Laurie Lawyer: Multiple Wavelengths of Language Processing
Complex word processing
The primary strand of my research looks at how the structure of complex words (such as those with prefixes or suffixes) impacts the way people learn, conceptualize, or perceive these words. The goal is to uncover how language users store and process information encoded within the structure of words. A number of projects are ongoing in this area, including recently published work (Lawyer et al. (2024)) looking at the conceptualization of noun class prefixes in Kinyarwanda (a Bantu language of Rwanda).
Closer to home, I have been developing methods to measure the processing of individual parts of words, such as specific prefixes or word roots, in complex words like unrelatable. Using a novel statistical approach to relating EEG neural responses to individual word parts, these methods provide evidence both for the existence of decomposition of spoken words in perception, and a way to trace the timecourse of specific sub-processes.
An offshoot of this project is a collaboration with Dr. Charles Redmon (Essex) which aims to create a very large, openly-available dataset of EEG recordings covering complex word processing across a significant portion of the English lexicon. To our knowledge, there are no publicly available data sets of word processing which include EEG data, let alone with data covering both visual and auditory processing of the same set of words. Although we are in early stages of this project, we anticipate it will make a significant contribution to the field, and open science more generally.
Language processing and memory
A strand of my work in collaboration with Dr. Willem van Boxtel (LSU) relates to exploring more generally how memory interacts with the processing of words and sentence structures, and particularly how this interacts with healthy aging. In van Boxtel & Lawyer (2024), we addressed memory for specific sentence structures and asked whether this kind of memory is stable in older adults, and showed that the answer to this question varies as a function of task. Despite what has been reported in studies using explicit measures where older adults need to recall specific sentence details, in our implicit studies where we measure reading times, older adults and younger adults show highly similar language processing profiles. These results added important nuance to our understanding of the interaction between language and memory in older adults. Although many studies have shown language processing abilities decline over time, our work shows some core processes are resistant to age-related change. A follow-on from this project aims to map out how long memory traces for sentence structures are active in reading, both in younger language users and in healthy older adults. We’re still in the early stages of analysis, but anticipate publishing results soon.
Language processing and linguistic bias
A project under development with a former PhD student, Dr. Xiaoduan Fan, addresses entrenched aspects of linguistic bias and how these affect language use and processing. We’re looking at gender-inclusive language in speakers of grammatically genderless languages (in this case, Mandarin). As a first step towards understanding how individuals process language differently when it agrees with or conflicts with their underlying social bias, a recent publication (Fan and Lawyer (2024)) explored to what degree Mandarin speakers are aware of and support using gender-fair language. Our findings showed resistance to truly gender-neutral language among older speakers and in males more generally, and involved producing the first measure for quantifying linguistic sexism in Chinese. This measure was used in a paper currently under review that explore asymmetries in the perception of explicitly gendered occupational terms (like 'male nurse' or 'female firefighter') in Mandarin. Looking forward, the next stages of this project will look at implicit attitudes towards gender-neutral language in Mandarin using implicit association tests to assess the degree to which violations of these biases disrupt sentence processing in real time.