A State-Transition-Free Delayed-Feedback Task Elicits Heterogeneous Human Responses.

Area of Science:

• Cognitive Psychology
• Behavioral Neuroscience
• Learning Sciences

Background:

• Action-outcome learning is fundamental for survival and adaptation.
• Previous research has focused on immediate rewards, neglecting delayed outcomes.
• Understanding delayed reward learning is crucial for explaining complex decision-making.

Purpose of the Study:

• To investigate human ability to associate actions with delayed outcomes following intervening choices.
• To explore individual differences in learning strategies and understanding of delayed reward contingencies.
• To build upon previous findings on action-outcome association.

Main Methods:

• A novel learning task with six visual stimuli across three trial types was designed.
• Participants made choices in A vs B (immediate reward), C vs D (delayed reward), and E vs F (cue) trials.
• Performance and verbal reports were analyzed to assess learning and understanding.

Main Results:

• Nine participants successfully learned to choose the action leading to a delayed reward.
• One participant verbally articulated the task structure; eight showed misunderstandings.
• Significant individual differences were observed in action selection biases and explicit/implicit understanding.

Conclusions:

• Humans possess the capacity to learn action-outcome associations with significant delays and intervening choices.
• Cognitive linking of actions to delayed outcomes is complex and shows considerable inter-individual variability.
• Further research is needed to understand the mechanisms underlying delayed reward learning and its subjective representation.