Jérémie Boulanger

Skill transfer specificity shapes perception and action under varying environmental constraints

Using an ecological dynamics framework, this study investigated the generality and specificity of skill transfer processes in organisation of perception and action using climbing as a task vehicle. Fluency of hip trajectory and orientation was assessed using normalized jerk coefficients exhibited by participants as they adapted perception and action under varying environmental constraints. Twelve recreational climbers were divided into two groups: one completing a 10-m high route on an indoor climbing wall; a second undertaking a 10-m high route on an icefall in a top-rope condition. We maintained the same level of difficulty between these two performance environments. An inertial measurement unit was attached each climbers hips to collect 3D acceleration and 3D orientation data to compute jerk coefficient values. Video footage was used to record the ratio of exploratory/performatory movements. Results showed higher jerk coefficient values and number of exploratory movements for performance on the icefall route, perhaps due to greater functional complexity in perception and action required when climbing icefalls, which involves use of specific tools for anchorage. Findings demonstrated how individuals solve different motor problems, exploiting positive general transfer processes enabling participants to explore the pick-up of information for the perception of affordances specific to icefall climbing.

Automatic Sensor-Based Detection and Classification of Climbing Activities

This paper presents a novel application of a machine learning method to automatically detect and classify climbing activities using inertial measurement units (IMUs) attached to the wrists, feet, and pelvis of the climber. This detection/classification can be useful for research in sport science to replace manual annotation where IMUs are becoming common. Detection requires a learning phase with manual annotation to construct statistical models. Full-body activity is then classified based on the detection of each IMU.

Full-body movement pattern recognition in climbing*

The aim of this study was to propose a method for full-body movement pattern recognition in climbing, by computing the 3D unitary vector of the four limbs and pelvis during performance. One climber with an intermediate skill level traversed two easy routes of similar rates of difficulty (5c difficulty on French scale), 10m in height under top-rope conditions. The first route was simply designed to allow horizontal edge-hold grasping, while the second route was designed with more complexity to allow both horizontal and vertical edge-hold grasping. Five inertial measurement units (IMUs) were attached to the pelvis, both feet and forearms to analyse the 3D unitary vector of each limb and pelvis. Cluster analysis was performed to detect the number of clusters that emerged from coordination of the four limbs and pelvis during climbing performance. Analysis revealed 22 clusters with 11 clusters unique across the two routes. Six clusters were unique to the simple hold design route and five clusters emerged only in the complex hold design route. We conclude that clustering supported identification of full-body orientations during traversal, representing a level of analysis that can provide useful information for performance monitoring in climbing.

Movement phase detection in climbing

The aim of this study was to propose a method to automatically detect the different types of behavioural states in climbing. One climber traversed an easy route (5c difficulty on French scale) of 10 m height with a top-rope. Five inertial measurement units (IMU) (3D accelerometer, 3D gyroscope, 3D magnetometer) were attached to the pelvis, both feet and forearms to analyse the direction (3D unitary vector) of each limb and pelvis in ambient space. Segmentation and classification processes supported detection of movement and immobility phases for each IMU. Depending on whether limbs and/or pelvis were moving, four states of behaviour were detected: immobility (absence of limb and pelvis motion), hold exploration (absence of pelvis motion but at least one limb in motion), pelvis movement (pelvis in motion but absence of limb motion) and global motion (pelvis in motion and at least one limb in motion). Results indicated that the climber spent 10% of time immobile, 65% exploring holds, 1% with pelvis in motion (indicating posture regulation) and 24% in global movement (could indicate transition between holds). This new method allows automatic, rapid and reliable detection of climbing behavioural states to facilitate assessment and monitoring of climbing performance.

Comparing dynamics of fluency and inter-limb coordination in climbing activities using multi-scale Jensen–Shannon embedding and clustering

This paper reports the results of two studies carried out in a controlled environment aiming to understand relationships between movement patterns of coordination that emerge during climbing and performance outcomes. It involves a recent method of nonlinear dimensionality reduction, multi-scale Jensen–Shannon neighbor embedding (Lee et al., 2015), which has been applied to recordings of movement sensors in order to visualize coordination patterns adapted by climbers. Initial clustering at the climb scale provides details linking behavioral patterns with climbing fluency/smoothness (i.e., the performance outcome). Further clustering on shorter time intervals, where individual actions within a climb are analyzed, enables more detailed exploratory data analysis of behavior. Results suggest that the nature of individual learning curves (the global, trial-to-trial performance) corresponded to certain behavioral patterns (the within trial motor behavior). We highlight and discuss three distinctive learning curves and their corresponding relationship to behavioral pattern emergence, namely: no improvement and a lack of new motor behavior emergence; sudden improvement and the emergence of new motor behaviors; and gradual improvement and a lack of new motor behavior emergence.