Table 3.
Descriptive characteristics of DNT and TMB samples.
| Digital Neuropsychology Toolkit Sample |
TMB Normative Sample |
|||||
|---|---|---|---|---|---|---|
| Test Score | N | Mean (SD) | Split-Half Reliability | N | Mean (SD) | Split-Half Reliability |
| Forward Digit Span – Score | 1213 | 6.31 (1.82) | a | 9858 | 6.43 (1.58) | a |
| Backward Digit Span – Score | 1149 | 5.22 (2.04) | a | 6269 | 5.33 (1.58) | a |
| GradCPT – D prime | 2727 | 2.46 (0.92) | .86 | 20,197 | 2.88 (0.85) | 0.86 |
| GradCPT – Criterion | 2727 | 0.61 (0.50) | .86 | 20,197 | 0.82 (0.37) | 0.82 |
| Simple RT – RT (ms) | 1982 | 367.47 (146.0) | .99 | 49,668 | 303.21 (63.06) | 0.97 |
| Choice RT – Accuracyb | 2745 | .94 (.13) | .93 | 17,121 | .95 (.08) | 0.73 |
| Choice RT – Correct RT (ms)b | 2743 | 1172.86 (688.05) | .97 | 17,121 | 895.6 (257.47) | 0.96 |
| Digit Symbol Matching – Score | 2642 | 44.02 (13.33) | .90 | 36,000 | 51.87 (12.20) | 0.91 |
| Trail-Making Test Part A – Median RT (ms) | 2604 | 869.20 (405.01) | .87 | 21,278 | 691.09 (230.84) | 0.86 |
| Trail-Making Test Part A – Total Time (ms) | 2604 | 35,551(30,488) | .97 | 21,278 | 27,420 (11,038) | 0.94 |
| Verbal Paired Associates – Score | 1027 | 16.90 (6.65) | .92 | 8496 | 18.21 (5.63) | 0.89 |
| Visual Paired Associates – Score | 1730 | 14.44 (5.27) | .84 | 9280 | 15.97 (4.53) | 0.78 |
| Trail-Making Test Part B – Median RT (ms) | 2564 | 1407.88 (1235.40) | .76 | 21,343 | 1065.43 (410.12) | 0.73 |
| Trail-Making Test Part B – Total Time (ms) | 2564 | 58,163 (65,371) | .94 | 21,343 | 41,776 (16,774) | 0.93 |
| Matrix Reasoning – Score | 2058 | 24.22 (7.59) | .94 | 3674 | 26.43 (5.87) | 0.91 |
a
The design of the Forward and Backward Digit Span tests did not allow for computation of split-half reliability. In an alternate version of these tests where separate participants completed two interleaved, independently scored digit spans, reliability was .73 for the Forward Digit Span Test, and .68 for the Backward Digit Span Test (Chaytor et al., 2021).
b
Differing sample sizes because two individuals did not produce any correct trials, so median correct reaction time calculation would be infeasible.