An Extended Interactive Evolutionary Computation Using Heart Rate Variability as Fitness Value for Composing Music Chord Progression

Interactive Evolutionary Computation (IEC) is known as an efficient method to create media content suited to the individual user. To reduce user’s fatigue, which remains as a serious problem in IEC, extended IEC that uses physiological information as a fitness value have been proposed. As a new extended IEC, this study proposed extended IEC using Heart Rate Variability (HRV), which reflects autonomic nervous activity. A High Frequency (HF) component of HRV was used as the fitness value. Two listening experiments were conducted to determine the efficacy of the proposed method. In experiment 1, with a concrete system of the proposed method creating music chord progression, a change in the fitness value was observed. In experiment 2, representative created music chord progressions were evaluated subjectively. The change in the fitness value of the HF component showed no gradual increase. Subjective evaluation results showed that the lowest fitness value was observed in the 1st generation, and the fitness value in the 10th generation significantly increased from the 1st generation ( P < 0.05). The result of the subjective evaluation showed the efficacy of the proposed method.