Grant McGuire

Novelty and social preference in phonetic accommodation

Abstract Phonetic imitation is the unintentional, spontaneous acquisition of speech characteristics of another talker. Previous work has shown that imitation is strongly moderated by social preference in adults, and that social preference affects childrens speech acquisition within peer groups. Such findings have led to the suggestion that phonetic imitation is related to larger processes of sound change in a change-by-accommodation model. This study examines how preferential processing of particular voice types affects spontaneous phonetic accommodation, interpreting the results in the context of how sound change can be propagated through a speech community. To explore this question eight model talkers previously rated as attractive, unattractive, typical, and atypical for each gender were used in an auditory naming paradigm. Twenty participants completed the task, and an AXB measure was used to quantify imitation. Female participants imitated more than male participants, but this varied across model voices. Females were found to rely more on social preference than men, while both groups imitated the atypical voices. The results suggest that females adapt their speech to auditory input more readily, but the nature of the accommodation does not qualify as direct evidence for a change-by-accommodation model given the constrained context of the task.

Perceptual fluency and judgments of vocal aesthetics and stereotypicality.

Research has shown that processing dynamics on the perceivers end determine aesthetic pleasure. Specifically, typical objects, which are processed more fluently, are perceived as more attractive. We extend this notion of perceptual fluency to judgments of vocal aesthetics. Vocal attractiveness has traditionally been examined with respect to sexual dimorphism and the apparent size of a talker, as reconstructed from the acoustic signal, despite evidence that gender-specific speech patterns are learned social behaviors. In this study, we report on a series of three experiments using 60 voices (30 females) to compare the relationship between judgments of vocal attractiveness, stereotypicality, and gender categorization fluency. Our results indicate that attractiveness and stereotypicality are highly correlated for female and male voices. Stereotypicality and categorization fluency were also correlated for male voices, but not female voices. Crucially, stereotypicality and categorization fluency interacted to predict attractiveness, suggesting the role of perceptual fluency is present, but nuanced, in judgments of human voices.

The role of voice similarity in accommodation

Studies of accommodation show that some talkers are perceived as accommodating more than others. One possibility is that the similarity of the shadowers voice to a model talkers can account, in part, for the amount of perceived accommodation. To determine this, we conducted an auditory naming task having eight model talker voices previously rated for attractiveness and prototypicality, such that the Most Attractive and Least Attractive and Most Typical and Least Typical voices for each gender were used as models. Twenty participants completed an auditory naming task with these eight voices. A separate group of twenty listeners rated the similarity of model tokens and shadowers baseline productions using a visual analogue scale. The results of this task were compared to the perceived accommodation results from a separate AXB rating task. Overall, female voices that were more different from the models showed more accommodation. This effect was not found for males, who generally showed less accommodation ...

Sub-phonemic correlates of gender and regional identity in California

Given that the Northern and Southern California have large metropolitan areas geographically and culturally separated from each other, it is to be expected that each is developing a unique linguistic identity. Despite a handful of ethnographic studies showing otherwise (e.g., Hall-Lew 2009), the West has generally been lumped into a single dialect region (Labov etal. 2006). This paper presents data showing sub-phonemic differences between the regions that break along gender lines. Vowel productions from 14 (female = 8) Northern Californians (NCs) and 15 (female = 8) Southern Californians (SCs) were analyzed for regional differences in normalized vowel quality, voice quality (spectral tilt), pitch, and duration. No major differences in vowel quality were found. However, interactions were found between region and gender for duration and voice quality. Specifically, NC females had significantly longer word durations than NC males, with no difference between genders for SC. For voice quality, H1-H2 and H1-A3 ...

Tongue body shape during the production of Irish palatalization and velarization

Irish is an endangered Celtic (Goidelic) language that has a rare phonemic contrast between palatalized and velarized consonants for all places and manners of articulation. Using ultrasound tongue body imaging of 15 native speakers we provide comparative data on the phonetic realization of the contrast, specifically focusing on how place, manner, and vowel context each affect tongue body position during production for each of the three major dialects. Using principal components analysis, we find evidence for the role of tongue root advancement in the contrast independent of the (less surprising) roles of tongue body frontness and raising. This data have consequences for our understanding of the relationship between phonetic and phonological categories as well as the role of perceptual saliency in shaping inventories.

