<b>FACILITATING SENTENCE PROCESSING IN APHASIA: EVIDENCE FROM SELF-PACED READING</b>

It is well known that persons with aphasia (PWA) demonstrate deficits in sentence processing. Specifically, many show difficulties with syntactic re-analysis, or the ability to revise one’s interpretation of a sentence due to a temporary ambiguity. Emerging evidence suggests that structural priming, individuals’ tendency to unconsciously re-use a previously encountered structure, can facilitate production and comprehension of different sentence structures in PWA. However, there are inconsistencies within the literature regarding the robustness of priming across different sentence structures and tasks, and the existing findings are limited to off-line measures which do not capture real-time sentence processing. Furthermore, a range of individual variability has been shown in magnitudes of priming effects.This current project addressed these critical gaps by investigating the effects of structural priming on comprehension of sentences with reduced relative (RR) clauses. Study 1 examined whether PWA show immediate structural priming and longer-term cumulative priming in a computerized self-paced reading paradigm. In this study, participants read RR target sentences that were preceded by prime sentences with either an RR or a main clause (MC) interpretation. Study 2 examined if and how individuals’ cognitive (memory) skills account for individual variability shown in structural priming. Participants completed a Serial Reaction Time task as a measure of implicit memory, and the Verbal Paired Associates I task as a measure of explicit memory. Eighteen healthy adults (HA) and 18 PWA participated in both studies.In Study 1, PWA demonstrated robust and timely immediate structural priming effect, providing novel evidence that structural priming can be captured in real-time using self-paced reading, and that priming can help ameliorate syntactic re-analysis deficits in PWA. However, these online changes were not reflected in the behavioral measures, highlighting the sensitivity of self-paced reading in detecting the dissociation between automatic, real-time processing and overt behavioral responses. Cumulative priming as traditionally defined was also not significant for either group, although participants became faster at reading RR target sentences as the session progressed. Study 2’s findings suggest that implicit memory may be an important cognitive skill to promote sentence processing, whereas having strong explicit memory may not be facilitative of the learning process.