The General Principle of Emotion is the theory that differential emotion values may prove to be more accurate in defining emotion qualia than the experience of emotion qualia at each point in time. These Delta Emotion values infer the emotional response of a given time point (i.e. event) by capturing and analysing the emotional responses at either side of the time point, and could provide a more genuine emotion signal for predictive analysis.
Delta emotions are geared towards long-term analysis; specifically, it could be thought of as measuring shifts in perception. The General Principle of Emotion states that an emotional response of an event at time t, E(t), can be inferred by calculating the change in emotion, delta(E) = E(t+1) - E(t-1). We are currently researching into how different periods of t can affect any subsequent entropy in emotional memory, and other phenomenon such as Emotion Injection (altering 'normal' levels after an event) and Emotion Bursts (identical 'normal' levels after an event, with significant emotion within the period).
While our data is shedding light on more powerful emotion prediction techniques and more precise measurements of mood, the GPE has solid roots in psychology, and could have interesting uses outside of AI and robotic technologies. For example, once we have explored Emotion Injection in more detail, we would be able to accurately alter the perception of people by selecting their emotion response before they experience it.