Abstract
The aim of this thesis is to examine how variations in conversant speech style affect the dynamics of conversational behaviour. Three key issues are addressed. Firstly, given that women and men have been shown to differ in speech style, how do conversants respond to these different styles of speech? In other words, to what extent does one conversant's speech behaviours influence the linguistic responses of the other interlocutor? Secondly, do these reactions to conversant speech behaviours depend on the gender of the respondent? Finally, in what way does conversant speech style change as a function of conversational context and duration?
Experiment 1 investigated the impact of confederate speech style on the speaking time, interruption frequency, minimal response frequency, and reduced eye contact of conversants. This experiment further examined participants' convergence towards the speech behaviours of confederates and the relative influence of confederate behaviours in predicting participant speech behaviours. To examine this, four female and four male confederates were trained to employ a facilitative and a non-facilitative style of speech in interactions with 64 other university students in same- and mixed-sex dyads. Two different topics were specified and counter -balanced across gender and conversational style. The results indicated that confederate speech style, and not confederate gender, was most likely to predict participant behaviour in conversations. Participants who conversed with a non-facilitative partner spoke less, offered fewer minimal responses, interrupted more, and looked away more than participants who conversed with a facilitative partner. Subtle differences in how women and men responded to confederate behaviour were demonstrated. Male participants appeared more reactive to the speech style of the confederate than did female participants. For example, males increased their minimal response frequency when they spoke with a facilitative female confederate, whereas female participants maintained a similar frequency of minimal responses regardless of speech style or confederate gender. Regression analyses revealed that increases in confederate speaking time were related to significantly less speech by females, more frequent instances of looking away by females and increased interruptions by males. Speech accommodation to interruptions was evident by all participants, whereas speech accommodation to minimal responses was demonstrated only by male participants to facilitative female confederates.
Experiment 2 further examined the relationship of gender and speech style using naturally occurring variations in speech style rather than the manipulated speech style of trained confederates. 48 women and men between the ages of 30 and 60 years of age each participated in one same- and two mixed-sex dyadic conversations with no topic restraints imposed. When participants were categorised on the basis of high or low minimal response frequency, we observed other behaviours concomitant with this categorisation and unrelated to gender. Further analyses showed that women and men differed in how they responded to specific speech behaviours of their conversational partners. The results suggested that women and men might differ in their interpretation of their speaking partner's speech and in the functions specific language devices serve in conversation. For example, it appeared that female participants in this experiment were interpreting minimal responses during conversation as verbal reinforcers offering encouragement to talk. It was also evident that, across the three conversations, women and men gradually shifted in their speech style towards patterns noted in the literature. By the third conversation, men talked more than women and took longer turns; women used more frequent minimal responses; and men marginally reduced their frequency of questions.
Together, these results suggest that the speech style of conversants is a more reliable predictor of participants' speech behaviour than the gender of the conversant. Individuals who were paired with a facilitative conversational partner talked more than individuals paired with a non-facilitative partner, regardless of their gender. The results further demonstrated subtle differences in how women and men responded to the behaviour of their conversational partners. Further implications of these findings and future directions of gender and language research are discussed.