ABSTRACT
The present research examined whether American listeners express different attitudes toward different British accents (i.e. RP, Northern, Scottish, and Welsh) and, if so, which cognitive mechanisms underlie that attitudinal variation. Across two studies, we found that Americans’ attitudes toward British accents were differentiated. RP-accented speakers were consistently rated more favourably than speakers of regional British accents, with some regional accents (e.g. Northern) denigrated more than others (e.g. Scottish). This attitudinal variation was associated with variation in how speakers of different accents were categorised (Study 1), how prototypical of their respective groups they were perceived to be (Study 2), and how much difficulty listeners experienced processing their speech (Studies 1 and 2). These findings contribute to our understanding of language attitudes processes and have methodological implications for the elicitation of language attitudes using audio stimuli.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The data that support the findings of this research are available from the corresponding author upon reasonable request.
Notes
1 Stated differently, social categorization is the process of construing people as members of social groups (rather than as individuals) and stereotyping is the processes of attributing people traits and characteristics believed to be associated with those groups (i.e., stereotypes).
2 In the context of speech processing, processing fluency is conceptually indistinguishable from comprehensibility—defined as the perceived ease or difficulty in understanding speech—but different from intelligibility—defined as the extent to which speech is actually understood (see Munro and Derwing Citation1995). In the present research, we use the term processing fluency, rather than comprehensibility, because the former is a broader construct, applicable to a wider range of domains beyond speech processing (see Alter and Oppenheimer Citation2009).
3 These particular British accents were selected because they represent some of the major linguistic distinctions on the island of Great Britain (see Wells Citation1986) and correspond to the three countries located on the island (i.e., England, Scotland, and Wales).
4 Some participants met multiple exclusion criteria.
5 We first converted Pearson correlation coefficients into Fischer’s z-scores (Fisher Citation1921), which we used to calculate averages and corresponding confidence intervals, and then converted these values back into Pearson correlation coefficients.
6 We also computed correlations based only on the ratings for the four British accents. The same pattern of results emerged. Familiarity and processing fluency were positively correlated (correlation of averages: r = .97, p = .03; average correlation: r = .57, 95% CI = .44, .68). Processing fluency was positively correlated with both competence (correlation of averages: r = .97, p = .03; average correlation: r = .71, 95% CI = .61, .79) and warmth (correlation of averages: r = .65, p = .35; average correlation: r = .41, 95% CI = .25, .55).
7 As one reviewer pointed out, some Americans may use the terms England and Britain interchangeably. To what extent Americans associate these terms with the same or different identities and stereotypes remains an important direction for future research.
8 Some participants met multiple exclusion criteria.
9 Analyses using the entire sample produced an identical pattern of results.
10 For a discussion of distinctive Welsh varieties, see Evans, Dunbar, and Chartier (Citation2020); for Northern varieties, see Beal (Citation2004) and Wales (Citation2006); for Scottish varieties, see Chambers (Citation2019); and for RP varieties see Crystal (Citation2020).