Emotion Circuits in the Brain

LeDoux, J.E. (2000). Emotion Circuits in the Brain. Annual Reviews in Neuroscience, 23, 155-184.

Emotion research was largely lost for some time in the wake of the cognitive revolution. However, people soon realized a purely cognitive view of the brain -- leaving out emotions, motivations, and the like -- is likely to paint an unrealistic view of real minds. Unfortunately, attempts to dig into emotions once again were hamstrung by the limbic system concept, a flawed and inadequate theory of the emotional brain: cognition does not only reside in the neocortex and emotions do not only reside within the limbic system (a moving target itself).

Emotion research began its official resurgence with a bottoms-up examination of fear conditioning, with a bulk of the work focused on the auditory modality. Research soon named amygdala as centrally important, a site where transmission of information about the CS and US converged and output projections controlled fear reactions. On the input side, CS sensory inputs terminate in the lateral amygdala (LA), coming from both the auditory thalamus and the auditory cortex, although plasticity seems to occur initially through the thalamic pathway. US information also seems to converge in the amygdala, receiving inputs from the spino-thalamic tract, cortical areas that process somatosensory stimuli including nociceptive stimuli, the parabrachial area, and the spinal cord. On the outbound side, the central nucleus of the amygdala (CE) projects to autonomic (hypothalamus) and defensive motoric (periaqueductal gray) centers. Methodologies used have largely been single unit recordings, long-term potentiation (LTP) studies, and pharmacological experiments which block LTP. Studies have focused on two types of fear learning: simple fear conditioning (a benign tone comes to evoke a fear response) and contextual fear conditioning (fear responsivity to environmental cues). Research agrees that the amygdala seems to be required for Pavlovian fear conditioning to occur, although the site of long-term fear memory storage is still unknown: it may very well exist in the amygdala, but it may also be distributed across multiple structures or transferred off to cortical areas over time. However, plasticity within the amygdala is probably not required for learning cognitive aspects of fear.

Human studies have echoed many of the results from animal literature. Additionally, they have found perceptual deficits of the emotional meaning of faces in patients with amygdalar damage. The amygdala also appears activated more strongly in the presence of fearful and angry faces than of happy ones. Further, when the activity of the amygdala during fear conditioning is cross-correlated with other regions of the brain, the strongest relations are seen in subcortical areas, emphasizing the importance of the direct thalamo-amygdala pathway in the human brain. Although a fear conditioning approach cannot account for all aspects of human fear and anxiety disorders, it may be especially elucidating for PTSD, panic disorders, and phobias. Difficulty in extinguishing fear memories witnessed in human disorders may also involve the medial prefrontal cortex circuitry.

Future research needs to integrate both cognition and emotion. How fear processing in the amygdala can influence perceptual, attentional, and memory functions of the cortex, and vice versa, is begging for additional research, although it is known that the amygdala does receive input from cortical sensory processing regions and projects back to these both directly and indirectly. How conscious emotional feelings are manifest is also relatively unexplored, although the models posit that feelings may arise from interactions between the amygdala and prefrontal working memory areas, sensory processing areas in cortex, long-term memory systems in the temportal lobe, and arousal systems which maintain global projections.