Comparison of Parkinson's disease patients and controls in neuropsychological tests
| Parkinson's disease patients, n=25, mean (SD) | Controls, n=25, mean (SD) | One-tailed p-value* | |
| Global cognitive assessment | |||
| Trail making test | |||
| Part A times (s) (1) | 44.6 (16.8) | 45.2 (15.8) | 0.433 |
| Part B times (s) (2) | 91.9 (38.1) | 75.9 (28.3) | 0.090 |
| Trail Making Test (B–A) (2)–(1) | 47.3 (36.9) | 30.8 (20.6) | 0.084 |
| Stroop test†—inhibition cost index | 66.0 (33.1) | 53.8 (22.0) | 0.097 |
| Benton Visual Retention Test | 12.8 (1.5) | 13.4 (1.3) | 0.054 |
| Baddeley's Working Memory dual task | |||
| Digit Span (n) | 5.6 (0.9) | 6.0 (1.2) | 0.092 |
| Digit span task—digits sequences recalled (n) | 6.8 (4.2) | 7.6 (4.1) | 0.225 |
| Updating: n-back task | |||
| 0-back | |||
| Response times (ms) | 418.2 (60.1) | 411.4 (59.4) | 0.688 |
| Errors‡ | 0.7 (1.5) | 0.2 (0.4) | 0.261 |
| 1-back | |||
| Response times (ms) | 464.8 (80.2) | 425.7 (61.0) | 0.058 |
| Errors‡ | 0.7 (1.7) | 0.1 (0.2) | 0.038 |
| 2-back | |||
| Response times (ms) | 733.5 (189.1) | 632.1 (133.8) | 0.033 |
| Errors‡ | 4.4 (1.2) | 1.4 (0.5) | 0.002 |
| Flexibility: plus–minus task | |||
| Addition list completion times (s) (5) | 60.4 (16.4) | 50.8 (12.7) | 0.013 |
| Subtraction list completion times (s) (6) | 83.7 (29.5) | 65.4 (22.8) | 0.005 |
| Shift list completion times (s) (7) | 92.7 (31.2) | 77.3 (19.6) | 0.043 |
| Shift cost (7)−[(6)+(5)]/2 | 20.7 (15.8) | 19.2 (14.5) | 0.392 |
| Information-processing speed: Stroop test† | |||
| Colour naming completion time (s) | 65.0 (9.5) | 62.2 (7.1) | 0.217 |
Variables in italics are the variables of interest. Values in bold are significant.
↵* Student t test or Mann–Whitney U test. One-tailed analyses were applied in accordance with our hypothesis: we expected Parkinson's disease patients to be significantly impaired compared with controls. Two-tailed analyses did not alter our main findings (data not shown).
↵† One patient was excluded from the analysis because the data could not be analysed.
↵‡ Errors were calculated by the addition of the number of hits and false alarms.
Variables in italics are the variables of interest.