Skip to main content
. Author manuscript; available in PMC: 2015 Dec 1.
Published in final edited form as: Cogn Psychol. 2014 Aug 27;0:1–27. doi: 10.1016/j.cogpsych.2014.07.003

Table 4.

Results of logistic regression in Experiment 3. The intercept indicates subjects were at chance in choosing novel-grammatical sentences over novel-ungrammatical sentences. The significant trial type effect demonstrates that subjects were more likely to choose repeated over scrambled sentences than they were to choose novel-grammatical over novel-ungrammatical.

Estimate Std. Error Wald’s z p-value
Intercept (novel-grammatical vs. novel-
ungrammatical)		 −.01053 .11512 −.091 .927
Trial Type (contrasting repeated vs. novel
against intercept)		 .67053 .15004 4.469 7.9×10−6