The numberof senses effect in polysemous adjective recognition

Abstract

Previous research revealed a significant polysemy effect: namely, it found that words with multiple related senses (polysemous words) are recognised faster compared to the words with multiple unrelated meanings (homonymous words) and words with only one meaning/sense (unambiguous words; Rodd et al., 2002). The measure of ambiguity in polysemous words was the number of senses (NoS), derived from the meanings/senses provided by native speakers. NoS was a significant predictor of reaction time in visual lexical decision task (VLDT) experiments (Filipović Đurđević, 2007). This is in accordance with various models of lexical ambiguity processing. Although some attribute the effect to an increased semantic activation due to the facilitation among the related senses (Armstrong & Plaut, 2016; Rodd et al., 2004), whereas others attribute it to the differences at the level of responding (Hino & Lupker, 1996), they agree in predicting the processing advantage in polysemous word recognition. Research in Serbian revealed this effect in noun and verb processing (Filipović Đurđević & Kostić, 2008; Mišić & Filipović Đurđević, 2019; 2020). The aim of this research was to further generalize the findings and to test whether the NoS effect is present in polysemous adjective recognition. The prediction was that the increase in the NoS and word frequency would be followed by faster adjective recognition. In this research, the participants were presented with a VLDT consisting of 107 polysemous Serbian adjectives. They were presented in all three grammatical genders using the Latin square design between participants, which allowed each participant to see only one form of the same adjective. Multiple regression revealed that the NoS and frequency were significant predictors of the reaction time: polysemous adjectives with higher NoS and higher frequency were processed faster (NoS: β = -.199, S.E. = .093, df = 106, t = -2.143, p < .05; frequency: β = -.281, S.E. = .093, df = 106, t = -3.036, p < .05). These findings are in accordance with our hypothesis and concur with the previous findings from the experiments with nouns and verbs (Filipović Đurđević & Kostić, 2008; Mišić & Filipović Đurđević, 2019; 2020), as well as various models regarding word ambiguity processing (Armstrong & Plaut, 2016; Hino & Lupker, 1996; Rodd et al., 2004). Together they converge to the conclusion that the NoS facilitates recognition of polysemous words in the VLDT across different parts of speech.