Home > Research > Publications & Outputs > Atypically-reading adults: a profile

Electronic data

  • 2024MillsPhD

    Final published version, 13.8 MB, PDF document

    Available under license: CC BY-NC-ND: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

Text available via DOI:

View graph of relations

Atypically-reading adults: a profile: an exploratory, longitudinal study of single word recognition processes

Research output: ThesisDoctoral Thesis

Publication date2024
Number of pages456
Awarding Institution
Thesis sponsors
  • Economic and Social Research Council
Award date28/03/2024
  • Lancaster University
<mark>Original language</mark>English


Approximately 16% of school leavers cannot read to a sufficient skill level so
as to be called “functionally literate” (Castles et al., 2018; Leitch, 2006). This
exploratory study explores the single word recognition processes of a group of
atypically-reading adults in comparison with groups of younger and older readers.
In the main study, we assessed orthographic, phonological and semantic skills
longitudinally. We estimated their influence plus that of psycholinguistic properties such as word-frequency, consistency and neighbourhood-size on single word recognition processes by way of reaction time and accuracy data from four experimental tasks (letter search, lexical decision, single word naming and sentence reading).
To support the estimation of our statistical models for the main study, we
conducted a wide ranging meta-analysis of psycholinguistic predictor effects. We
report the findings here and introduce the study as an accessible resource for use by the research community.
Linear-mixed-effects-models estimated that the rate of change in reading-related skills was either too small or too slow to detect within the time-frame
or data. Adult-learners perform similarly to all comparison groups in response
latencies across all tasks. They perform similarly to 11-12- and 16-17-year-old readers in the lexical decision and sentence reading accuracy measures. They are more accurate in letter search and less accurate in word naming accuracy measures.
Nonword reading skill, rather than word reading skill, is a reliable predictor
in this sample. Word-frequency, age-of-acquisition, consistency and neighbourhood size show influence across tasks.
We interpret the results through the lexical quality hypothesis. The predictors that are influential across the models, and the similarity of adult-learners’
performance to younger readers suggests that their orthographic, phonological and semantic knowledge is weakly correlated. Further, adult-learners may be using a dominant reading strategy that reflects sublexical processing, thereby impeding development of orthographic learning and knowledge over the longer term.