The Discriminative Lexicon: A Unified Computational Model for the Lexicon and Lexical Processing in Comprehension and Production Grounded Not in (De)Composition but in Linear Discriminative Learning

Table 7

Statistics for the partial effects of a generalized additive mixed model fitted to the relative duration of edge segments in the Buckeye corpus. The trend for log frequency is positive accelerating. TPRS: thin plate regression spline.

A. parametric coefficients	Estimate	Std. Error	t-value	p-value
Intercept	-1.9165	0.0592	-32.3908	0.0001
Log LDL Edge Weight	-0.1358	0.0400	-3.3960	0.0007
Phrase Rate	0.0047	0.0021	2.2449	0.0248
Log Ncount	0.1114	0.0189	5.8837	0.0001
B. smooth terms	edf	Ref.df	F-value	p-value
TPRS log Frequency	1.9007	1.9050	7.5495	0.0004
random intercepts word	1149.8839	1323.0000	31.4919	0.0001
random intercepts speaker	30.2995	39.0000	34.5579	0.0001