Table 3.

Linear regression models.

Neuropsychological test sample (n = 125)	All (n = 125)		Men (n = 51)		Women (n = 74)
(a) Predicting Long-Term Memory Recognition	B (SE)	p	B (SE)	p	B (SE)	p
Constant	2.42 (0.3)	<0.01	2.55 (0.48)	<0.01	2.35 (0.39)	<0.01
Age	0 (0)	0.34	−0.01 (0.01)	0.28	0 (0.01)	0.70
Education	0.1 (0.02)	<0.01	0.1 (0.04)	0.01	0.1 (0.03)	<0.01
Depressive symptoms	−0.01 (0)	<0.01	−0.01 (0.01)	0.13	−0.02 (0.01)	0.00
Brokerage	0.07 (0.08)	0.41	0.01 (0.11)	0.93	0.12 (0.11)	0.29
Depressive symptoms × Brokerage	0.01 (0.01)	0.29	−0.04 (0.02)	0.04	0.05 (0.02)	0.01
Gender	−0.13 (0.06)	0.02
Depressive symptoms × Brokerage × Gender	−0.08 (0.02)	<0.01
MRI subsample (n = 65)	All (n = 65)		Men (n = 25)		Women (n = 40)
(b) Predicting adjusted left hippocampal volume	B (SE)	p	B (SE)	p	B (SE)	p
Constant	3800.98 (456.25)	<0.01	3694.14 (635.57)	<0.01	3884.84 (635.57)	<0.01
Age	−14.05 (6.36)	0.03	−14.78 (8.89)	0.09	−13.65 (8.89)	0.13
Depressive symptoms	−2.12 (5.95)	0.72	−9.79 (9.42)	0.24	5.66 (9.42)	0.55
Brokerage	−149.41 (121.72)	0.22	−366.58 (172.18)	0.03	−9.77 (172.18)	0.96
Depressive symptoms × Brokerage	19.84 (18.57)	0.29	−42.30 (25.56)	0.10	58.50 (25.56)	0.03
Gender	−292.76 (74.22)	<0.01
Depressive symptoms × Brokerage × Gender	−100.09 (38.23)	0.01

B = Unstandardized coefficient, SE = Standard error.

Brokerage: Social network structural constraint (range 0–1), a smaller value indicates a larger opportunity for brokerage; Social network size: number of social ties in the global network.