cable into the midbrain. While recording the activity of randomly-sampled neurons, we could ask each neuron whether it responded to a pulse of light from a laser. If it did, then we knew it was a dopaminergic (or GABAergic) neuron.
We recorded from identified dopaminergic or GABAergic neurons while mice smelled odors that predicted rewards or punishments. We found that dopaminergic neurons signaled reward prediction error and that nearby GABAergic neurons signaled reward expectation.
Dopaminergic neurons = received reward – GABAergic neurons
Our findings give a new framework in which to think about how drugs hijack the reward system. Previous studies have shown that GABAergic neurons–that we found send an “expectation” signal to dopaminergic neurons–are inhibited by drugs such as opiates and cannabinoids (i.e., morphine and marijuana). Our findings suggest that this, in turn, deprives dopaminergic neurons of their “expectation” signal, causing a reinforcement signal that is normally suppressed by the expectation signal. This may cause people take drugs compulsively, like the self-stimulating rats in Olds and Milner’s experiment.
Hear more on neuropod (January 2012 issue: “Great expectations“)
[January 18, 2012]