Perceived tempo change is dependent on base tempo and direction of change: Evidence for a generalized version of Schulze's (1978) internal beat model

Assumptions of a theory, still influential, of tempo perception by Schulze (1978) are reconsidered in view of counter-evidence from other time-perception and timing studies. Specifically, it is shown that the role of base tempo of stimuli and the orientation of tempo change (acceleration versus deceleration) have to be incorporated in them. Therefore, Schulze's assumptions were extended so as to include the two stimulus parameters. A generalized version of Schulze's internal beat model, one of three formal elaborations of the theory, was tested in an experiment which required subjects (N = 17) to tell the direction of change (acceleration or deceleration) in metronomic tone sequences presented at tempi of 1, 2, or 4 Hz, the amount of tempo change being varied in steps of 2% between –10% (acceleration) and + 10%, including zero change. Psychometric curves obtained from logistic regression analysis showed the limitation of the original assumptions and supported the proposed generalization of Schulze's model. The validity of the theoretical interpretation of the data was discussed with respect to some methodological constraints in the experiment, such as the presentation of the different tempo conditions and the time length of the metronomic stimuli used, and with respect to their applicability to the performance of expressive drift (rubato versus accelerando) in music.