Musical structure operates across multiple nested timescales from microsound through the perceptual present to formal sections

Perception and memory impose different time constants on musical experience. At the smallest scale (below 50 ms), individual sound grains are fused into texture. The perceptual present: the window of events that feel simultaneous: is roughly 2 to 3 seconds. Short-term auditory memory spans approximately 8 to 10 seconds and is where rhythmic grouping and local melodic patterns are tracked. Beyond that, long-term memory handles structural forms, themes, and narrative arcs. Composers working algorithmically must address structure at all these levels or risk producing event-level richness with formal incoherence.

Examples

In Strudel and TidalCycles, nested patterns address multiple timescales simultaneously. In SuperCollider Patterns, Pbind handles note-level events while Ppar handles phrase-level.

Assessment

At what timescale does a listener perceive individual events as simultaneous rather than successive? Why must an algorithmic composer address multiple timescales rather than just note-level selection?