The somatic marker hypothesis proposes that both the amygdala and the orbitofrontal cortex are parts of a neural circuit critical for judgment and decision-making. Although both structures couple exteroceptive sensory information with interoceptive information concerning somatic/emotional states, they do so at different levels, thus making different contributions to the process. We define "primary inducers" as stimuli that unconditionally, or through learning (e.g., conditioning and semantic knowledge), can (perceptually or subliminally) produce states that are pleasurable or aversive. Encountering a fear object (e.g., a snake), a stimulus predictive of a snake, or semantic information such as winning or losing a large sum of money are all examples of primary inducers. "Secondary inducers" are entities generated by the recall of a personal or hypothetical emotional event or perceiving a primary inducer that generates "thoughts" and "memories" about the inducer, all of which, when they are brought to memory, elicit a somatic state. The episodic memory of encountering a snake, losing a large sum of money, imagining the gain of a large sum of money, or hearing or looking at primary inducers that bring to memory "thoughts" pertaining to an emotional event are all examples of secondary inducers. We present evidence in support of the hypothesis that the amygdala is a critical substrate in the neural system necessary for triggering somatic states from primary inducers. The ventromedial cortex is a critical substrate in the neural system necessary for the triggering of somatic states from secondary inducers. The amygdala system is a priori a necessary step for the normal development of the orbitofrontal system for triggering somatic states from secondary inducers. However, once this orbitofrontal system is developed, the induction of somatic states by secondary inducers via the orbitofrontal system is less dependent on the amygdala system. Perhaps the amygdala is equivalent to the hippocampus with regard to emotions, that is, necessary for acquiring new emotional attributes (anterograde emotions), but not for retrieving old emotional attributes (retrograde emotions). Given the numerous lesion and functional neuroimaging studies illustrating the involvement of the amygdala in complex cognitive and behavioral functions, including "social cognition," we suggest that this involvement is a manifestation of a more fundamental function mediated by the amygdala, which is to couple stimuli/entities with their emotional attributes, that is, the processing of somatic states from primary inducers.