2
Background
Internal Clocks and the Universalist Assumption
                         Two models dominate cognitive research on musical timing. Scalar Expectancy Theory (SET) proposes a pacemaker-accumulator system where pulses are generated, accumulated, and compared to stored references, predicting Weber's law conformity wherein timing variability scales proportionally with interval length (Gibbon, Church, & Meck, 1984, 52-60). Dynamic Attending Theory (DAT) proposes that attention operates via internal oscillators entraining to external regularities, with expectancy peaks at anticipated event times (Jones & Boltz, 1989, 459-470; Large & Jones, 1999, 119-130). Neural implementations ground these models in oscillatory dynamics: Large and Palmer's (2002) resonance model demonstrates that coupled oscillator networks account for meter perception and tempo flexibility (Large & Palmer, 2002, 527-535), while striatal beat-frequency models localize timing to cortico-basal ganglia circuits (Grahn & Brett, 2007, 893-900).


Wang and Wöllner's (2020) review concludes that intrinsic models apply primarily to subsecond timing while central clock models explain the suprasecond range, but neither fully accounts for music's effects on subjective time (Wang & Wöllner, 2020, pp. 4-14). Crucially, neither model explicitly theorizes cultural variation. Culture appears implicitly as parameter modulation, meaning that it causes arousal affecting pacemaker rate, or influences oscillator frequencies shaped by training, but has no formal status as variable. The assumption is that timing mechanisms are biologically universal, with cultural experience providing inputs these mechanisms process neutrally. 


                   This assumption reflects broader debates in cognitive science. London's (2012) Hearing in Time argues that metric well-formedness constraints are cognitive universals, with different musics instantiating different surface patterns within universal structural limits (London, 2012, 15-25). Savage et al.'s (2015) statistical universals project similarly posits shared deep structure beneath surface diversity (Savage et al., 2015, 1-8). The statistical learning literature reinforces this view: Saffran, Aslin, and Newport's (1996) demonstration that infants extract transitional probabilities from auditory input suggests universal learning mechanisms operating over variable cultural input (Saffran, Aslin, & Newport, 1996, 1926-1928). Against this, Agawu (2003) argues that concepts like "African rhythm" function as Western scholarly impositions rather than indigenous categories (Agawu, 2003, 55-62), suggesting universalist frameworks may encode cultural bias as scientific neutrality.


Cross-cultural evidence has accumulated rapidly. Jacoby et al.'s (2024) study of 923 participants across 39 groups found both universal features, such as discrete rhythm categories, biases toward small-integer ratios present in all groups, and cultural specificity: Malian drummers exhibited 7:2:3 rhythm priors rare elsewhere (Jacoby et al., 2024, 4-8). Hannon and Trehub (2005) demonstrated developmental perceptual narrowing: six-month-olds detect metric violations equally in Western isochronous and Balkan non-isochronous meters, but twelve-month-olds show culture-specific patterns (Hannon & Trehub, 2005, 12639-12642). Trainor's (2005) review extends this to pitch and timbre, suggesting domain-general sensitive periods (Trainor, 2005, 262-268).


     Ethnomusicological research documents temporal systems challenging Western assumptions. Balinese gamelan organizes time through cyclic punctuation with tempo flexibility (majalan) fundamentally different from Western rubato (McGraw, 2008, 38-44). Indian tala involves cycles up to 128 beats with single beats lasting six seconds (Clayton, 2000, 15-23). West African drumming employs non-isochronous subdivisions achieving precision equal to isochronous ones (Polak & London, 2014, 533-538).

←2