Describing and remembering complex motion events in ‘real-world' videos
Abstract
Languages lexicalize motion events in different ways. Previous studies, using short clips of isolated motion events, have shown that these cross-linguistic differences can affect attention and memory. In this study, we use videos of ‘real world' complex motion events to test the effects of language patterns on perceived saliency of event elements as well as the granularity of speakers' descriptions. To increase the perceived importance of the task and its applicability to forensic contexts, participants (N=64) were informed that they were viewing CCTV footage of potential criminal activity, before completing a surprise or expected memory task. Participants watched a two-minute video depicting multiple interconnected motion events with several paths and manners of motion, placement events, as well as different event endings. Results unpack the extent of thinking for speaking effects on later ‘eye witness' description, as well as nuanced patterns of description change on iterated descriptions of the same video.