Poster #42 - The “Gleam-Glum” Sound Symbolic Effect: English-speaking Children and Adults link Phonemic Vowel Sounds with Emotional Valence

• In Event: Poster Session 12

Abstract

Contrary to a traditional linguistic view that word-to-meaning mappings are arbitrary (de Saussure, 1959), sound-symbolic phenomena demonstrate non-arbitrary associations between the phonetic properties of words and their meanings (Blasi, et al., 2016; Perry et al., 2015). Most known examples are like the bouba-kiki effect where the words “bouba” and “kiki” respectively refer to round and spiky shapes. Recently, a novel type of sound symbolism has been found, the gleam-glum effect, for which phonemes can carry emotional valence. The findings support that [i]-phonemic words (like “gleam”) are perceived as emotionally positive while [Ʌ]-phonemic words (like “glum”) are perceived as emotionally negative (Yu et al., 2021). In the current pre-registered study, we extend this prior work by testing if the gleam-glum effect 1) is present in a new pseudoword-to-scene paradigm, and 2) is robust across adults and young children.

English-speaking college adults (n=105, Mage=19.18, 41 with second language experience) and 5- to 7-year-olds (n=31, Mage=6.55, 10 with second language experience, ongoing data collection) were tested on the same pseudoword-to-scene mapping paradigm. In each trial, participants heard a pair of novel matched [i]- vs [Ʌ]-pseudowords contrasting only in the interposed phoneme (e.g. “zeem” vs “zum”), while seeing two cartoon scenes contrasting in valence, an emotionally positive vs negative scene (see Figure 1). The participants’ task was to match up one of the novel words with either the positive or negative scene. Adults self completed the study online, and children were tested over zoom, moderated by an experimenter.

Results replicated the gleam-glum effect among adults (M=.75, SD=.21, p<.001, d=1.18). In addition, we found the first evidence of the effect among young children (M=.62, SD=.14, p<.001, d =.91, see Figure 2A). The effect sizes for both adults and children were large, but with children exhibiting a somewhat smaller and less varied effect (ΔM=.13, p<.001, d=.63). Figure 2B and 2C illustrates the effect size distribution by pseudoword for each age group. We also found that age did not predict the magnitude of the effect within the 5-7 year-old range (b=.01, p=.755).
Our findings support the robustness of the gleam-glum effect in a pseudoword-to-scene mapping task, and provide the first empirical evidence that the effect is robust across young children (5- to 7-year-olds) and adults. Importantly, the magnitude of the gleam-glum effect is already quite substantial in children of this age, while still not quite as large as adults. This supports that the gleam-glum effect may emerge at a younger age, but also appear to continue to grow stronger with age or age-related experience (e.g. language use). Overall, our findings support the robustness of this new sound symbolic effect, where phonemes can carry emotional valence, occurring across testing methods and across development. The presence of the gleam-glum effect in young children invites further investigation regarding how much earlier in age the effect emerges, and the role of learning such as in child-directed speech.

Authors

• Ye Li, Arizona State University
  Presenting Author

• Shin-Phing Yu, Arizona State University

• Jada Randall, Arizona State University

• Skylar Keefer, Arizona State University

• Amanda Sather, Arizona State University

• Michael K McBeath, Arizona State University

• Viridiana L. Benitez, Arizona State University