Steady state responding based upon simple and compound stimuli.

Researchers taught pigeons to peck one key when they saw a red vertical line and another key when they saw a green horizontal line.

The birds first learned with both color and tilt cues together. Later the team faded one cue so only color or only tilt stayed.

They tracked how fast the birds chose and how often they got it right.

The pigeons learned fastest when both cues were shown together. They almost never made mistakes.

When only one cue stayed, the birds still picked the right key. They had learned to watch both parts and then go with the cue that had paid off most often.

Bennett et al. (1973) saw the opposite: compound cues hurt matching accuracy. The clash is solved by the task. Catania (1973) used steady-state choice where both cues pointed to the same answer. S et al. used matching-to-sample where cues sometimes disagreed, so extra info created conflict.

Snapper et al. (1969) showed pigeons can learn color-form compounds, but form usually wins. Catania (1973) adds that redundancy, not just pairing, is key. When cues agree, control is stronger.

Thrailkill et al. (2025) later showed birds give more attention to parts that paid better before. This explains why C’s birds leaned on the cue with the richer history once the partner cue was removed.

Start teaching new conditional discriminations with redundant cues that all signal the same answer. Once the learner is accurate, fade the extra cues and keep only the target feature. This builds strong stimulus control fast and prevents errors later.