“…The experiments use a well-established sentence-completion task that is designed to elicit errors of subject-verb agreement in speakers' sentences. The task is the same as in most studies Bock & Cutting, 1992;Bock, Eberhard, Cutting, Meyer, & Schriefers, 2001;Bock & Miller, 1991;Eberhard, 1999;Gillespie & Pearlmutter, 2011; K. R. Humphreys & Bock, 2005;Solomon & Pearlmutter, 2004;Penta & Pearlmutter, 2015;Thornton & MacDonald, 2003;Vigliocco, Butterworth, & Garrett, 1996;Vigliocco, Butterworth, & Semenza, 1995): Participants are presented with test preambles, which they repeat or read aloud and for which they provide an ending of their own creation; participants' utterances are transcribed, coded, and analyzed; and error rates, calculated as the number of errors out of the total of number-inflected verbs produced in participants' responses, are computed and compared to quantify differences in effects across conditions.…”