Pavlov’s dogs were conditioned to go to their treat. Why do some animals learn to interact with the bell instead?

High school students learn that Pavlov’s dogs were conditioned to associate the sound of a bell with getting food. The association was so strong that the dogs would begin to salivate when they heard the bell, before there was even a whiff of food. When they were finally presented with the food, they ate it.

They did not lick the bell.

But that’s just what some animals will do when presented with a stimulus, or cue, that has been paired with a reward: interact with the cue. Sometimes they’ll paw at it and even gnaw at it, said Sara Morrison, research assistant professor in the Department of Neuroscience at Pitt’s Kenneth P. Dietrich School of Arts and Sciences. Only once the reward has been delivered will they turn their attention to that tasty sugar pellet.

Animals who display this behavior are known as sign trackers. Alternatively, goal trackers head toward the area where they expect their reward to be delivered. A study from Morrison’s lab, published in the Journal of Neuroscience on June 25, found that sign tracking is an altogether different learning process than goal tracking.

The research showed that in rats, sign trackers, learning to value a cue relies on the availability of the neurotransmitter dopamine in a particular brain region at the time they receive their reward. Neither inhibiting nor increasing dopamine had any effect on goal trackers, Morrison said, a finding contrary to the way researchers have historically thought of cue-reward associations: as dopamine-dependent. While that seems to be the case for sign trackers, Morrison’s research indicates goal tracking is reliant on a different, non-dopamine-dependent learning mechanism.

A better understanding of the neurological basis of sign tracking, and how to unlearn it, may help researchers better understand risk factors for related psychiatric disorders.

Sign tracking has been linked to risk-taking, impulsivity and substance abuse relapse. And it’s “sticky” — even after a reward is taken away, sign trackers are more likely to react to a cue than goal trackers. Even if the reward is switched to something the animals aren’t interested in, sign trackers continue to react to the cue.

The research team, led by first author Ethan Herring (A&S ’23), used rats engineered with dopamine neurons that could be turned on and off using light, a method known as optogenetics, in the brain’s ventral tegmental area. The team could inhibit or increase the release of dopamine at will.

After giving rats the 8-second cue that they had been conditioned to associate with sugar pellets, the researchers turned off dopamine neurons at the time the reward was delivered.  “Inhibiting dopamine prevented the rats from learning to sign track,” Morrison said. “When we stopped the inhibition, after a few days some went on to become sign trackers again.”

Adding additional dopamine when rats received the reward, however, did not help sign trackers learn the association any faster. And when the additional stimulation was taken away, the rats stopped developing sign-tracking abilities for a few days.

“This says something really cool,” Morris said. “The signal seemed to scale to the amount of dopamine provided. It was like we were giving a bigger reward, then took half of that away.” The rats reacted as if the higher dopamine output was their baseline for a few days, after which they again began to improve as sign trackers.

That response makes sense, according to Morrison. “We all learn equally well from all different kinds of rewards. There must be some way our brain scales the rewards to the appropriate learning rate.”