Meeting Abstract

P2.49  Thursday, Jan. 5  Functional data analysis of mammalian masticatory jaw movements CRANE, E.; CHILDERS, D.; ROTHMAN, E.; GERSTNER, G.E.*; Univ. of Michigan; Univ. of Michigan; Univ. of Michigan; Univ. of Michigan geger@umich.edu

Mammals are characterized by complex jaw movements associated with food ingestion and reduction. Despite the diversity of mammalian dentoskeletal forms, very few of the 5400+ identified mammalian species have been studied in terms of feeding kinematics, biomechanics and neuromotor control. One of our interests involves creating methods for quantifying oromandibular movements in ways that: (1) are minimally invasive for ethological reasons, (2) increase data analysis rates to expedite studies of additional species, (3) are rigorously quantitative and (4) allow statistically-legitimate comparative studies. Functional Data Analysis (FDA) is a set of sophisticated new methods that show great promise in these regards. Although FDA is gaining a foothold in human movement sciences, it is less well known among other disciplines and to our knowledge has only recently been applied to oral movement studies. One advantage of FDA lies in its treating time series data as functional observations, which offers the advantage of retaining information in continuous signals, e.g., masticatory jaw movements, by representing movements as functions of time rather than reducing the signal into discrete variables. Statistical analyses test hypotheses about the shape of movements, as opposed to traditional analyses based on landmarks, e.g., maximum gape, time of maximum velocity. We have been developing and testing FDA; here, we present FDA results, wherein we demonstrate: (1) automated jaw movement feature detection, e.g., gape cycles, transitions between fast and slow gape-cycle phases, etc., (2) extraction and functional rendering of gape cycles and (3) statistical tests to identify the location of significant differences in gape cycles between species.