Effects of Acoustic Speech Variation on Personality Trait Perception
| dc.contributor.advisor | Pape, Daniel | |
| dc.contributor.author | Pearsell, Sara | |
| dc.contributor.department | Cognitive Science of Language | en_US |
| dc.date.accessioned | 2024-07-12T20:12:30Z | |
| dc.date.available | 2024-07-12T20:12:30Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | This thesis examines acoustic properties of speech which influence perceptions of personality traits, specifically charisma. The following questions are addressed: How does amplitude variation influence ratings of dominance (i), how does voice quality affect personality trait attribution (ii), and how does allophonic variation affect ratings of charisma (iii). Chapter 2 addresses question (i), finding that certain linguistic levels (increased amplitude in sentence and syllable levels) affected dominance ratings while others (increased amplitude at word level and reduction at syllable level) did not. Increased sentence amplitude increased dominance ratings while increased syllable amplitudes had inverse effects. Additionally, two types of dominance were examined (social and physical dominance) but no statistically significant differences were found between the two. Chapter 3 examines question (ii). All voice qualities investigated (modal, creaky, breathy, nasal, and smiling) were found to be statistically significant. Effect sizes for statistical significance varied for each voice quality. Creaky voice (rated the lowest/ most negative) and smiling voice (rated the highest/most positive) had the strongest effects. Chapter 4 examines question (iii). Experiment 1 (in-person) and Experiment 2 (online) examined the effects of allophonic variation, final consonant devoicing (FCD), and /t/ variation, on ratings of charisma. Experiment 1 found statistically significant rating differences for FCD. Final voiced items were rated higher compared to devoiced ones. For the /t/ variation, only speaker differences were found to be statistically significant. Experiment 2 showed no statistically significant results for FCD, whereas /t/ variation found statistical significance for [t] productions versus the glottal stop, and for flap productions versus the glottal stop. No rating differences were found between [t] and flap. Overall, this thesis demonstrates that some acoustic variations within speech affect personality trait ratings, specifically charisma, while others do not. I discuss reasons for these outcomes and their utilization in various domains, including AI. | en_US |
| dc.description.degree | Doctor of Philosophy (PhD) | en_US |
| dc.description.degreetype | Dissertation | en_US |
| dc.description.layabstract | This research explores the effects of different aspects of speech on the impressions of the speaker’s personality. It examines three questions: (i) how loudness affects the perception of dominance, (ii) how voice quality influences personality traits, and (iii) how pronunciation variations impact charisma. Chapter 2 (i) found that for sentences, increases in loudness increases perceptions of dominance, while for syllables they reduce them. Chapter 3 (ii) found that each voice quality investigated affects personality trait ratings, but creaky voice was perceived most negatively and smiling voice most positively. Chapter 3 (iii) found that voiced final consonants are rated higher in charisma than devoiced ones for in-person participants, but not for online participants. Regular [t] and flap pronunciations differ from glottal stops but not from each other only for online participants. The findings suggest that certain aspects of speech variation influence personality trait ratings and offer applications to teaching and AI. | en_US |
| dc.identifier.uri | http://hdl.handle.net/11375/29947 | |
| dc.language.iso | en | en_US |
| dc.subject | speech variation | en_US |
| dc.title | Effects of Acoustic Speech Variation on Personality Trait Perception | en_US |
| dc.type | Thesis | en_US |