Semantic and Associative Factors in Lexical Processing of Polysemous Bionyms in Russian: Evidence from an Eye-Tracking Study in the Visual World Paradigm

The aim of the study is to provide experimental results of lexical processing of polyse-mous Russian bionyms. This study takes into account “online” mechanisms of lexical processing and uses eye movement parameters of subjects during listening to a polysemous word out of context. The Visual World Paradigm was used for experimental design modeling. This technique allows revealing some underpinnings of psychological representations of different meanings of a polysemous word. The choice of Russian bionyms (names of living organisms and plants) is due to the fact that this class of words falls into the core lexical system of almost any world language. This makes it possible to compare experimental results in languages of different structures. One important aspect of such modeling is not only the consideration of semantic, but also associative factors of lexical access to the meaning of polyse-mous bionyms. Thirty one Russian speakers were recruited for an eye-tracking experiment using the Visual World Paradigm. A dataset of audio stimuli contained 60 Russian nouns with varying degrees of word frequency, length and number of meanings. The visual panel contained different types of word referents: Т (target representing the dominant meaning of a spoken word), TD (target representing a diminutive spoken word), Hype (competitor representing hypo-hyperonymic relations with a spoken word), Hypo (competitor representing hypo-hyponymic relations with a spoken word), Meta (competitor representing the metaphorical meaning of a spoken word), Meto (competitor representing the metonymic meaning of a spoken word), A (competitor representing the associate for a spoken word), D (distractor unrelated to a spoken word). Every spoken word and visual panel with one target, two competitors and one distractor were presented for 1 200 ms without any context. After each trial participants were asked to respond how many meanings and associations of the spoken word were presented on the visual panel. Linear mixed effects modeling showed significant differences in several eye movement parameters of the subjects. The calculated models have moderate explanatory effect and this indicates the prospect of a further search for fixed effects controlling. The most surprising finding was the difference between the early (first fixation time) and late measures of polysemous word meaning processing. While the late measures of lexical processing are guided by a constant comparison of the input (auditory stimulus) with the irrelevant or weakly related referent, the early measures are influenced by the sequence of revisions of areas of interest that represent different referents. The first fixation on the first area of interest tended to be the shortest. At all stages of the ambiguous bionyms processing, the referents that reflected the dominant meaning were more activated in the subjects’ visual field during the first entering the area of interest, and later this activation was inhibited. Associative relations had significant effects in this process, interacting with other processes of extracting representations of word meanings from the subjects’ memory.

Keywords

Authors

References