Transformer language models reveal distinct patterns in aphasia subtypes and longitudinal change

Abstract Language impairments in aphasia are characterized by various representational disruptions that may be reflected in discourse production. This research examines the capacity of transformer-based language models, particularly GPT-2, to serve as a computational framework for analyzing variations in aphasic narrative speech. A longitudinal dataset of narrative speech samples collected at six time points from individuals with aphasia (N = 47; baseline cross-sectional sample N = 57) was utilized as part of an intervention study. All transcripts were processed via the GPT-2 language model to obtain activation values from each of the 12 transformer layers. Statistically significant differences in activation values across aphasia subtypes were found at every layer (all p < .001), with the most pronounced effects in the deeper layers. Pairwise Tukey HSD tests revealed consistent distinctions between Broca’s aphasia and both Anomic and Wernicke’s aphasia, suggesting a shared activation profile between the latter two. Longitudinal tests revealed significant changes over time, especially in the final three layers (10–12). These findings suggest that transformer-based activation patterns reflect meaningful variation in aphasic discourse and may align with clinical observations. Overall, GPT-2 provides a scalable tool to model representational dynamics in aphasia and may offer exploratory insights into language representation in aphasic discourse.