2015 年 1 巻 1 号 p. 87-92
In daily life, we are required to adapt our behavior continuously in situations in which much of our incoming information is emotional and unrelated to our immediate behavioral goals. Such information is processed both ‘with’ and ‘without’ our consciousness. However, it has not been clarified how such emotional factor, which is independent from reward, affects the learning process. Here, we addressed this issue with the reinforcement learning model and identified the neural substrates that underlie emotion-learning interaction by functional magnetic resonance imaging (fMRI). We observed that the emotional stimulus, represented consciously or unconsciously prior to the cue signal, enhanced the learning rate and accelerated the speed of probabilistic association learning. The fMRI results indicated that these phenomena were caused by the enhancement of reward prediction error signal by the striatum–amygdala interaction.