Parameter Estimation of Linear Sensorimotor Synchronization Models: Phase Correction, Period Correction, and Ensemble Synchronization

in Timing & Time Perception
Restricted Access
Get Access to Full Text
Rent on DeepDyve

Have an Access Token?



Enter your access token to activate and access content online.

Please login and go to your personal user account to enter your access token.



Help

Have Institutional Access?



Access content through your institution. Any other coaching guidance?



Connect

Linear models have been used in several contexts to study the mechanisms that underpin sensorimotor synchronization. Given that their parameters are often linked to psychological processes such as phase correction and period correction, the fit of the parameters to experimental data is an important practical question. We present a unified method for parameter estimation of linear sensorimotor synchronization models that extends available techniques and enhances their usability. This method enables reliable and efficient analysis of experimental data for single subject and multi-person synchronization. In a previous paper (Jacoby et al., 2015), we showed how to significantly reduce the estimation error and eliminate the bias of parameter estimation methods by adding a simple and empirically justified constraint on the parameter space. By applying this constraint in conjunction with the tools of matrix algebra, we here develop a novel method for estimating the parameters of most linear models described in the literature. Through extensive simulations, we demonstrate that our method reliably and efficiently recovers the parameters of two influential linear models: Vorberg and Wing (1996), and Schulze et al. (2005), together with their multi-person generalization to ensemble synchronization. We discuss how our method can be applied to include the study of individual differences in sensorimotor synchronization ability, for example, in clinical populations and ensemble musicians.

Parameter Estimation of Linear Sensorimotor Synchronization Models: Phase Correction, Period Correction, and Ensemble Synchronization

in Timing & Time Perception

Sections

References

AitkenA. C. (1935). On least squares and linear combinations of observations. Proc. R. Soc. Edinburgh554248.

BoydS. P. & VandenbergheL. (2004). Convex optimization. Cambridge, UK: Cambridge University Press.

ButterfieldM. (2010). Participatory discrepancies and the perception of beats in jazz. Music Percept. 27157176.

ChenY.ReppB. H. & PatelA. D. (2002). Spectral decomposition of variability in synchronization and continuation tapping: Comparisons between auditory and visual pacing and feedback conditions. Hum. Movement Sci. 21515532.

CramérH. (1999). Mathematical methods of statistics. Princeton, NJ, USA: University Press.

DarabiN.SvenssonP. & FarnerS. (2008). Quantifying the strategy taken by a pair of ensemble hand-clappers under the influence of delay. Paper presented at the 125th Audio Engineering Society Convention San FranciscoCA, USA.

Deligniè resD.TorreK. & LemoineL. (2009). Long-range correlation in synchronization and syncopation tapping: A linear phase correction model. PloS One11e7822. DOI: 10.1371/journal.pone.0007822.

DesainP. & HoningH. (1994). Does expressive timing in music performance scale proportionally with tempo? Psychol. Res. 56285292.

DingM.ChenY. & KelsoJ. A. (2002). Statistical analysis of timing errors. Brain Cogn. 4898106.

FairhurstM. T.JanataP. & KellerP. E. (2014). Leading the follower: An fMRI investigation of dynamic cooperativity and leader–follower strategies in synchronization with an adaptive virtual partner. NeuroImage84688697.

FellerW. (2008). An introduction to probability theory and its applications. New York, NY, USA: John Wiley & Sons.

GoeblW. & PalmerC. (2009). Synchronization of timing and motion among performing musicians. Music Percept. 26427438.

HaryD. & MooreG. (1987a). On the performance and stability of human metronome — Synchronization strategies. Br. J. Math. Stat. Psychol. 40109124.

HaryD. & MooreG. (1987b). Synchronizing human movement with an external clock source. Biol. Cybern. 56305311.

JacobyN. & ReppB. H. (2012). A general linear framework for the comparison and evaluation of models of sensorimotor synchronization. Biol. Cybern. 106135154.

JacobyN.KellerP. E.ReppB. H.AhissarM. & TishbyN. (2015). Lower bound on the accuracy of parameter estimation methods for linear sensorimotor synchronization models. Timing Time Percept. 3XXYY.

JanataP.TomicS. T. & HabermanJ. M. (2012). Sensorimotor coupling in music and the psychology of the groove. J. Exp. Psychol. Gen. 1415475.

KellerP. E. (2008). Joint action in music performance. Emerg. Commun. 10205221.

KellerP. E. (2014). Ensemble performance: Interpersonal alignment of musical expression. In FabianD.TimmersR. & SchubertE. (Eds) Expressiveness in music performance: Empirical approaches across styles and cultures (pp. 260282). Oxford, UK: Oxford University Press.

