Table 1.
Confusion matrix for sinewave vowels in experiment 1, prior to training. The response means given in the last row indicate the percentage of trials in which a given vowel was used as a listener response. Each vowel was presented on 8.3% of the trials.
| Vowel identified by listener | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ∕i/ | ∕ɪ∕ | ∕e/ | ∕ɛ∕ | ∕æ∕ | ∕ɑ∕ | ∕ɔ∕ | ∕o/ | ∕ʊ∕ | ∕u/ | ∕ʌ∕ | ∕ɚ∕ | ||
| Vowel intended by talker | ∕i/ | 76.7 | 3.9 | 3.1 | 1.4 | 0.5 | 0.5 | 0.2 | 1.3 | 2.3 | 9.3 | 0.9 | 0.0 |
| ∕ɪ∕ | 12.8 | 44.7 | 2.3 | 8.5 | 2.9 | 1.6 | 0.7 | 1.7 | 4.4 | 12.8 | 5.6 | 2.0 | |
| ∕e/ | 40.8 | 2.8 | 41.0 | 1.8 | 0.8 | 0.8 | 0.4 | 2.1 | 1.5 | 6.4 | 0.6 | 1.0 | |
| ∕ɛ∕ | 2.6 | 17.1 | 8.2 | 30.4 | 9.1 | 3.4 | 1.7 | 7.3 | 6.3 | 6.0 | 5.6 | 2.2 | |
| ∕æ∕ | 1.5 | 8.3 | 6.9 | 16.3 | 36.9 | 2.4 | 3.8 | 4.7 | 2.3 | 5.3 | 2.1 | 9.6 | |
| ∕ɑ∕ | 0.6 | 0.8 | 2.9 | 2.5 | 5.5 | 33.1 | 15.5 | 5.0 | 5.0 | 2.1 | 17.7 | 9.3 | |
| ∕ɔ∕ | 0.1 | 0.3 | 0.4 | 0.5 | 1.1 | 12.8 | 52.3 | 19.4 | 3.5 | 2.0 | 6.3 | 1.2 | |
| ∕o/ | 0.0 | 0.0 | 0.1 | 0.2 | 0.1 | 0.6 | 0.6 | 69.7 | 2.6 | 24.7 | 0.7 | 0.7 | |
| ∕ʊ∕ | 0.2 | 1.2 | 1.4 | 1.3 | 0.7 | 1.2 | 1.0 | 1.5 | 66.7 | 11.3 | 10.9 | 2.5 | |
| ∕u/ | 0.1 | 0.1 | 0.1 | 0.1 | 0.0 | 0.2 | 0.1 | 12.9 | 8.1 | 75.9 | 2.1 | 0.5 | |
| ∕ʌ∕ | 0.2 | 1.0 | 1.1 | 2.1 | 1.4 | 3.7 | 3.5 | 7.2 | 33.5 | 5.4 | 39.8 | 1.3 | |
| ∕ɚ∕ | 0.3 | 0.1 | 0.7 | 0.5 | 0.2 | 0.2 | 0.1 | 0.7 | 1.4 | 2.7 | 0.8 | 92.4 | |
| Response means: | 11.3 | 6.9 | 5.7 | 5.5 | 4.9 | 5.0 | 6.7 | 11.1 | 11.5 | 13.7 | 7.8 | 10.2 | |