Table 1. Mean posterior parameter fit for the DDM-equivalent model and exact input model over the four difficulty levels for 24 participants.
| Parameter | DDM-Equivalent Model |
Exact Input Model |
||||||
|---|---|---|---|---|---|---|---|---|
| D1 (hard) | D2 | D3 | D4 (easy) | D1 (hard) | D2 | D3 | D4 (easy) | |
| Noise SD | 236.70 (22.16) | 76.90 (5.78) | 70.73 (3.94) | 51.29 (4.99) | 95.63** (7.31) | 58.27* (2.93) | 52.10** (2.82) | 38.22* (2.57) |
| NDT mode | 563.26 (52.72) | 453.31 (43.54) | 431.24 (40.67) | 430.51 (36.10) | 520.73** (47.29) | 415.96** (37.83) | 382.16** (30.47) | 399.35* (33.16) |
| NDT SD | 195.17 (20.38) | 121.61 (13.13) | 137.60 (16.62) | 164.97 (17.95) | 201.66 (16.42) | 181.27** (12.06) | 172.54* (12.17) | 167.54 (10.00) |
| Bound | 0.71 (0.01) | 0.74 (0.01) | 0.83 (0.02) | 0.83 (0.02) | 0.65** (0.01) | 0.77 (0.02) | 0.87 (0.02) | 0.88 (0.02) |
| Prior | 0.46 (0.01) | 0.46 (0.01) | 0.46 (0.01) | 0.43 (0.02) | 0.47* (0.01) | 0.47 (0.01) | 0.44 (0.01) | 0.42 (0.02) |
| Lapse prob. | 0.03 (0.01) | 0.03 (0.00) | 0.02 (0.00) | 0.03 (0.00) | 0.02 (0.00) | 0.02 (0.00) | 0.02 (0.00) | 0.02* (0.00) |
| Lapse (TO) prob. | 0.24 (0.02) | 0.16 (0.02) | 0.18 (0.02) | 0.14 (0.01) | 0.24 (0.02) | 0.17 (0.02) | 0.18 (0.02) | 0.15 (0.01) |
TO = Timeout, NDT = Non-decision time, SD = Standard deviation. NDT mode and NDT SDs are reported in milliseconds. Bound and prior are both probabilities (range: 0 to 1). For the prior, 0 means complete right bias and 1 complete left bias. Values in parentheses are SE over participants. The asterisks indicate differences between DDM-equivalent model and exact input model for specific conditions (*p < 0.05, **p < 0.01, paired t-test over 24 subjects). Most notably, both the noise SD and NDT mode were higher for the DDM-equivalent model, for each condition. As expected, with increasing difficulty level the bound for making decisions decreases. The prior (i.e. a bias for reporting one of the two alternatives) indicates that participants slightly favour the right hand target. This is in line with the observed data; participants made on average approximately 5% more right decisions (data not shown).