Interactions between emotions and decision making: a SEEG study with focus on the role of the cingulate cortex

Introduction Depending on their content, negative or positive, emotions can promote or slow down the cognitive functions. The inhibiting influence of negative affects on the cognitive control has been attributed to the reciprocal inhibition of the neuronal systems implicated in emotion and cognitive functions, but the mechanisms and the temporal process are not well established. The anterior cingulate cortex has been classically subdivided in two major regions with distinct functions: a dorsal-cognitive division and a rostral–ventral affective division. Recent evidence, however, do not support this traditional differentiation but indicate that both subdivisions of the anterior cingulate make important contributions to emotional processing. Aim of the study The project consisted on performing an analysis of the cerebral activity (mainly in the gamma band: > 30Hz) during a protocol of alternating tasks which necessitated the taking of simples decisions, either within an emotionally neutral and within an emotionally negative context, with the aim (i) to compare the performance in an emotionally-negative condition with the performance in an emotionally-neutral condition, (ii) to search the electrophysiological responses of the cerebral areas implanted, implicated in the cognitive and in the emotional control in the different conditions, (iii) to compare the frequency modulations of the different subregions of the cingulate cortex implanted in the different conditions. Patients and methods The study was realized with 6 adult patients (female/male: 3/3) undergoing invasive presurgical evaluation for epilepsy surgery by stereo-electroencephalography (SEEG) in the Claudio Munari Epilepsy Surgery Center, Niguarda Hospital. Monopolar SEEG data were converted to the format of the software Elpho-SEEG developed in LabView, in the Besta Institute. A total of 1365 contacts were analysed. Contacts exploring lesional areas and those showing ictal paroxystic abnormalities were excluded. Frequency changes between the range 1-150 Hz were included in the study. Results In 3 patients reaction time was not associated with the emotional content of the pictures but rather with the complexity of the Raven matrices; furthermore the mean reaction time, as well as the total number of erroneous responses, progressively decreased from the first to the last task, indicating learning mechanisms/adaptability, independently of the emotional valence of the associated pictures. On the other hand, 3 patients did more errors and showed a slightly faster or slower reaction time in the Raven trial in the negative condition compared with the other two ones. Analysis of frequency modulations during the exposure (300msec) to the negative images compared with the affectively neutral conditions, showed a power increase in the anterior aspect of midcingulate cortex, the pregenual aspect of anterior cingulated cortex, the short gyri of insula, the supplementary motor area, the frontal antero-mesial cortex as well as the temporal pole. Interestingly, only the short gyri of insula (anterior insular cortex), but not the long gyri, were activated during the exposure to the negative pictures. Our results are in line with the existing human imaging studies focusing on empathy for other's aversive events that have highlighted the role of these regions involved in the direct pain experience. Furthermore, we provide evidence that dorsal cingulate codes emotional processing. Moreover, our results support the suggestion that "second-hand" experience of pain follows only an anterior activation pattern of the insular cortex.