Finding

walsh-2024-prior-cue.psychometric.cue-valid

Walsh K, McGovern DP, Dully J, Kelly SP, O'Connell RG. Prior probability cues bias sensory encoding with increasing task exposure. eLife 12:RP91135, 2024. · Human visual contrast 2AFC keyboard task · psychometric

Observed curve and fits

Joint-fit companions

Other findings jointly fit by the same model fits.

Fit diagnostics

Caveats on these fits (1 term)

Each tag below is attached to one or more candidate fits for this finding. The chip column in the table flags which row carries which tag.

aggregate DDM RT approximation
DDM likelihood uses aggregate chronometric constraints rather than a full trial-level response-time likelihood.
Variant AIC Δ AIC RMSE Caveats
logistic-4param winner 31831.8 0.0 0.0646 —
ddm-starting-point-bias 34872.2 3040.3 0.1076
  • aggregate DDM RT approximation
ddm-drift-bias 34956.2 3124.3 0.1137
  • aggregate DDM RT approximation
sdt-2afc 35466.1 3634.3 0.1180 —
ddm-vanilla 37864.9 6033.1 0.1384
  • aggregate DDM RT approximation
Full diagnostics (logL · n · params · max abs error)
Variant logL n Free params Max |error| Predicted points
logistic-4param -15911.9 44,255 4 0.1933 38
ddm-starting-point-bias -17432.1 44,255 4 0.2358 38
ddm-drift-bias -17474.1 44,255 4 0.2616 38
sdt-2afc -17731.0 44,255 2 0.2637 38
ddm-vanilla -18929.5 44,255 3 0.3186 38

Residuals

Fit parameters and provenance

logistic-4param.walsh-2024-prior-cue.psychometric.cue-valid

dirty fit

bias=-1.713 · lower_lapse=0.221 · slope=1.179 · upper_lapse=0.049

Method
scipy.optimize.minimize
Commit
24fd416
Predicted points
38
BIC
31866.6

Fitted with scipy.optimize.minimize via behavtaskatlas.model_fits.logistic-4param; success=True; CONVERGENCE: RELATIVE REDUCTION OF F <= FACTR*EPSMCH

ddm-starting-point-bias.walsh-2024-prior-cue.psychometric.cue-valid

dirty fit

boundary=0.763 · drift_per_unit_evidence=0.204 · non_decision_time=0.680 · starting_point=0.651

Method
scipy.optimize.minimize
Commit
10e1c18
Predicted points
38
BIC
34906.9

Fitted with scipy.optimize.minimize via behavtaskatlas.model_fits.ddm-starting-point-bias; success=True; CONVERGENCE: RELATIVE REDUCTION OF F <= FACTR*EPSMCH

ddm-drift-bias.walsh-2024-prior-cue.psychometric.cue-valid

dirty fit

boundary=0.896 · drift_bias=0.834 · drift_per_unit_evidence=0.169 · non_decision_time=0.680

Method
scipy.optimize.minimize
Commit
10e1c18
Predicted points
38
BIC
34990.9

Fitted with scipy.optimize.minimize via behavtaskatlas.model_fits.ddm-drift-bias; success=True; CONVERGENCE: RELATIVE REDUCTION OF F <= FACTR*EPSMCH

sdt-2afc.walsh-2024-prior-cue.psychometric.cue-valid

dirty fit

criterion=-0.406 · d_prime=0.084

Method
scipy.optimize.minimize
Commit
24fd416
Predicted points
38
BIC
35483.5

Fitted with scipy.optimize.minimize via behavtaskatlas.model_fits.sdt-2afc; success=True; CONVERGENCE: RELATIVE REDUCTION OF F <= FACTR*EPSMCH

ddm-vanilla.walsh-2024-prior-cue.psychometric.cue-valid

dirty fit

boundary=0.747 · drift_per_unit_evidence=0.186 · non_decision_time=0.680

Method
scipy.optimize.minimize
Commit
10e1c18
Predicted points
38
BIC
37891.0

Fitted with scipy.optimize.minimize via behavtaskatlas.model_fits.ddm-vanilla; success=True; CONVERGENCE: RELATIVE REDUCTION OF F <= FACTR*EPSMCH

Observed points

Signed target-distractor contrast p_right n
-28.000 0.1579 266
-26.000 0.1351 259
-25.000 0.0281 249
-22.000 0.2148 1,038
-20.000 0.1480 757
-18.000 0.2185 769
-17.000 0.1648 540
-16.000 0.3633 933
-15.000 0.1677 1,324
-14.000 0.2579 3,207
-13.000 0.1458 391
-12.000 0.1994 2,452
-11.000 0.1709 275
-10.000 0.2619 3,467
-8.000 0.2571 1,330
-7.000 0.2038 2,076
-6.000 0.1979 1,142
-4.000 0.4077 390
-3.000 0.3756 402
3.000 0.9215 484
4.000 0.9389 442
6.000 0.9677 1,209
7.000 0.9786 2,198
8.000 0.9261 1,434
10.000 0.9831 4,087
11.000 0.9659 293
12.000 0.9719 2,737
13.000 0.9610 410
14.000 0.9302 3,282
15.000 0.9157 1,399
16.000 0.9812 959
17.000 0.9634 574
18.000 0.9497 776
20.000 0.9137 765
22.000 0.9158 1,152
25.000 0.9524 252
26.000 0.7930 256
28.000 0.8136 279

Take it with you

Cover sheet

Self-contained Markdown for citation, slides, or notebooks

The cover sheet pins the finding to the atlas commit and includes the observed points, fit ranking, caveats, and provenance — everything needed to drop into a paper or notebook without losing the trail back to the deploy.