Laura Peeters

Trunk, head and pelvis interactions in healthy children | 53 3 Flexion, lateral bending to the right and rotation to the right were defined as positive. Movements of the different trunk segments were named after the more cranial segment (e.g. upper thoracic angle represents the orientation of the upper thoracic segment relative to the lower thoracic segment). ‘Total trunk movement’ is used for the summation of all trunk segments. Pelvis, trunk and neck angles during a recording while sitting quietly (i.e. sitting upright with both hands on the legs (see Figure 2)) were used to zero angles in the movement trials. This was done by post-multiplying the orientation matrix of all segments with the inverse of the orientation matrix while sitting quietly. All kinematics for the two left-hand dominant subjects were transformed to match the kinematics for the right- hand dominant subjects. To determine maximum ranges of trunk motion, the trial in which the summed angle of all trunk segments and pelvis was maximal in the requested movement plane was selected. Similarly, the trial with the maximum range of neck motion was selected. For all reaching tasks, the instant of task execution that was used for analyses of segment angles was defined as the first instant where the wrist velocity reached zero after the maximum wrist velocity. For the drink task, this instant was at the point where the hand was the closest to the mouth (i.e. peak of the wrist movement path) and for the plate task, this was where the hands grabbed the plate on the left side (i.e. peak of right wrist movement path). For all of these tasks, the start was identified as the instant where the velocity of the wrist exceeded 5% of its peak velocity. All instants were selected by a computer algorithm and afterwards visually confirmed. For the drawing and dexterity tasks, the instant at task execution was midway between start and end. Start and end were defined manually with the use of video and kinematics recordings, since rendering automatic detection was unfeasible due to low wrist velocity. The ROM was defined as the segment angles at the instant of task execution, subtracted by the segment angles at the start position of the same task. Kinematics of the arms are not reported in this article. Head movements relative to the upper thoracic segment were categorized in three different strategies: no relative movement between head and trunk, relative movement of the head in the same direction as the trunk, or relative movement of the head in opposite direction to the trunk. The range where the head movement was defined none, was in between plus or minus two times the standard deviation obtained from the head movement during the quiet sitting task. The maximum standard deviation of all participants, in each direction was used for this. For each subject and trial, the head strategy was determined and the percentage of subjects using each strategy was calculated. Movement onsets of the head and trunk were defined relative to hand movement