Two very different models of hemispheric specialization provide the starting point for this theory-driven research approach to examine the laterality of emotional learning and memory. The first is known as the valence hypothesis. It accounts for a number of observations on lateralized emotion processing, but was not designed to address issues of encoding and retrieval of affective information. The second is the HERA model, which was developed to account for a variety of findings on encoding and retrieval processes, but does not consider the affective content of the processed information. The work proposed here uses functional magnetic resonance imaging (fMRI) to test differing predictions made by these models about structures involved in the encoding and retrieval of affective information and has these two specific aims.
Aim 1 : To determine whether specific brain structures are laterally activated by the encoding or retrieval of affective stimuli. A set of three experiments is proposed in which subjects encode affective stimuli designed to elicit broadly negative or positive affect, or the basic emotions of happiness or disgust (Exp. 1 a/b), engage in an emotional encoding and retrieval task (Exp. 2 a/b), and use imagery and recall of previously viewed films to experience emotions of happiness or disgust (Exp. 3).
Aim 2 : To determine whether specific brain structures are laterally activated by affective judgment based on the retrieval of neutral stimuli (the 'mere exposure' effect). The 'mere exposure' effect refers to the phenomenon that subjects prefer stimuli to which they have been previously exposed, irrespective of their awareness for the exposure. This observation led to a debate between Zajonc and Lazarus about whether higher cognitive (i.e., cortical) processing is necessary for affective judgment. A direct test is proposed, in which fMRI is used to identify cortical and subcortical systems that are activated in the mere exposure paradigm.
