Event-related brain potentials and the study of reward processing: Methodological considerations.

Abstract

There is growing interest in using electroencephalography and specifically the event-related brain potential (ERP) methodology to study human reward processing. Since the discovery of the feedback related negativity (Miltner et al., 1997) and the development of theories associating the feedback related negativity and more recently the reward positivity with reinforcement learning, midbrain dopamine function, and the anterior cingulate cortex (i.e., Holroyd and Coles, 2002) researchers have used the ERP methodology to probe the neural basis of reward learning in humans. However, examination of the feedback related negativity and the reward positivity cannot be done without an understanding of some key methodological issues that must be taken into account when using ERPs and examining these ERP components. For example, even the component name - the feedback related negativity - is a source of debate within the research community as some now strongly feel that the component should be named the reward positivity (Proudfit, 2015). Here, ten key methodological issues are discussed - confusion in component naming, the reward positivity, component identification, peak quantification and the use of difference waveforms, frequency (the N200) and component contamination (the P300), the impact of feedback timing, action, and task learnability, and how learning results in changes in the amplitude of the feedback-related negativity/reward positivity. The hope here is to not provide a definitive approach for examining the feedback related negativity/reward positivity, but instead to outline the key issues that must be taken into account when examining this component to assist researchers in their study of human reward processing with the ERP methodology.