Associating everything with everything else, all at once: Semantic associations facilitate visual working memory...

Area of Science:

• Cognitive Psychology
• Neuroscience
• Artificial Intelligence

Background:

• Integrating prior semantic knowledge with environmental input is crucial for cognition.
• The impact of this integration on perception and memory, particularly visual working memory (VWM), is not fully understood.

Purpose of the Study:

• To investigate how associative semantic knowledge interacts with perceptual constraints in VWM.
• To determine if semantic associations aid VWM for real-world objects under brief encoding times.

Main Methods:

• Reanalyzed data from two studies (N=125, 25,630 trials) using a VWM task with objects and verbal distraction.
• Utilized natural language processing to estimate semantic associations and computer vision for image similarity.
• Manipulated encoding times (0.2 s vs. 1-2 s).

Main Results:

• Semantic associations predicted VWM accuracy significantly under brief encoding (0.2 s).
• This effect was not observed with prolonged encoding times (1-2 s).
• Image similarity did not influence VWM accuracy, suggesting a semantic, not visual, mechanism.

Conclusions:

• Associative semantic knowledge facilitates object grouping in VWM, mitigating perceptual constraints.
• Semantic associations play a significant role in VWM formation, especially under time pressure.
• Findings highlight the pervasive influence of semantic knowledge on cognitive processes like VWM.