Table 2:

Model set fixed effects structure

Model #	Model hierarchy	SES and SES^2	ESL	URG	Sex	Age	WRAT	APOE	Age^2	Sex * Age	Sex * Age^2	WRAT * Age	WRAT * Age^2	APOE * Age	APOE * Age^2	Sex * WRAT	Sex * APOE	Sex * WRAT * Age	Sex * WRAT * Age^2	Sex * APOE * Age	Sex* APOE * Age^2
1	baseline model	x	x	x	x	x
2	M1 + WRAT	x	x	x	x	x	x
3	M1 + APOE	x	x	x	x	x		x
4	M1 + Age^2	x	x	x	x	x			x
5	M1 + WRAT + APOE	x	x	x	x	x	x	x
6	M1 + WRAT + Age^2	x	x	x	x	x	x		x
7	M1 + APOE + Age^2	x	x	x	x	x		x	x
8	M1 + WRAT + APOE + Age^2	x	x	x	x	x	x	x	x
9	M1 + Sex * Age	x	x	x	x	x				x
10	M9 + Sex * Age^2	x	x	x	x	x			x	x	x
11	M2 + WRAT * Age	x	x	x	x	x	x					x
12	M11 + WRAT * Age^2	x	x	x	x	x	x		x			x	x
13	M3 + APOE * Age	x	x	x	x	x		x						x
14	M13 + APOE * Age^2	x	x	x	x	x		x	x					x	x
15	M9 + WRAT * Age	x	x	x	x	x	x			x		x
16	M9 + APOE * Age	x	x	x	x	x		x		x				x
17	M11 + APOE * Age	x	x	x	x	x	x	x				x		x
18	M15 + APOE * Age	x	x	x	x	x	x	x		x		x		x
19	M2 + Sex * WRAT	x	x	x	x	x	x									x
20	M3 + Sex * APOE	x	x	x	x	x		x									x
21	M9 + Sex * WRAT	x	x	x	x	x	x			x						x
22	M9 + Sex * APOE	x	x	x	x	x		x		x							x
23	M19 + Sex * APOE	x	x	x	x	x	x	x								x	x
24	M21 + Sex * APOE	x	x	x	x	x	x	x		x						x	x
25	M1 + Sex * WRAT * Age	x	x	x	x	x	x			x		x				x		x
26	M1 + Sex * WRAT * Age^2	x	x	x	x	x	x		x	x	x	x	x			x		x	x
27	M1 + Sex * APOE * Age	x	x	x	x	x		x		x				x			x			x
28	M1 + Sex * APOE * Age^2	x	x	x	x	x		x	x	x	x			x	x		x			x	x

Note: Models represent several possible hypothesized associations between sex, age, WRAT, and APOE. For example, Model 1 includes within person random effects (intercept and age-related slope) and SES, ESL, URG, sex and age. Model 1 terms are included in all other models. Given the potential for differences in age-related trajectories, Models 9, 11, and 13 examine whether sex, WRAT, or APOE influence age-related change in a linear way while models 10, 12, and 14 examine non-linear (i.e., quadratic) differences.