Cognitive Inertia: Cyclical Interactions Between Attention and Memory Shape Learning.

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Machine Learning

Background:

• Current frameworks for selective attention often overlook real-world constraints.
• Computational models typically assume perfect feature encoding and accuracy maximization.
• Existing models inadequately explain attention allocation based on individual learning histories.

Purpose of the Study:

• To propose a comprehensive view of selective attention by integrating computational modeling with memory.
• To investigate the interplay between selective attention and memory in human learning.
• To understand how memory influences the goal-directed nature of attention.

Main Methods:

• Utilized computational modeling to simulate the interaction between attention and memory.
• Developed a framework where attention is constrained by memory of past experiences.
• Analyzed how memory influences the selective allocation of attention.

Main Results:

• Relevance for attention is established through memories of previous experiences.
• Attention is initially used for feature encoding and memory creation.
• Subsequent attention operates selectively based on stored memories, not just task demands.

Conclusions:

• Selective attention is fundamentally shaped by memory.
• Deviations from ideal attention allocation are goal-directed, constrained by individual memory.
• A memory-informed approach is crucial for understanding human attention and learning.