Individuals with aphasia often exhibit difficulty producing accurate complex syntactic structures (agrammatism). Syntactic priming—the tendency to use a recently heard syntactic structure—is an avenue for therapeutic intervention in this population. Traditional priming paradigms rely on the implicit prediction error inherent in a repetition task. Recently, Jackson and colleagues (Gruter et al., 2021; under review; Jackson & Gruter, under review) have shown that L2 learners benefit more from a task that more explicitly induces prediction errors—guessing what their interlocutor will say—than from the traditional repetition priming. In this talk, Sandberg will discuss their recent work extending this paradigm to monolingual older adults and people with aphasia to examine the effect of increased prediction error on syntactic priming. The results of this work are expected help in the development of more effective therapeutic techniques for agrammatism in aphasia.