Driving is a sensitive task that is strongly affected by the driver’s emotions. Negative emotions, such as anger, can evidently lead to more driving errors. In this work, we introduce a concept of detecting and influencing driver emotions using psycho-physiological sensing for emotion classification and ambient light for feedback. We detect arousal and valence of emotional responses from wearable bio-electric sensors, namely brain-computer interfaces and heart rate sensors. We evaluated our concept in a static driving simulator with a fully equipped car with 12 participants. Before the rides, we elicit negative emotions and evaluate driving performance and physiological data while driving under stressful conditions. We use three ambient lighting conditions (no light, blue, orange). Using a subject-dependent random forests classifier with 40 features collected from physiological data we achieve an average accuracy of $78.9$% for classifying valence and $68.7$% for arousal. Driving performance was enhanced in conditions where ambient lighting was introduced. Both blue and orange light helped drivers to improve lane keeping. We discuss insights from our study and provide design recommendations for designing emotion sensing and feedback systems in the car.