Table 6 o1-Preview performance on the algorithmic thinking test for adults (ATTA).
From: Comparative evaluation of OpenAI O1 and human performance in higher order cognition
Participant category | Human overall | o1-Preview | Z-score |
|---|---|---|---|
Experts | 14.63 ± 3.81 | 20.00 ± 0.00 | 1.41 |
Novices | 9.11 ± 3.81 | 20.00 ± 0.00 | 2.86 |
Social Sciences | 9.11 ± 4.54 | 20.00 ± 0.00 | 2.40 |
Mathematics | 15.70 ± 3.71 | 20.00 ± 0.00 | 1.16 |
Physics | 15.26 ± 3.92 | 20.00 ± 0.00 | 1.21 |
Engineering | 14.37 ± 3.17 | 20.00 ± 0.00 | 1.78 |
Computer Science | 14.00 ± 3.80 | 20.00 ± 0.00 | 1.58 |
