Orbitofrontal Cortex and Representation of Incentive Value in Associative Learning

Michela Gallagher, Robert W. McMahan, and Geoffrey Schoenbaum

The Journal of Neuroscience (1999)

DOI: 10.1523/JNEUROSCI.19-15-06610.1999

The conceptual gap or the scientific question: the functional role of the orbifrontal cortex (OFC) in incentive value based conditioning?

How the authors proposed the question: two previous papers from the same lab had reported that neurons in the OFC could encode the incentive value of outcomes during associative learning (orbitofrontal cortex and basolateral amydgala encode expected outcomes during learning, 1998, nature neuroscience; Neural encoding in orbitofrontal cortex and basolateral amygdala during olfactory discrimination learning, JNS, 1999). So the authors sought to test the functional roles of the OFC in supporting the incentive value based conditioning.

Brief summary: by training rats to learn a reward devaluation task, the authors shown that the lesions of the OFC impaired the reward-devaluation induced decrease in conditioned response, while left the initial learning of cue-reward association and taste aversion learning unaffected. It implied that the OFC may be critical for bridging the associations of predicting cues and unconditioned outcomes.

How the authors tell the story: it's a coherent inquiry about the functions of the OFC by considering their previous recording experiments of OFC neurons. First, previously they found that neurons in the OFC could encode the expectation of the incentive value of the outcomes. And then they want to know that whether those signals are the essential driving force for incentive value based behaviors. So they performed lesion experiments. However, unexpectedly they observed that the OFC-lesions did not affect the initial learning of cue-reward association. So it seems that the OFC is not necessary for driving this kind of associative learning. Facing the negative results, many people may give up promptly. But the authors performed a further test after manipulating the incentive value of the unconditoned outcomes, that is devaluation of the unconditioned outcome by pairing it with the aversive lithium chloride, and found that the OFC-lesions impaired the devaluation induced decrease in conditioned response compared with the control group. The results suggested that the OFC played an important role in the incentive value changed conditions, but not the value-constant condition. Moreover, the OFC-lesioned group also shown normal taste aversion learning, indicating that OFC lesions did not impaired the ability to detect the change of the incentive value. Taken together, the multi-step behavioral tests in this study jointly implied that the OFC may play a pivitol role in remapping the cue-reward contingency under the value changeable conditions.  

This study instantiates the essences of scientific research. Scientists are the guys who keep figuring out the constraints which define the working conditions of the Nature. The key is keep moving from one condition to another, especially when you are comfronting negative results. 

The ongoing question(s) inspired by the current study: what's the core the functions of the OFC in the devaluation test? It's not outcome expectancy, not value updating, not the building of cue-reward contingency. Then what other functions? Remapping the cue-reward contingency? Why specifically for'remapping'? It's quit perplexing.