KellerP. E. & AppelM. (2010). Individual differences, auditory imagery, and the coordination of body movements and sounds in musical ensembles. Music Percept.282746.

KellerP. E. & ReppB. H. (2005). Staying offbeat: Sensorimotor syncopation with structured and unstructured auditory sequences. Psychol. Res. 69292309.

KonvalinkaI.VuustP.RoepstorffA. & FrithC. D. (2010). Follow you, follow me: Continuous mutual prediction and adaptation in joint tapping. Q. J. Exp. Psychol. 6322202230.

LargeE. W.FinkP. & KelsoS. J. (2002). Tracking simple and complex sequences. Psychol. Res. 66317.

LargeE. W. & JonesM. R. (1999). The dynamics of attending: How people track time-varying events. Psychol. Rev. 106119159.

LawsonC. L. & HansonR. J. (1974). Solving least squares problems. Englewood Cliffs, NJ, USA: Prentice Hall.

LiS.LewandowskyS. & DeBrunnerV. E. (1996). Using parameter sensitivity and interdependence to predict model scope and falsifiability. J. Exp. Psychol. Gen. 125360369.

LjungL. (1998). System identification. Berlin, Germany: Springer.

LoehrJ. D. & PalmerC. (2011). Temporal coordination between performing musicians. Q. J. Exp. Psychol. 6421532167.

LondonJ. (2012). Hearing in time. Oxford, UK: Oxford University Press.

MadisonG. & MerkerB. (2004). Human sensorimotor tracking of continuous subliminal deviations from isochrony. Neurosci. Lett. 3706973.

MarchiniM.PapiotisP. & MaestreE. (2012). Timing synchronization in string quartet performance: A preliminary study. Paper presented at the 9th International Symposium on Computer Music Modelling and Retrieval (CMMR12)London, UK.

MatesJ. (1994a). A model of synchronization of motor acts to a stimulus sequence. Biol. Cybern. 70463473.

MatesJ. (1994b). A model of synchronization of motor acts to a stimulus sequence. II. Stability analysis, error estimation and simulations. Biol. Cybern. 70475484.

McAuleyJ. D. & JonesM. R. (2003). Modeling effects of rhythmic context on perceived duration: A comparison of interval and entrainment approaches to short-interval timing. J. Exp. Psychol. Hum. Percept. Perform. 2911021125.

MerkerB. H.MadisonG. S. & EckerdalP. (2009). On the role and origin of isochrony in human rhythmic entrainment. Cortex45417.

MichonJ. (1967). Timing in temporal tracking. Soesterberg, The Netherlands: Institute for Perception RVO-TNO.

MooreG. P. & ChenJ. (2010). Timings and interactions of skilled musicians. Biol. Cybern. 103401414.

NowickiL.PrinzW.GrosjeanM.ReppB. H. & KellerP. E. (2013). Mutual adaptive timing in interpersonal action synchronization. Psychomusicol. Music Mind Brain23620.

PearsonK. (1896). Mathematical contributions to the theory of evolution. III. Regression, heredity and panmixia. Philos. Trans. R. Soc. Lond. 187253318.

PressingJ. (1998a). Error correction processes in temporal pattern production. J. Math. Psychol. 4263101.

PressingJ. (1998b). Referential behavior theory: A framework for multiple perspectives on motor control. In PiekJ. P. (Ed.) Motor behavior and human skill: A multidisciplinary perspective (pp. 357384). Champaign, IL, USA: Human Kinetics.

PressingJ. & Jolley-RogersG. (1997). Spectral properties of human cognition and skill. Biol. Cybern. 76339347.

RaoC. R. (1992). Information and the accuracy attainable in the estimation of statistical parameters. In KotzS. & JohnsonN. L. (Eds) Breakthroughs in statistics (pp. 235247). New York, NY, USA: Springer.

RaschR. A. (1988). Timing and synchronization in ensemble performance. In SlobodaJ. A. (Ed.) Generative processes in music: The psychology of performance improvisation and composition (pp. 7090). Oxford, UK: Clarendon Press.

ReppB. H. (1990). Patterns of expressive timing in performances of a Beethoven minuet by nineteen famous pianists. J. Acoust. Soc. Am. 88622641.

ReppB. H. (1995). Quantitative effects of global tempo on expressive timing in music performance: Some perceptual evidence. Music Percept. 133957.

ReppB. H. (2001). Phase correction, phase resetting, and phase shifts after subliminal timing perturbations in sensorimotor synchronization. J. Exp. Psychol. Hum. Percept. Perform. 27600621.

