DD-011 v2
Selective-attention experiment extension
Nine synthetic treatments separate public-clue audience, precision, recommendation, license, reward, and conditional policy. Six appended hypotheses define their contrasts and estimands.
Scroll horizontally to see every column.
| Cell | Public access | Precision | Institution | Policy recommendation | Reward |
|---|---|---|---|---|---|
| T20-public-all | all | full | none | none | equal_split |
| T21-random-reader | one_random | full | none | none | equal_split |
| T22-selected-audience | selected_audience | full | none | none | equal_split |
| T23-reduced-precision | all | reduced | none | none | equal_split |
| T24-designated-reader | designated | full | none | none | equal_split |
| T25-ignore-recommendation | all | full | nonbinding_ignore | none | equal_split |
| T26-attention-license | all | full | license | none | equal_split |
| T27-off-signal-rescue | all | full | none | conditional | sole_rescue |
| T28-conditional-visible | all | full | none | contrarian_visible | equal_split |
Scroll horizontally to see every column.
| ID | Question | Outcome | Multiplicity family |
|---|---|---|---|
| H9 | Does broadcast induce more public-signal use than one-reader access? | public_signal_use | attention |
| H10 | Does single-reader access improve discovery relative to full broadcast? | discovery | attention |
| H11 | Does an attention license outperform a nonbinding ignore recommendation? | one_reader_alignment | implementation |
| H12 | Does a designated reader implement one-reader allocation over broadcast? | one_reader_alignment | implementation |
| H13 | Do conditional recommendations improve discovery over broadcast? | discovery | attention |
| H14 | Does action visibility induce the registered contrarian response? | contrarian_action | attention |