Table 2.
The percentage of correct recall as predicted by the final model across target-to-masker-ratio (TMR) conditions for both groups.
| Plogis transformed model estimates for the % of correctly recalled words by both groups | ||
|---|---|---|
| TMR condition (dB) | Musicians (%) | Non-musicians (%) |
| −3 | 92.4 | 85.8 |
| −5 | 82.7 | 70.3 |
| −7 | 63.7 | 46.5 |
| −9 | 35.6 | 21.5 |