Attention to aversive emotion and specific activation of the right insula and right somatosensory cortex

Abstract

The evaluation of emotional stimuli is based on different levels of information processing, ranging from rather automatic processes to focused attention to the emotional relevance of stimuli. The role of specific brain areas for these processes is a matter of debate. In this event-related fMRI study, we varied the information processing mode of participants exposed to aversive and neutral pictures. Based on four different tasks, participants' attentional focus onto the emotional quality of the stimuli and the own emotional involvement was increased systematically across tasks. Regardless of task, stronger activation to threatening vs. neutral pictures was found in several regions such as the amygdala, anterior insula, anterior cingulate cortex, primary somatosensory cortex and medial prefrontal cortex. However, there was a parametric increase of activation with increasing attention to one's own emotion specifically in the right posterior insula and right primary and secondary somatosensory cortex, i.e. in areas implicated in self-awareness of a person's own body. These findings are in accordance with theories suggesting a crucial role of the perception of bodily states for emotional experiences.

Introduction

The processing of emotional stimuli is associated with different stages of information processing. While implicit processing modes are related to rather automatic responses to emotional stimuli, explicit processing requires that subjects' attention is directed to the emotional significance of stimuli or to one's own emotional involvement, i.e. to own feelings (e.g., LeDoux, 1996, Damasio, 1999). These different levels of information processing seem to depend partially on different neural systems (e.g., LeDoux, 1996, Damasio, 1999, Phan et al., 2002). Several prefrontal areas and the anterior cingulate cortex were suggested to be involved in cognitive-emotional interactions, such as certain high level appraisal processes, as well as in the generation and regulation of emotional responses (e.g., Lane et al., 1998, Phan et al., 2002, Ochsner et al., 2004, Straube et al., 2006, Straube et al., 2009a). In contrast, subcortical areas such as the amygdala have been proposed to mediate the rapid processing of emotional, especially of threatening stimuli, given a minimum of processing resources (Vuilleumier et al., 2001, Pessoa et al., 2002, Straube et al., 2007).

Besides these areas, brain regions that are involved in the direct representation of bodily states, such as the insula and the somatosensory cortex, were also proposed to be relevant for emotion processing (Damasio, 1999; Adolphs et al., 2000; Craig, 2002, Craig, 2009). This hypothesis is based on the idea that the perception of real or simulated bodily responses is an important factor for emotional experiences (James, 1884, Damasio, 1999, Craig, 2002, Craig, 2009). While the anterior insula has been suggested to serve as an integrative center for the representation and evaluation of interoceptive information, posterior areas of the insula, and the primary and secondary somatosensory cortex are more strongly involved in general awareness of the own body and several visceral and somatosensory sensations (Craig, 2002; Adolphs et al., 2000; Karnath et al., 2005, Khalsa et al., 2009).

While several brain areas have been proposed to participate in different aspects of emotion processing, the effect of the processing mode on brain activation warrants further examinations. To elucidate the specific role of a brain region as a function of the information processing mode, it is necessary to systematically vary of the level of evaluation of emotional stimuli and to control for other factors (see also Lieberman et al., 2007). In the present event-related fMRI-study, we systematically increased subjects' attention to one's own emotional involvement by means of four different tasks. If the experience of emotion is closely linked to the perception of bodily responses, increasing attention to own feelings should be associated with increased activation of relevant brain structures, such as the insula or the somatosensory cortex, which are involved in the processing of bodily states. In contrast, those areas that are involved in other or more general processes should not show an increase of brain activation across tasks.