Table 2.

Environment rewards across endophenotypes controlled by the RL model

Transition Description	Probability for each phase
	P (f1)	P (f2)	P (f3)	P (f4)
T(s=i\|s=i,a=a^s=i), i neutral state	0	0	0	0	From States
T(s=i+j\|s=i,a=a^s=i+j), j=+1/-1, i neutral state, i+j neutral state	0	0	0	0
T(s=i\|s=i,a=a^s=i+j), j=+1/-1, i neutral state, i+j neutral state	0	0	0	0
T(s=i+k\|s=i,a=a^s=i+k),k!=+1/-1, i neutral state, i+k neutral state	-0.3	-0.3	-0.3	-0.3
T(s=i\|s=i,a=a^s=i+k),k!=+1/-1, i neutral state, i+k neutral state	0	0	0	0
T(s=i\|s=i,a=a^w), i neutral state	0	0	0	0
T(s=1\|s=2,a=a^g)	0	0	0	0
T(s=i\|s=i,a=a^g), i!=2 neutral state	0	0	0	0
T(s=8\|s=7,a=a^d)	0	10	-1	10
T(s=i\|s=i,a=a^d), i!=7 neutral state	0	0	0	0
T(s=i\|s=i,a=a^g), i drug/aft state	-0.3	-1.2	-1.2	-1.2	From Drug/aft States
T(s=4\|s=i,a=a^g), i drug/aft state	-4	-4	-4	-4
T(s=i\|s=i,a=a^s=*), i drug/aft state	-0.3	-1.2	-1.2	-1.2
T(s=4\|s=i,a=a^s=*), i drug/aft state	-4	-4	-4	-4
T(s=j\|s=i,a=a^w), i!=15 drug/aft state, j next or previous drug/aft state	-0.3	-1.2	-1.2	-1.2
T(s=4\|s=i,a=a^w), i!=15 drug/aft state	-4	-4	-4	-4
T(s=14/16\|s=15,a=a^w)	-0.3	-1.2	-1.2	-1.2
T(s=4\|s=15,a=a^w)	-4	-4	-4	-4
T(s=j\|s=i,a=a^d), i drug/aft state, j next drug/aft state	-0.3	-1.2	-1.2	-1.2
T(s=j\|s=i,a=a^d), i drug/aft state, j previous drug/aft state	-0.3	-1.2	-1.2	-1.2
T(s=4\|s=i,a=a^d), i drug/aft state	-4	-4	-4	-4
T(s=4\|s=1,a=a^g)	1	1	1	1	Goal
T(s=1\|s=1,a=a^s=*)	0	0	0	0
T(s=1\|s=1,a=a^w)	0	0	0	0
T(s=1\|s=1,a=a^d)	0	0	0	0