Table 1. Best-fitting parameter estimates, shown as median plus quartiles across conditions.
| γ* | Δ+mf | Δ+mb | Δ_mf** | Δ_mb | |
|---|---|---|---|---|---|
| CONTROL | |||||
| 25th | 0.23 | 0.70 | 0.10 | 0.12 | -0.41 |
| Median | 0.47 | 1.09 | 0.33 | 0.48 | -0.14 |
| 75th | 0.60 | 1.67 | 0.80 | 0.78 | 0.04 |
| T | 8.41** | 3.30* | 2.00 | 5.17** | -2.74 |
| STRESS | |||||
| 25th | 0.56 | 0.04 | -0.23 | -0.10 | -0.13 |
| Median | 0.73 | 0.50 | -0.04 | 0.02 | -0.01 |
| 75th | 0.95 | 1.06 | 0.18 | 0.14 | 0.11 |
| T | 11.50** | 3.84* | .02 | .62 | -.89 |
Notes: Δ+mb and Δ_mb are parameters which represent the model-based tendency after reward and no-reward, respectively. Δ+mf and Δ+mf are parameters which indicate the model-free tendency after reward and no-reward, respectively.
* p < 0.01
** p < 0.001. T is the t value from the paired t-test which was performed to investigate whether each parameter was significantly different from zero.