Table 1.
Fixed effects for RT model.
| Fixed effect | Estimate (ms) | Standard error (ms) | t-value |
|---|---|---|---|
| Intercept | 722.0 | 18.6 | 38.9 |
| Phoneme (plosive) | −53.5 | 16.2 | −3.3 |
| Experiment (2) | −3.8 | 22.7 | −0.2 |
| Cue (non-predictive) | 141.9 | 19.9 | 7.5 |
| Pronunciation (wrong) | 15.8 | 9.3 | 1.7 |
| Phoneme (plosive):Experiment (2) | −3.7 | 5.2 | −0.7 |
| Phoneme (plosive):Cue (non-predictive) | 1.9 | 22.5 | 0.1 |
| Phoneme (plosive):Pronunciation (wrong) | 4.5 | 4.9 | 0.9 |
| Experiment (2):Cue (predictive) | −3.1 | 13.5 | −0.2 |
| Experiment (2):Pronunciation (wrong) | 5.0 | 11.3 | 0.4 |
| Cue (non-predictive):Pronunciation (wrong) | 19.4 | 6.5 | 3.0 |
Highlighted rows indicate significant (|t| > 2) and marginally significant predictors in final model. The predictors, all bivariate categorical variables were coded using treatment coding, making plosive, predictive, correct, Experiment 1 trials as the reference level. Final model specification: RT ~ Phoneme *(Experiment + Cue + Pronunciation) + Experiment* (Cue + Pronunciation) + Cue * Pronunciation + (1 + Phoneme + Experiment + Cue + Pronunciation + Cue * Pronunciation | subject) + (1 | Cue image).