Unravelling individual rhythmic abilities using machine learning

Abstract

Humans can easily extract the rhythm of a complex sound, like music, and move to its regular beat, for example in dance. These abilities are modulated by musical training and vary significantly in untrained individuals. The causes of this variability are multidimensional and typically hard to grasp with single tasks. To date we lack a comprehensive model capturing the rhythmic fingerprints of both musicians and non-musicians. Here we harnessed machine learning to extract a parsimonious model of rhythmic abilities, based on the behavioral testing (with perceptual and motor tasks) of individuals with and without formal musical training (n = 79). We demonstrate that the variability of rhythmic abilities, and their link with formal and informal music experience, can be successfully captured by profiles including a minimal set of behavioral measures. These profiles can shed light on individual variability in healthy and clinical populations, and provide guidelines for personalizing rhythm-based interventions.

Mobile version of the Battery for the Assessment of Auditory Sensorimotor and Timing Abilities (BAASTA): Implementation and adult norms

“…Notably, the aforementioned tasks showed excellent test-retest reliability (BAT and paced tapping with music) and are also those found to capture most variability in rhythmic capacities across different BAASTA tests as shown in a recent machine-learning study (Dalla Bella et al, 2023). This finding supports the choice of the two tasks to compute a composite scorethe BTIas a global measure of rhythmic abilities (see also, Puyjarinet et al, 2017 for a first presentation of the BTI).…”

“…This complexity is difficult to capture through isolated tasks and instead requires multiple assessments, as provided by comprehensive test batteries. Multiple tests afford precise measurement of timing and rhythm abilities, and the detection of individual profiles (Dalla Bella et al, 2023), allowing isolation of the locus of impairment in individuals with rhythm disorders. Existing batteries for testing timing and rhythm abilities include both perceptual and sensorimotor tasks.…”

“…The battery is sensitive to individual differences in timing and rhythm abilities. It can distinguish between individuals with and without musical training or informal music experience (Dalla Bella et al, 2023), as well as impaired rhythm perception and performance in healthy and patient populations (e.g., Bégel et al, 2017Bégel et al, , 2022Benoit et al, 2014;Dalla Bella et al, 2023;Falk et al, 2015;Fiveash et al, 2022;Hidalgo et al, 2021;Oschkinat et al, 2022;Puyjarinet et al, 2017Verga et al, 2021;Zagala et al, 2021). Hence, the battery can serve as a screening tool for rhythm disorders, and pave the way to individualized rhythm-based interventions (Dalla Bella et al, 2018).…”

The Ramp protocol: Uncovering individual differences in walking to an auditory beat using TeensyStep

“…These studies have demonstrated significant individual differences in rhythm perception and rhythm production tasks, both with and without an external stimulus (Dalla Bella et al, 2023;Fiveash et al, 2022;Kragness et al, 2021;Palmer et al, 2014;Rajan et al, 2019;Sowiński & Dalla Bella, 2013;Tierney et al, 2017;Tranchant et al, 2016). Rhythmic abilities fall on a continuum spanning from experts (Dalla Bella et al, 2024), such as professional musicians, to individuals showing poor rhythmic abilities, such as beat deafness or poor synchronization (Bégel et al, 2017;Palmer et al, 2014;Phillips-Silver et al, 2011;Sowiński & Dalla Bella, 2013;Tranchant et al, 2021;Tranchant & Peretz, 2020). At the low end of the continuum, individuals can be impaired specifically in rhythm production (Sowiński & Dalla Bella, 2013), in rhythm perception (Bégel et al, 2017), or both (Palmer et al, 2014;Phillips-Silver et al, 2011).…”

“…At the low end of the continuum, individuals can be impaired specifically in rhythm production (Sowiński & Dalla Bella, 2013), in rhythm perception (Bégel et al, 2017), or both (Palmer et al, 2014;Phillips-Silver et al, 2011). Thus, measuring both rhythm perception and motor production is paramount to account for individual differences in rhythmic abilities (Dalla Bella et al, 2023, 2024.…”