Modeling Motor Learning Using Heteroscedastic Functional Principal Components Analysis

Daniel Backenroth, Jeff Goldsmith, Michelle D. Harran, Juan C. Cortes, John Krakauer, Tomoko Kitago

Research output: Contribution to journalArticle

Abstract

We propose a novel method for estimating population-level and subject-specific effects of covariates on the variability of functional data. We extend the functional principal components analysis framework by modeling the variance of principal component scores as a function of covariates and subject-specific random effects. In a setting where principal components are largely invariant across subjects and covariate values, modeling the variance of these scores provides a flexible and interpretable way to explore factors that affect the variability of functional data. Our work is motivated by a novel dataset from an experiment assessing upper extremity motor control, and quantifies the reduction in movement variability associated with skill learning. The proposed methods can be applied broadly to understand movement variability, in settings that include motor learning, impairment due to injury or disease, and recovery. Supplementary materials for this article are available online.

Original languageEnglish (US)
Pages (from-to)1003-1015
Number of pages13
JournalJournal of the American Statistical Association
Volume113
Issue number523
DOIs
StatePublished - Jul 3 2018

Fingerprint

Functional Principal Component Analysis
Covariates
Functional Data
Principal Components
Modeling
Motor Control
Random Effects
Quantify
Recovery
Invariant
Learning
Principal component analysis
Experiment
Movement
Principal components

Keywords

  • Functional data
  • Kinematic data
  • Motor control
  • Probabilistic PCA
  • Variance modeling
  • Variational Bayes

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Modeling Motor Learning Using Heteroscedastic Functional Principal Components Analysis. / Backenroth, Daniel; Goldsmith, Jeff; Harran, Michelle D.; Cortes, Juan C.; Krakauer, John; Kitago, Tomoko.

In: Journal of the American Statistical Association, Vol. 113, No. 523, 03.07.2018, p. 1003-1015.

Research output: Contribution to journalArticle

Backenroth, Daniel ; Goldsmith, Jeff ; Harran, Michelle D. ; Cortes, Juan C. ; Krakauer, John ; Kitago, Tomoko. / Modeling Motor Learning Using Heteroscedastic Functional Principal Components Analysis. In: Journal of the American Statistical Association. 2018 ; Vol. 113, No. 523. pp. 1003-1015.
@article{c3fc3d813ad043aab8e47ff1c76e9ed0,
title = "Modeling Motor Learning Using Heteroscedastic Functional Principal Components Analysis",
abstract = "We propose a novel method for estimating population-level and subject-specific effects of covariates on the variability of functional data. We extend the functional principal components analysis framework by modeling the variance of principal component scores as a function of covariates and subject-specific random effects. In a setting where principal components are largely invariant across subjects and covariate values, modeling the variance of these scores provides a flexible and interpretable way to explore factors that affect the variability of functional data. Our work is motivated by a novel dataset from an experiment assessing upper extremity motor control, and quantifies the reduction in movement variability associated with skill learning. The proposed methods can be applied broadly to understand movement variability, in settings that include motor learning, impairment due to injury or disease, and recovery. Supplementary materials for this article are available online.",
keywords = "Functional data, Kinematic data, Motor control, Probabilistic PCA, Variance modeling, Variational Bayes",
author = "Daniel Backenroth and Jeff Goldsmith and Harran, {Michelle D.} and Cortes, {Juan C.} and John Krakauer and Tomoko Kitago",
year = "2018",
month = "7",
day = "3",
doi = "10.1080/01621459.2017.1379403",
language = "English (US)",
volume = "113",
pages = "1003--1015",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "523",

}

TY - JOUR

T1 - Modeling Motor Learning Using Heteroscedastic Functional Principal Components Analysis

AU - Backenroth, Daniel

AU - Goldsmith, Jeff

AU - Harran, Michelle D.

AU - Cortes, Juan C.

AU - Krakauer, John

AU - Kitago, Tomoko

PY - 2018/7/3

Y1 - 2018/7/3

N2 - We propose a novel method for estimating population-level and subject-specific effects of covariates on the variability of functional data. We extend the functional principal components analysis framework by modeling the variance of principal component scores as a function of covariates and subject-specific random effects. In a setting where principal components are largely invariant across subjects and covariate values, modeling the variance of these scores provides a flexible and interpretable way to explore factors that affect the variability of functional data. Our work is motivated by a novel dataset from an experiment assessing upper extremity motor control, and quantifies the reduction in movement variability associated with skill learning. The proposed methods can be applied broadly to understand movement variability, in settings that include motor learning, impairment due to injury or disease, and recovery. Supplementary materials for this article are available online.

AB - We propose a novel method for estimating population-level and subject-specific effects of covariates on the variability of functional data. We extend the functional principal components analysis framework by modeling the variance of principal component scores as a function of covariates and subject-specific random effects. In a setting where principal components are largely invariant across subjects and covariate values, modeling the variance of these scores provides a flexible and interpretable way to explore factors that affect the variability of functional data. Our work is motivated by a novel dataset from an experiment assessing upper extremity motor control, and quantifies the reduction in movement variability associated with skill learning. The proposed methods can be applied broadly to understand movement variability, in settings that include motor learning, impairment due to injury or disease, and recovery. Supplementary materials for this article are available online.

KW - Functional data

KW - Kinematic data

KW - Motor control

KW - Probabilistic PCA

KW - Variance modeling

KW - Variational Bayes

UR - http://www.scopus.com/inward/record.url?scp=85054631988&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054631988&partnerID=8YFLogxK

U2 - 10.1080/01621459.2017.1379403

DO - 10.1080/01621459.2017.1379403

M3 - Article

C2 - 30416231

AN - SCOPUS:85054631988

VL - 113

SP - 1003

EP - 1015

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 523

ER -