. Author manuscript; available in PMC: 2009 Nov 3.

Published in final edited form as: Cogn Sci. 2008 Jun 1;32(4):643–684. doi: 10.1080/03640210802066816

Table 1.

Results of three hierarchical linear models for the relationship between ambiguity^a, task, proximity and recency scores, and competitor fixations.

Model^b	Fixed Effects	coefficient	SE	t-ratio	df^c	p-value
#1	task	−.011851	.017426	−0.680	11	=.510
	proximity	−.025193	.006550	−3.847	11	<.01
	recency	.000044	.000045	0.972	11	=.352
	ambiguity	.046458	.018959	2.450	11	<.05
#2	task	.061205	.061159	1.001	11	=.339
	ambiguity	.127344	.045326	2.810	11	<.05
#3	task	−.039824	.014661	−2.716	11	<.05
	ambiguity	.022439	.024970	0.899	11	=.388

Ambiguity was coded as 0=ambiguous; 1=disambiguated; Task was coded as 0=predicted by the task; 1=not predicted by the task.

Model 1 included each eye-tracking trial (n=751). Model 2 included trials (n=130) on which there was a competitor block adjacent to the last mentioned block. Model 3 included trials (n=621) on which there was not a competitor block adjacent to the last mentioned block.

(Approximate).