Table 2:
Summary of Infant and WOLVES model performance in a canonical CSWL task
| Measure | Smith & Yu (2008) | Yu & Smith (2011) | Range | WOLVES | RMSE | MAPE | |
|---|---|---|---|---|---|---|---|
| TEST TRIALS | |||||||
| Mean looking time per 8s trial | 6.10 | 5.92 | 5.92 – 6.10 | 6.26 | 0.26 | 4.22 | |
| Preferential looking time ratio | 0.60 | 0.54 | 0.54 – 0.60 | 0.54 | 0.04 | 6.10 | |
| Mean words learned (of 6) | 4.0 | 3.5 | 3.5 – 4 | 4.0 | 0.35 | 7.14 | |
| Proportion of Strong (S) vs Weak (W) Learners | N/A | 0.67 | N/A | 0.74 | 0.07 | 10.45 | |
| Mean looking per trial to Target | 3.6 | 3.25 | 3.25 – 3.6 | 3.36 | 0.19 | 5.03 | |
| Mean looking per trial to Distractor | 2.5 | 2.67 | 2.5 – 2.67 | 2.89 | 0.32 | 11.92 | |
|
| |||||||
| TRAINING TRIALS | S | W | |||||
| Mean looking time per 4s trial | 3.04 | 2.96 | 3.07 | 2.96 – 3.07 | 3.01 | 0.02 | 0.71 |
| Mean fixations per trial | N/A | 2.75 | 3.82 | 2.75 – 3.82 | 2.89 | 0.22 | 6.98 |
| Mean fixation duration | N/A | 1.69 | 1.21 | 1.21 – 1.69 | 1.31 | 0.22 | 14.38 |