Learned changes in outcome associability.

Area of Science:

• Cognitive psychology
• Learning sciences
• Behavioral neuroscience

Background:

• Associative learning theories posit that cues change associability based on prediction.
• These theories typically do not consider an outcome's prior learning history as influential for new learning.
• This asymmetry suggests outcomes are treated as passive in the learning process.

Purpose of the Study:

• To investigate whether an outcome's learning history influences subsequent associative learning.
• To test the assumption that outcomes do not retain learning-related properties.
• To examine the impact of prior outcome predictability on new cue-outcome associations.

Main Methods:

• Two experiments utilized a serial letter-prediction task.
• Participants learned associations in two stages.
• Stage 1 involved predictable (X) and unpredictable (Z) outcomes; Stage 2 involved novel cues equally predicting X and Z.

Main Results:

• Learning was significantly faster for the previously predictable outcome (X) in Stage 2 compared to the previously unpredictable outcome (Z).
• This indicates that the outcome's prior learning history affected the speed of new associations.
• Participants demonstrated faster acquisition of new predictive relationships for outcomes with a history of predictability.

Conclusions:

• The results challenge the traditional associative learning view that outcomes are passive in learning.
• An outcome's prior learning history, specifically its predictability, influences subsequent associative learning rates.
• This suggests a more active role for outcomes in associative learning than previously theorized.