Fig. 3
From: Quantifying spontaneous infant movements using state-space models

State movement dynamics. (a) Direction and magnitude of frame-to-frame movements of each body point in each state, centered on the group average position. Movements were randomly selected (n = 5000) from frames in each state in all videos. (b). For each body point, the top 10% high velocity frames (those containing most movement) were identified and grouped by state membership. Plots shows average % high velocity frames for each group of bodypoints in each state for every video (n = 486). Boxplot shows median and interquartile range.