While sociolinguistic research is accumulating an increasing body of evidence on language production across the lifespan (Sankoff & Blondeau 2007, Wagner 2012, inter alia), we have yet to understand how perception is influenced by speaker and listener age. Given that previous research found effects for listener’s cognitive and attitudinal disposition (Buchstaller & Levon 2015, Wagner & Hesson 2014), what exactly is the impact of hearer’s cognitive backgrounds in the way they hear, notice and evaluate variability in language? My project aims to address these vital questions with an eye on the perceptual aspects of ongoing language change, including the question of how we can utilize perception data to explain patterns in production data and vice versa. Data collection and analysis are conducted in cooperation with the Language and Variation across the Lifespan (LaVaLi) project, which investigates lifespan change using trend and panel data from the North East of England.

The study focuses on Tyneside English, a variety that is well described regarding the processes of variation and change in production (Watt 2002, Buchstaller & Corrigan 2015, inter alia). To date, however, the variety has only been subject to relatively few perception experiments, none of which examines the relevance of speaker and listener age for sociolinguistic perception of change in progress (Buchstaller & Levon 2015, Levon, Buchstaller, Mearns 2020). This study addresses this gap by exploring the relevance of maturation in the perception of stimuli-rich speech samples from the North East of England.

The study draws on naturally occurring, stimulus-rich sound samples from four speakers who differ in age and gender (young/old, male/female). Careful sampling of the large LaVaLi corpus, that contains approximately 100 hours of sociolinguistic interviews, ensures that the four guises were stringently controlled for sound quality, fundamental frequency as well as voice quality. Each speech stimulus is about forty seconds in length.

Using an online survey surface (Qualtrics), the samples were played to two listener groups in an inter-speaker design: half of the informants listening to the stimuli were asked to rate the speakers on a “professionalism” scale in the context of a job application. The other half of informants were told that the speakers were introducing themselves to a local volunteer group via voice message (online due to the ongoing pandemic). This listener group was then asked to respond according to perceived friendliness. Finally, participants were asked to answer questions regarding their attitudes towards the North East varieties (e.g. North, North East, Newcastle) and – to explore the impact of cognitive factors – they filled out the diagnostic questions of the BAPQ (Hurley et al. 2007). In a mixed effects regression model, the effects of age, gender, attitudes, and results of the BAPQ were tested.

Results suggest that the perception of vernacular linguistic features is highly contingent on both speaker and listener age. These findings allow us to develop a more holistic understanding of the cognitive underpinnings of age-related variability in language perception and the impact of such findings for language change across the lifespan.