Whoosh! visual depictions of direction, speed, and temporality: a corpus analysis of motion events in global comics

Representing motion in static images is a challenge due to the limitations of two-dimensional visual media. Such motion depictions are common in visual narratives, yet no studies have extensively analyzed their properties. This study examines how lexicalized visual cues contribute to motion events by analyzing 315 comics across three studies. Study 1 focuses on the frequency of motion cues and path segments, revealing that midpoints of paths were depicted more frequently than endpoints or starting points, consistent with prior corpus findings. Study 2 then explores how different cues relate to the explicitness of direction, manner of speed, and temporality. While most cues were associated with explicit paths, others, like partial repetition of postures, were more implicit. Also, backfixing and suppletion lines showed no positive relationship with any kind of paths but related to speed of motion. Cues that rely on indexicality also related to implied moments of action. Finally, Study 3 used Principal Component Analysis to uncover broader patterns in the data, revealing more nuanced groupings of properties of motion events than a binary explicit/implicit distinction. Overall, this study offers new insights into the ways static images convey dynamic information by showing they are systematized in a visual vocabulary.