ReppB. H. (2005). Sensorimotor synchronization: A review of the tapping literature. Psychonom. Bull. Rev.12969992.

ReppB. H. (2011). Tapping in synchrony with a perturbed metronome: The phase correction response to small and large phase shifts as a function of tempo. J. Motor Behav. 43213227.

ReppB. H. & BruttomessoM. (2009). A filled duration illusion in music: Effects of metrical subdivision on the perception and production of beat tempo. Adv. Cogn. Psychol. 5114134.

ReppB. H. & KellerP. E. (2004). Adaptation to tempo changes in sensorimotor synchronization: Effects of intention, attention, and awareness. Q. J. Exp. Psychol. A57499521.

ReppB. H. & KellerP. E. (2008). Sensorimotor synchronization with adaptively timed sequences. Hum Movement Sci. 27423456.

ReppB. H. & SuY. (2013). Sensorimotor synchronization: A review of recent research (2006–2012). Psychonom. Bull. Rev. 20403452.

ReppB.KellerP. & JacobyN. (2012). Quantifying phase correction in sensorimotor synchronization: Empirical comparison of three paradigms. Acta Psychol. 139281290.

SchulzeH. & VorbergD. (2002). Linear phase correction models for synchronization: Parameter identification and estimation of parameters. Brain Cogn. 488097.

SchulzeH.CordesA. & VorbergD. (2005). Keeping synchrony while tempo changes: Accelerando and ritardando. Music Percept. 22461477.

SemjenA.VorbergD. & SchulzeH.-H. (1998). Getting synchronized with the metronome: Comparisons between phase and period correction. Psychol. Res. 614455.

ShafferL. H. (1984). Timing in solo and duet piano performances. Q. J. Exp. Psychol. 36577595.

SnyderJ. S.HannonE. E.LargeE. W. & ChristiansenM. H. (2006). Synchronization and continuation tapping to complex meters. Music Percept. 24135146.

StrangG. (2006). Linear algebra and its applications. Belmont, CA, USA: Thomson Brooks Cole / Boston MA, USA: Cengage Learning.

ThautM. H.MillerR. A. & SchauerL. M. (1998). Multiple synchronization strategies in rhythmic sensorimotor tasks: Phase vs period correction. Biol. Cybern. 79241250.

TorreK. & BalasubramaniamR. (2009). Two different processes for sensorimotor synchronization in continuous and discontinuous rhythmic movements. Exp. Brain Res. 199157166.

TorreK.BalasubramaniamR. & Deligniè resD. (2010). Oscillating in synchrony with a metronome: Serial dependence, limit cycle dynamics, and modeling. Motor Contr.. 14323343.

TorreK. & Deligniè resD. (2008a). Distinct ways of timing movements in bimanual coordination tasks: Contribution of serial correlation analysis and implications for modeling. Acta Psychol. 129284296.

TorreK. & Deligniè resD. (2008b). Unraveling the finding of 1/f β noise in self-paced and synchronized tapping: A unifying mechanistic model. Biol. Cybern. 99159170.

VorbergD. (2005). Synchronization in duet performance: Testing the two-person phase error correction model. Tenth Rhythm Perception and Production Workshop Alden Biesen Belgium.

VorbergD. & HambuchR. (1984). Timing of two-handed rhythmic performance. Ann. N. Y. Acad. Sci. 423390406.

VorbergD. & SchulzeH. (2002). Linear phase-correction in synchronization: Predictions, parameter estimation, and simulations. J. Math. Psychol. 465687.

VorbergD. & SchulzeH. H. (2013). Modeling synchronization in musical ensemble playing: Parameter estimation and sensitivity to assumptions. Fourteenth Rhythm Production and Perception Workshop University of BirminghamBirmingham, UK.

VorbergD. & WingA. (1996). Modeling variability and dependence in timing. In HeuerH. & KeeleS.W. (Eds) Handbook of perception and action Vol. 2 (pp. 181262). London, UK: Academic Press.

WingA. M. (2002). Voluntary timing and brain function: An information processing approach. Brain Cogn. 48730.

WingA. M. & KristoffersonA. B. (1973). Response delays and the timing of discrete motor responses. Percept. Psychophys. 14512.

WingA.DaffertshoferA. & PressingJ. (2004). Multiple time scales in serial production of force: A tutorial on power spectral analysis of motor variability. Hum. Movement Sci. 23569590.

