What is 'musical ability' and how do we measure it?

Abstract

There is little consensus on what exactly constitutes musical ability and how to best measure it. Past research has used various tasks; most commonly assessing perceptual skills (e.g., same/different judgments in sequentially presented melodies), but also sometimes production tasks (e.g., singing a series of pitches or tapping along with a musical sequence). Outcome measures have ranged from single indices (e.g., “pitch ability”) to composite scores from multiple tasks (e.g., pitch, rhythm, loudness, timbre, etc.). To date, it remains unclear how these different measures/scores relate to one another, limiting the ability to generalize across tasks and results. To address these issues, we assessed 165 participants’ performance on 15 representative musical ability tasks to model the unity and diversity of musical abilities. Latent variable model comparisons suggest that musical ability is best represented by related but separable pitch, timing, perception, and production factors.