Temporal rescaling of sample and complex rations in rhythmic tapping

Two well-established phenomena in temporal performance--preference for simple ratios over complex ratios and the ability to proportionately rescale temporal patterns--were examined together. Unlike the case with simple ratios, participants (3 trained musicians) showed only a limited ability to learn complex ratios and no ability to proportionately rescale them. These differences suggest that different mechanisms are used to produce the 2 ratio types. Systematic biases observed in the performances of the simple ratios (6 trained musicians) were modeled by assuming that an additive timing component, which is unequal for short and long intervals and changes with base rate, overlays a system that preferentially produces simple ratios. The general pattern of results was consistent with the view that rhythmic learning builds on or modifies a system of innate preferences.