WingA. M.EndoS.BradburyA. & VorbergD. (2014). Optimal feedback correction in string quartet synchronization. J. R. Soc. Interface111125. DOI: 10.1098/rsif.2013.1125.

2

Figure 1 of Jacoby et al. (2015) provides an explanation why constraining the parameter space could dramatically reduce estimation errors. A detailed analysis of this issue can be found in the classical works of Cramér and Rao (Cramér 1999 originally 1946; Rao 1992 originally 1945). Intuitively the interdependence comes from points in the parameter space that do not satisfy eqn. (4). When these points are not considered for parameter estimation the problem disappears and the original method performs well for all cases. This is explained in detail in Jacoby et al. (2015).

Figures

  • View in gallery

    OkR and OkS are the onsets of the response and stimulus at beat k. Sk=OkSOk1S and Rk=OkROk1R are the inter-stimulus and inter-response intervals, respectively. Ak=OkROkS is the asynchrony. This figure is published in color in the online version.

  • View in gallery

    Comparison of the unbounded acvf method [circles] and the unbounded Generalized Least Squares method [thin line]. The correct underlying parameters are plotted as the dashed line. The figure shows the mean estimates based on 100 simulated renditions of the phase correction model of eqn. (3), for different α values (x-axis) with parameters σT2=100 , σM2=25 , nseq = 15, and N = 30. Error bars represent the standard deviation of the estimates. Both methods show extremly large estimation errors for large α values. This demonstrates the necessity of using further constraints in order to obtain a reliable estimates. This figure is published in color in the online version.

  • View in gallery

    The bGLS method with a non-isochronous sequence. Ideal performance is marked in dashed line. We simulated 1000 iteration for different α values (x-axis) with parameters ( σT2=100 and σM2=25 , nseq = 15, N = 30). The top, middle, and bottom graphs show the mean and standard deviation of the estimates for phase correction, timekeeper variance, and motor variance, respectively. This figure is published in color in the online version.

  • View in gallery

    Comparison of mean and standard deviation of the estimations based on motor noise and least square estimation (LS) that ignores motor noise. Motor variance was fixed ( σM2=25 ), and the different lines present different σT2=100,300,500 timekeeper variance. In all simulations nseq = 15 and N = 30. This figure is published in color in the online version.

  • View in gallery

    Schematic illustration of onsets for synchronizing in multi-person cases. We denote the onsets of the responses of the current synchronizer as OkR . We denote by OkSi the onsets of the P other partners. The asynchronies between response and partner i are given by Aki . This figure is published in color in the online version.

  • View in gallery

    Simulated string quartet with a setting similar to the data in Wing et al. (2014); namely, the same number of synchronizers and the same number of trials and blocks. The top graph shows the phase constants of all four simulated players. The bottom graph shows the different timekeeper standard deviations (σT = 10, 13, 15, and 18 for violin 1, violin 2, viola, and cello, respectively) and motor variances (σM = 5, 6, 7, 8). Estimation of 100 iterations of the simulations with nseq = 16, and N = 40 are depicted. Error bars represents standard deviation of the estimation error. This figure is published in color in the online version.

  • View in gallery

    Reanalyis of real string quartets using the data from Wing et al. (2014). The top figures show the estimated coupling constants for quartet 1 (left) and quartet 2 (right). The bottom figures show the estimated timekeeper and motor variance for each of the players. This figure is published in color in the online version.

  • View in gallery

    Simulation results for the period correction experiment in Jacoby and Repp (2012). Note that because of the high accuracy of the method there is a large overlap between the simulated results. (a): mean and standard deviation of the estimated α based on 1000 simulations for the three possible β values and six α values. (b): mean and standard deviation of the estimated β based on 1000 simulations for the six possible α values and three β values. (c): mean and standard deviation of the estimated σT based on 1000 simulations. (d): mean and standard deviation of the estimated σM based on 1000 simulations. This figure is published in color in the online version.

  • View in gallery

    Simulated string quartet with period and phase correction. The top graphs show the phase and period constants of all four simulated players. The bottom graphs show the different timekeeper standard deviations (σT =10, 13, 15, and 18 for violin 1, violin 2, viola, and cello, respectively) and motor variances (σM = 5, 6, 7, 8). Estimation of 1000 iterations of simulations with nseq = 16, and N = 40 are depicted in circles; error bars show the standard deviation of the estimates. Note the small bias in the estimation of the timekeeper, probably due a lack of sufficient data points in each of the simulated datasets. This figure is published in color in the online version.

Information

Content Metrics

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 35 35 12
Full Text Views 93 93 55
PDF Downloads 3 3 1
EPUB Downloads 0 0 0