Blockade of NMDA receptors in the dorsomedial striatum prevents action–outcome learning in instrumental conditioning

Henry H. Yin, Barbara J. Knowlton and Bernard W. Balleine

European Journal of Neuroscience (2005)

doi:10.1111/j.1460-9568.2005.04219.x

The gap/question: where in the brain action–outcome associations are encoded?

How the authors proposed the question: Bernard W. Balleine is a graduate student of Anthony Dickinson, who is a representative figure in the instrumental conditioning field. Before this study, Balleine and colleagues had performed a series of lesions experiments targeting many brain areas, including insular cortex, hippocampus, nucleus accumbens, mediodorsal thalamus, anterior thalamic nuclei, basolateral amygdala, prelimbic cortex. And then the lateral orbitofrontal cortex (2007) after this study. So this study was a natural extension of their series’ mapping work.

Brief summary: by utilizing well-established behavioral assays (grounded on a series of classic works from Dickinson’s lab in 1990s and 1980s), Yin et al, shown that the lesions of the posterior dorsomedial striatum (pDSM) but not the dorsolateral striatum, before but not after, the acquisition of action-outcome associations eliminated the devaluation induced decrease of the lever-press rate for the devalued outcome, implying a functional role of pDMS in supporting the learning of action-outcome association.

In this study, the authors first trained rats to press two levers to get two different rewards. After the acquisition of the two action-outcome (A-O) associations, they applied a devaluation procedure during the following day. To selectively devalue a reward, rats were free to get the reward for 1 h before the afterwards extinction test (no reward was delivered for both lever-press). For the normal animals, the lever-press rate to the devalued reward was significantly lower than to the non-devalued one. However, lesions of the pDMS by the injection of NMDA receptor antagonist specifically before the acquisition of A-O associations eliminated the ‘decreased’ level-press rate for the devalued reward under the physiological condition. Notably, the lesions did not significantly affect the lever-press rate for both rewards during the acquisition phase per se, indicating that the effects was not due to the impairments in motor control. The deficits may more possibly be attributed to the failure in learning the A-O associations and, the failure could be caused by the insensitive detection of the reward-outcome when pDMS was inhibited.