Variations on variability: effects of display composition on same-different discrimination in pigeons

Area of Science:

• * Cognitive psychology
• * Animal behavior
• * Comparative cognition

Background:

• * Discriminating between same and different multi-item arrays is a fundamental cognitive task.
• * This discrimination can be framed as distinguishing between low and high variability arrays.
• * Previous research proposed an entropy account suggesting the method of variability manipulation should not affect performance.

Purpose of the Study:

• * To critically test the predictions of an entropy account in same-different array discrimination.
• * To investigate how different methods of manipulating array variability affect discrimination performance in pigeons.
• * To determine if equivalent variability always yields equivalent performance.

Main Methods:

• * Pigeons were trained on extreme variability conditions (16 identical vs. 16 nonidentical items).
• * Testing involved intermediate variability using two methods: mixture testing (varying item combinations) and number testing (varying item counts).
• * Performance on these intermediate arrays was compared to assess the impact of variability manipulation.

Main Results:

• * The entropy account explained a significant portion of the observed discrimination data.
• * However, the results indicated that the method of changing array variability did influence performance.
• * Pigeons' performance varied depending on whether mixture testing or number testing was employed.

Conclusions:

• * While entropy is a key factor, it does not fully account for same-different array discrimination.
• * The specific manipulation of variability (e.g., mixture vs. number changes) impacts cognitive performance.
• * Further research is needed to refine models of discrimination that incorporate these nuanced effects.