Table 1.
Goodness‐of‐Fit Statistics for the Latent Markov Models of the Conflicting Switching Inhibition and Flexibility Task
| Model | LL | npar | AIC | BIC | BICw |
|---|---|---|---|---|---|
| 1 state | −619.57 | 1 | 1,241.15 | 1,242.80 | 0.00 |
| Regression (age) | −608.44 | 2 | 1,220.80 | 1,224.19 | 0.00 |
| 2‐state | −517.68 | 5 | 1,045.36 | 1,053.60 | 0.00 |
| 2‐state + age | −515.55 | 6 | 1,043.10 | 1,052.99 | 0.00 |
| 3‐state | −508.17 | 9 | 1,034.34 | 1,049.18 | 0.00 |
| 3‐state + age | −500.85 | 9 | 1,019.70 | 1,034.54 | 0.94 |
| 4‐state | −505.51 | 10 | 1,031.02 | 1,047.51 | 0.00 |
| 4‐state + age | −494.62 | 14 | 1,017.24 | 1,040.32 | 0.05 |
The abbreviation “npar” denotes the number of parameters; it is corrected for parameters estimated at their boundary values (0 or 1). The small‐sample version of the BIC is reported here. Finally, BICw denotes the BIC weights. AIC = Akaike's information criterion; BIC = Bayesian information criterion.