Acoustic Determinants of Perceptions of Personality From Speech

Abstract

The studies reviewed in this paper are somewhat diverse. The one unifying feature in all of them is their purpose of identifying the ways in which non-content aspects of speech elicit personality impressions. Whether the speech variables are dialect (or sociolect) categories or personality-expressive characteristics, whether they are linguistically categorized or acoustically manipulated variables, the central interest is in identifying the "implicit personality theory" of judges in their reactions to various types of speech. THE QUANTIFIABILITY OF VOCAL VARIABLESStudies of "implicit personality theory" (Bruner and Tagiuri, 1954) and interpersonal perception have a long history in social psychology, but investigators have typically avoided using actual people as stimuli, probably because of the difficulties of quantifying the appearance and behavior of stimulus persons. It is easy enough to quantify the personality impression responses of judges by using adjective rating scales, but there are so many aspects to the appearance and observable behavior of the stimulus person that it becomes difficult to specify the cues by which the judge makes his judgments. Many investigators have avoided the problem by reading a list of adjectives to judges, pretending that the traits apply to some real person, and then asking judges to check other adjectives that they think would apply to the person. This procedure might provide useful information about which traits go together to form each judge's "implicit personality theory", but it also runs the risk of being merely a word definition game. 1 1 The same kind of information as to which traits go together in the "implicit personality theories" of persons has been gathered in the more concrete situations where actual people are rated on adjectives. The judgments are made either from acquaintance with the persons or from exposure to the vocal, visual or verbal characteristics of the person and often the ratings are factor analyzed to present the interrelations among the trait adjectives in simplified form. Numerous examples of such studies appear in this review.

British attitudes towards six varieties of English in the USA and Britain

“…In the study of Brown, Strong and Rencher (1975), however, Anglo-American students with no knowledge of the French language made exactly the same judgements as did French Canadians. At first sight, this study seems able to support the inherent value hypothesis.…”

User—robot personality matching and assistive robot behavior adaptation for post-stroke rehabilitation therapy

“…The similarity-attraction principle, which assumes that individuals are more attracted to others who manifest the same personality, has been studied in human-machine interaction (HCI) (e.g., [37]). Also, studies [6] have found that prosodic characteristics are linked with features of personality, e.g., excitement or arousal (extroversion-introversion) are strongly correlated to prosodic features such as pitch level [53], pitch range, and tempo [2]. The interaction scripts that we designed in this work display extroverted and introverted personality type through the choice of words and paralinguistic cues.…”

Modelling personality in voices of talking products through prosodic parameters

“…It has also been shown that louder voices and male voices are considered to be more logical than softer voices and female voices. We associated sincerity to benevolence, which is expressed by speech with more variable intonation [3], whereas high pitched voices sound less benevolent [2] and those with a high fundamental frequency more deceiving [18]. In order to express sophistication we focused on its facets softness and attraction, which are associated with voices that are less monotonous, with a low pitch and also a larger pitch range [20].…”