An investigation into the validation of pedometers to detect foreleg steps in horses (Equus caballus) at walk and trot
MetadataShow full item record
Background: Current research, surrounding motor laterality at a population level in horses, indicates that in order to stand with the left forelimb in advance of the other, it is achieved by taking a greater amount of steps with the left foreleg than the right foreleg (McGreevy & Thomson 2006; McGreevy et al. 2007); suggesting that pedometers could effectively estimate asymmetry of forelimb locomotion in grazing horses. This novel method of detecting forelimb preference also reduces the likelihood of operator influence and provides an inexpensive objective measure of vertical movement which has the advantage of scoring large sample sizes, avoids single-day anomalies by recording over multiple days and overcomes logistical challenges (Vincent & Sidman 2003; Chan et al. 2005; Silva et al. 2010; Warren-Smith & McGreevy 2010). However in order to determine the reliability and validity of this novel measure the relationship between pedometer data and those derived from direct observation will first need to be assessed (Warren-Smith and McGreevy 2010). Objectives: To determine: (1) if pedometers accurately record equine steps at walk and trot, when compared to video analysis, (2) if alternative positioning of these pedometers affected the accuracy of step detection when compared to video analysis. Method: Five horses each wearing a Yamax Digiwalk SW-200 (spring lever arm pedometer) and a Yamax Power Walker PW-610/611 (piezoelectric pedometer) positioned on the left foreleg (LF), left scapular (LS), chest (C), right scapular (RS) and right foreleg (RF) walked and trotted on a 20m circle 10 times on each rein to yield 100 results for each gait. Video recorded by GoPro Hero 3 Black edition camera positioned on the girth facing the forefeet using the following settings: 180 degree field of view, 720p, 120fps. Both visual and audio data were captured and recorded. Results: Kruskal-Wallis Test on the deviation from actual number of steps per unit for each type of pedometer and the position of the pedometer compared to the actual number of steps taken by both forelegs determined that there is a significant difference (H1= 340.31; P<0.0001) between readings in walk and also in trot (H1 = 483.49; P<0.0001). A further Kruskal-Wallis Test on the deviation from actual number of steps per unit for each type of pedometer and the position of the pedometer compared to the actual number of steps taken by individual forelegs determined that there is a significant difference (H = 403.36; p<0.0001) at walk and also at trot (H = 477.10; p<0.0001). Conclusion: In summary, the analysis of the pedometer data compared to video analysis indicate that pedometers are not useful for scoring forelimb movements in horses at walk and trot, regardless of positioning.
The following license files are associated with this item: