Although autonomous driving is expected to pave the way for the future of transportation, it is often met with resistance. One of the reasons for this may be that, as of this writing, autonomous driving still cannot meet the individual needs of people. Furthermore, the unfamiliarity and discomfort when riding in an autonomous vehicle can cause drivers to feel stressed and distrustful of the vehicle. To this end, we propose an Emotion Preference Style Adaptation (EPSA) framework. The framework can analyze and determine a driver's driving preferences from the emotion which is recognized from their EEG signals. And then it will adapt the style of the vehicle's driving behavior to suit the driver's preference.