Table 4. Generalized additive model fitted to simplex words from the English Lexicon Project using discrimination-based predictors.

A. parametric coefficients	Estimate	Std. Error	t-value	p-value
Intercept	-1.3186	0.0024	-543.1538	< 0.0001
Age = young	-0.3224	0.0034	-93.9030	< 0.0001
B. smooth terms	edf	Ref.df	F-value	p-value
te(G2L prior, G2L a-diversity)	3.0008	3.0016	44.5369	< 0.0001
te(L2L prior, L2L l-diversity): Age = old	4.5693	5.3621	5.3898	< 0.0001
te(L2L prior, L2L l-diversity): Age = young	6.9382	8.3329	17.1926	< 0.0001
te(sem-typicality, sem-density)	7.1147	8.8957	5.4650	< 0.0001
s(G2L activation): AgeSubject = old	2.4596	3.1492	2.6914	0.0430
s(G2L activation): AgeSubject = young	1.0003	1.0005	8.0608	0.0045
s(Written Spoken Frequency Ratio)	4.2019	5.2811	8.1942	< 0.0001
s(Concept Frequency)	8.1445	8.8031	27.3368	< 0.0001

te: tensor product smooth, s: thin plate regression spline smooth. (AIC = −6220.1; -ML = −3106.6; R-sq. (adj) = 0.752)