When high working memory capacity is and is not beneficial for predicting nonlinear processes.

Area of Science:

• Cognitive Psychology
• Decision Science

Background:

• Predicting dynamic processes is crucial across disciplines.
• Existing research on working memory capacity (WMC) and prediction strategy accuracy is inconclusive.
• It remains unclear if higher WMC always leads to better prediction strategies or sometimes to overly complex ones.

Purpose of the Study:

• To investigate the relationship between working memory capacity (WMC) and prediction strategy selection and accuracy.
• To determine if WMC influences the use of well-calibrated versus simpler strategies in predicting nonlinear dynamic processes.

Main Methods:

• Participants predicted systematically varied nonlinear dynamic processes based on exponential functions.
• The study manipulated whether prediction accuracy could be enhanced by well-calibrated rules.
• Working memory capacity (WMC) was assessed, and prediction strategies were analyzed.

Main Results:

• Higher WMC participants utilized more accurate, well-calibrated strategies, especially for highly nonlinear process trajectories.
• Lower WMC participants favored simpler, exemplar-based prediction strategies.
• Simpler strategies were as effective as complex ones for approximately linear prediction regions.

Conclusions:

• Working memory capacity (WMC) is not universally a strength or weakness in predicting dynamic processes.
• The effectiveness of prediction strategies, and the role of WMC, is contingent on the nonlinearity of the process being predicted.