A listener-based account for dispersion effects in sound change

Phonetic dispersion has been proposed as the driving force behind a number of closely related sound-change phenomena. Listener-based accounts of dispersion (Labov, 1994, 587; Wedel, 2006; Denby, 2013) posit that phonetically unambiguous productions influence future productions of the listener more than ambiguous productions. The mechanism that drives this is a filter by which ambiguous productions are not stored, and thus do not update the phonemic categories of the listener. In turn, they are not reflected in that listeners future productions. In a new experiment, subjects heard words in noise and were asked to identify them by responding using a keyboard, following Goldinger (1996). Stimuli were from monosyllabic stop-initial minimal pairs differing in initial voicing, e.g., pat/bat. Half of these pairs were unambiguous productions, while the stop-initial VOT of the other half were manipulated to be somewhat ambiguous. If subjects store ambiguous words normally, their accuracy should improve with every...

The effect of accommodation on perceived vocal aesthetics

We conducted an auditory naming task (n = 20) using eight model talker voices previously rated for attractiveness and prototypicality such that the most attractive, least attractive, most typical, and least typical voice for each gender served as a model talker. Female shadowers accommodated more than males, particularly to the Most Attractive Female model. This finding led us to question if in the course of accommodation to an attractive female voice, female shadowers themselves become more vocally attractive. We then conducted an AX task where listeners judged whether shadowers’ baseline or shadowed productions were more attractive. Our results suggest that shadowers do modulate their perceived attractiveness in the course of accommodating; in particular, the more females accommodated to the Most Attractive Female model, the more attractive her own voice became. We are currently running a second study exploring whether shadowers’ voices change in perceived typicality when accommodating the Most Typical ...

The effect of high and low variability conditions on phonetic convergence

Studies of phonetic convergence using single-word auditory naming tasks offer insight into how variability in stimuli affect the translation from speech perception to speech production. In this paper, we report on an experiment which compares phonetic convergence in single-word production between high variability (mixed talker condition) or low variability (blocked talker condition) using five female model talkers’ voices for the task. Twenty female participants participated in a production task where they produced baseline tokens and shadowed model talker productions in either the high or low variability condition. Phonetic imitation was quantified using listener judgments in an AXB similarity rating task where a model token was compared to a shadower’s baseline and shadowed token. The results indicate a trend towards more convergence in the low variability condition, but this was highly affected by model voice; one model voice was spontaneously imitated more in the high variability condition than the low variability condition. Several socio-cognitive tests were administered to shadowers, and continued analyses of the data will explore whether these individual socio-cognitive measures predict shadowers’ predispositions toward phonetic convergence. Studies of phonetic convergence using single-word auditory naming tasks offer insight into how variability in stimuli affect the translation from speech perception to speech production. In this paper, we report on an experiment which compares phonetic convergence in single-word production between high variability (mixed talker condition) or low variability (blocked talker condition) using five female model talkers’ voices for the task. Twenty female participants participated in a production task where they produced baseline tokens and shadowed model talker productions in either the high or low variability condition. Phonetic imitation was quantified using listener judgments in an AXB similarity rating task where a model token was compared to a shadower’s baseline and shadowed token. The results indicate a trend towards more convergence in the low variability condition, but this was highly affected by model voice; one model voice was spontaneously imitated more in the high variability condition than the lo...

On the relationship between vocal aesthetics and speech perception.

This study reports data from four experiments exploring the interplay of meta‐linguistic analyzes and lower level tasks with the goal of understanding how judgments of vocal aesthetics and voice typicality affect voice and phoneme processing. In the first, speakers of west coast North American English rated the attractiveness of 60 American English voices. Results from this experiment were compared against the following ones. In the second experiment listeners were asked to rate the typicality of each voice for its sex. Ratings for both showed a strong correlation, suggesting that vocal attractiveness and voice typicality are related. In the next experiment listeners were asked to quickly classify the voices as male or female. Faster reaction times correlated with judgments of higher typicality, but not with attractiveness. Finally a group of listeners were asked to classify the vowels produced by the voices. Here a correlation was found with both rating tasks such that listeners were faster at vowel clas...

Phonetic correlates of vocal attractiveness in American English.

This study explores fine‐grained phonetic vocal characteristics that underpin vocal attractiveness. In general, while it is well known that F0 plays a major role in such judgments [see, e.g., Riding et al. (2006)] there is a distinct lack of more detailed examinations of the phenomenon [see Zuta (2007) for a notable exception]. Moreover, the term “attractiveness” is generally ill‐defined and conflated with other terms (such as “pleasantness”). Therefore, the specific goal of this study is to replicate and extend such studies by including a large number of talkers, more detailed acoustic measures, and better definition the term “attractiveness”. Specifically, 60 talkers from California (30 female) produced isolated words controlled for phonetic content. These voices will be played to listeners who will judge the attractiveness of each talker. Ratings of these talkers will be compared against these acoustic measures: duration, average F0, F0 variation, spectral tilt, jitter, vowel space area, long term aver...