Fractionation of reward processing into independent neural representations by novel decoding principle

Abstract How to retrieve latent neurobehavioural processes from complex human neurobiological signals is an important yet unresolved challenge. Processing value and salience information are fundamental yet mutually confounded pathways of reward reinforcement. Here, we develop a novel analytical approach, orthogonal-Decoding multi-Cognitive Processes (DeCoP), to decode brain-wide responses into spatially overlapping, yet functionally independent, evaluation and readiness networks during the anticipation of reward or punishment. Our findings indicate that this functional independence is modulated differentially by those neural systems innervated by the cortico-mesolimbic and nigro-striatal dopamine projections. The segregation of evaluation and readiness neural responses from a unified input signal was also achievable using a theoretical neuronal population-coding model, hence further advancing our understanding of the neural basis involved in motivational processing. Furthermore, our novel theoretical approach could potentially be applied more generally to decode multiple latent neurobehavioral processes and thus advance both the design and hypothesis testing in cognitive task paradigms.