TABLE 1.
Performance During Word Encoding and Recognition Tasks in Patients With Schizophrenia and Healthy Comparison Subjects
| Task and Performance Measurea | Patients (N=14)
|
Comparison Subjects (N=14)
|
Analysis
|
|||
|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | F (df=1, 26) | p | |
| Shallow word classification | ||||||
| Correct button presses | 34.57 | 5.93 | 38.93 | 1.27 | 9.05 | 0.006 |
| Reaction time (msec) | 1513.50 | 291.94 | 1271.43 | 227.11 | 6.00 | 0.02 |
| Deep word classification | ||||||
| Correct button presses | 33.00 | 7.24 | 37.00 | 2.32 | 3.55 | 0.07 |
| Reaction time (msec) | 1339.36 | 346.09 | 1080.29 | 261.18 | 5.00 | 0.03 |
| Shallow recognitions | ||||||
| Discriminability scoreb | 0.25 | 0.13 | 0.25 | 0.16 | 0.01 | 0.92 |
| Response bias scoreb | 0.18 | 0.12 | 0.16 | 0.08 | 0.14 | 0.71 |
| True positive (correct “old” responses) | 7.43 | 3.08 | 7.21 | 3.40 | 0.03 | 0.86 |
| False negative (incorrect “new” responses) | 9.86 | 3.88 | 12.43 | 3.30 | 3.57 | 0.07 |
| Reaction time (msec) | 1478.50 | 369.94 | 1326.89 | 297.78 | 1.43 | 0.24 |
| Deep recognition | ||||||
| Discriminability score | 0.51 | 0.20 | 0.61 | 0.23 | 1.34 | 0.26 |
| Response bias score | 0.29 | 0.20 | 0.39 | 0.23 | 1.44 | 0.24 |
| True positive (correct “old” responses) | 13.07 | 3.95 | 14.71 | 5.25 | 0.87 | 0.36 |
| False negative (incorrect “new” responses) | 5.43 | 3.63 | 4.50 | 4.29 | 0.38 | 0.54 |
| Reaction time (msec) | 1407.75 | 305.59 | 1203.00 | 369.61 | 2.55 | 0.12 |
| Distractor word detection | ||||||
| True negative (correct “new” responses) | 13.79 | 4.95 | 15.14 | 2.44 | 0.85 | 0.37 |
| False positive (incorrect “old” responses) | 4.93 | 4.27 | 4.29 | 2.09 | 0.26 | 0.62 |
The Shallow/Deep Word Encoding and Recognition Task (11), adapted for event-related functional magnetic resonance imaging, was used to test word encoding and recognition. The test consisted of 40 shallow and 40 deep target words from a standard word list. During encoding, participants judged whether words were in uppercase or lowercase letters (shallow) or if they were abstract or concrete (deep) and made a left or right button press every time a word was presented. The recognition task combined 20 shallow and 20 deep target stimuli with 20 novel stimuli, and participants were required to press a left button to indicate that the word was from the encoding list (“old”) or a right button if it was not from the encoding list (“new”).
Recognition discriminability (recognition accuracy) and response bias were calculated by using the Two-High Threshold Theory (32). A higher discriminability score indicates greater accuracy. A response bias >0.5 was considered a liberal bias; a response bias <0.5 was considered a conservative bias.