Table 2 Computational model parameters and model-free task metrics by group and resistance condition.

From: Computational mechanisms underlying multi-step planning deficits in methamphetamine use disorder

Resistance Condition

HCs (n = 49)

iMUDs (n = 40)

Effect of Group

Contrast

Aversive Pruning (π)

F(1,100) = 16.46, p < 0.001, \({\eta }_{p}^{2}=0.14\)

iMUD = 0.33, HC = 0.15; t(100) = 4.06, p<0.001, d = 0.81

 No resistance

0.18 (0.21)

0.35 (0.19)

 Resistance

0.19 (0.21)

0.29 (0.18)

Large-Loss Discounting Probability \((1-{\gamma }_{S})\)

F(1,100) = 13.45, p < 0.001, \({\eta }_{p}^{2}=0.12\)

iMUD = 0.49, HC = 0.33; t(100) = 3.67, p<0.001, d = 0.74

 No resistance

0.35 (0.21)

0.50 (0.18)

 Resistance

0.36 (0.19)

0.45 (0.17)

No Large-Loss Discounting Probability \((1-{\gamma }_{G})\)

F(1,100) = 1.67, p = 0.199

ns

 No resistance

0.17 (0.06)

0.15 (0.06)

 Resistance

0.17 (0.06)

0.16 (0.06)

Reward Sensitivity (β)

F(1,100) = 3.01, p = 0.086

ns

 No resistance

0.08 (0.04)

0.06 (0.03)

 Resistance

0.07 (0.03)

0.06 (0.03)

Total Points Won

F(1,100) = 14.81, p < 0.001, \({\eta }_{p}^{2}=0.13\)

iMUD = 2034, HC = 3145; t(100) = 3.85, p<0.001, d = 0.77

 No resistance

3009.18 (1230.88)

2114.00 (1406.69)

 Resistance

3004.29 (1021.07)

2117.25 (1403.15)

Percent OLL Trials Correct

F(1,490) = 25.30, p < 0.001, \({\eta }_{p}^{2}=0.05\)

iMUD = 0.19, HC = 0.35; t(99) = 3.99, p<0.001, d = 0.45

 No resistance

Depth 3

0.44 (0.32)

0.26 (0.22)

 Resistance

0.41 (0.31)

0.26 (0.23)

 No resistance

Depth 4

0.38 (0.24)

0.26 (0.20)

 Resistance

0.34 (0.22)

0.25 (0.18)

 No resistance

Depth 5

0.16 (0.16)

0.10 (0.10)

 Resistance

0.18 (0.14)

0.11 (0.09)

Percent ONLL Trials Correct

F(1,473) = 0.33, p = 0.568

ns

 No resistance

Depth 3

0.87 (0.17)

0.84 (0.20)

 Resistance

0.91 (0.11)

0.84 (0.23)

 No resistance

Depth 4

0.78 (0.26)

0.77 (0.24)

 Resistance

0.79 (0.23)

0.70 (0.26)

 No resistance

Depth 5

0.65 (0.30)

0.59 (0.28)

 Resistance

0.61 (0.30)

0.55 (0.27)

  1. Columns on the right indicate statistics for the group effect in LMEs predicting each behavioral metric based on group, resistance condition, and their interaction. Results for contrasts show estimated marginal means (EMMs) for each group as well as a t-test comparing those values. These contrast results are only shown when the effect of group was significant.
  2. ns nonsignificant, OLL optimal large-loss, ONLL optimal no large-loss.