Proceedings of the Annual Meeting of the Cognitive Science Society

EARSHOT is a model of human speech processing that learns to map real speech to semantic representations (Magnuson et al., 2020). We report first steps towards simulating aspects of aphasia with EARSHOT by damaging increasing proportions of randomly selected weights in different model layers. Although the model is purely receptive, we simulated naming/identification tasks by presenting spoken words to damaged models and measuring the cosine similarity of the output to every word in the lexicon, with the model's naming/identification response operationalized as the word with highest cosine similarity. For errors, we evaluated phonetic and semantic similarity of the response to the target vector. We were interested in how robust EARSHOT would be to damage, and whether we might observe systematic patterns of phonetic vs. semantic deficits following damage to different model components. As expected, lower-level damage impaired phonology most, and higher-level damage most affected semantics. Further development could turn EARSHOT into a valuable tool for enhancing understanding of aphasia.