Figure 4.

The proportion of time during all caption episodes for a given subject and condition (captions only, top; captions + audio, bottom) that was allocated to: (a) viewing within the caption area, including pause durations between successive saccades and in-flight time of saccades; (b) viewing within the video area or traveling between caption and video areas, including pause durations between successive saccades and in-flight time of saccades; (c) the latency of the first saccade made into the caption area in response to the onset of a caption episode; (d) intervals in which tracker lock was lost.