Do complex span and content-embedded working memory tasks predict unique variance in inductive reasoning?

Area of Science:

• Cognitive Psychology
• Neuroscience
• Working Memory Research

Background:

• Working memory is crucial for cognitive tasks, involving both maintenance and processing.
• Complex span and content-embedded tasks are used to measure working memory capacity.
• A key difference lies in task relevance: complex span tasks maintain irrelevant information, while content-embedded tasks maintain relevant information.

Purpose of the Study:

• To test if content-embedded tasks explain more unique variance in inductive reasoning than complex span tasks.
• To investigate the role of maintaining task-relevant information in inductive reasoning.
• To extend research on predicting higher-level cognition using different working memory measures.

Main Methods:

• Administered three complex span tasks, three content-embedded tasks, and three inductive reasoning tasks to 384 participants.
• Utilized structural equation modeling to analyze the predictive power of each task type on inductive reasoning.
• Assessed the unique and shared variance explained by content-embedded and complex span tasks.

Main Results:

• The structural equation model explained 51% of the variance in inductive reasoning.
• Content-embedded tasks uniquely predicted 45% of the variance in inductive reasoning.
• Complex span tasks uniquely predicted less than 1% of the variance, with 6% shared variance.

Conclusions:

• Content-embedded tasks significantly outperform complex span tasks in predicting inductive reasoning.
• The findings support the hypothesis that maintaining task-relevant information is key for higher-level cognitive abilities.
• This research highlights the advantage of content-embedded tasks for assessing cognitive functions related to reasoning.