Sources of Microtemporal Clustering in Sociolinguistic Sequences

Area of Science:

• Sociolinguistics
• Psycholinguistics
• Corpus Linguistics

Background:

• Persistence in speech refers to repeating a sociolinguistic variant choice.
• A key debate questions if this repetition stems from direct sequential influence or shared contextual factors.

Purpose of the Study:

• To differentiate between sequential dependence and baseline deflection in explaining speech persistence.
• To analyze longer sequences of sociolinguistic variation beyond typical prime-target pairs.

Main Methods:

• Analyzed corpus data of conversational speech, specifically the /ing/ variant.
• Developed simulations using Bernoulli trials to model sequential dependence and baseline deflection mechanisms.
• Compared model predictions with corpus data properties.

Main Results:

• Identified two key effects: the frequency of the /ing/ variant in prior tokens and the status of the immediately preceding token.
• Found that models incorporating both baseline deflection and sequential dependence best replicate observed corpus data patterns.

Conclusions:

• Both sequential dependence and baseline deflection appear to be active mechanisms in producing sociolinguistic variation.
• This exploratory analysis of longer sequences offers a promising avenue for future research into speech production mechanisms.