Seminar Series

Does a Fast, Fear-Related, Subcortical Visual Pathway to the Amygdala exist? Counter-evidences from Affective Blindsight and Monkey’s LFPs

It has been suggested that fear-related stimuli such as fearful or threatening facial expressions are initially processed via a fast, subcortical pathway involving the amygdala. A subcortical pathway to the amygdala has been well-documented in rodents, but its existence and relevance for the processing of facial expressions in human and non-human primates remains controversial. Intracranial recordings in epileptic patients show a relatively late contribution of the amygdala (200ms) to the evaluation of emotionally laden faces. On the other hand, there is evidence on amygdala involvement in the residual processing of affective faces in patients suffering from affective blindsight. This has been taken as an evidence of the early involvement of the amygdala in the processing of fear related signals. During this talk we will present results from analysis of EEG recordings carried out in a patient suffering from affective blindsight as well as direct LFP recordings in monkey’s amygdala. During recordings, the monkeys passively viewed different categories of images that included humans, food, monkey perinea, and monkey faces displaying threatening, appeasing, neutral, and fearful facial expressions. For the facial expression subset of our recordings, differences in evoked LFP alterations across facial expression groups were evaluated over time using a non-parametric ANOVA (P<.01) and multiple comparison post hoc tests using Tukey's honestly significant difference criterion. Evoked changes in LFP amplitude were not found to be specific to any facial expression until at least 100ms after the image presentation. This timing does not suggest a fast pathway or necessitate an entirely subcortical process. Furthermore, these differences were not specific to threatening or fearful faces but mainly to gaze direction. In the blindsight patient we used multivariate pattern recognition (MPR) of source localized EEG recordings was used to evaluate the consistent oscillatory patterns induced by facial expressions. All faces elicited distinctive oscillatory EEG patterns that were correctly identified by the MPR algorithm as belonging to the class of facial expression actually presented. Earliest non-specific differences were detected at superior temporal sulcus (STS) followed by prefrontal areas. Amygdala activation was observed only later, at around after 200ms after emotional face presentation. These results suggest that affective blindsight might activate the amygdala indirectly rather than through a direct subcortical pathway as previously thought. Prefrontal activity specific to emotional faces might explain behavior in forced choice tasks, although neither extra-striate visual areas nor prefrontal cortex activation are sufficient for conscious perception. Consequently, our results do not support the involvement of the right amygdala in a fast, fear-related visual pathway.