Please join me on zoom at umd.zoom.us/my/hmuller to discuss this poster between 12noon and 2pm on Thursday March 19, 2020.
Negative polarity illusions were once taken as a prime example of the role
of memory retrieval operations in sentence processing. However, recent work
shows that the illusion profile is more restricted than this account
predicts. We show the division of labor between early interpretive
processes and processes that occur at the NPI in creating NPI illusions. We
consider two hypotheses for how prior material could be (mis)represented
–which focus, respectively, on the scope of negative quantifiers and on the
recency of the relative clause meaning –and demonstrate in two experiments
that only the second captures the profile of the illusion.