Direct access to working memory contents.

Area of Science:

• Cognitive Psychology
• Neuroscience
• Human Memory

Background:

• Working memory (WM) involves maintaining and updating information.
• The process of updating information in WM, especially when it involves accessing stored data, is not fully understood.
• Previous research has explored the efficiency of information retrieval and updating within WM.

Purpose of the Study:

• To investigate whether accessing information from working memory (WM) influences the time costs associated with updating that information.
• To determine if the retrieval of WM data is a bottleneck in updating processes.
• To examine the automaticity of WM access during cognitive tasks.

Main Methods:

• Two experiments were conducted using a working memory task involving three digits in different colors.
• Participants updated digits using arithmetic equations, with conditions varying whether WM data was needed for the equation (memory-access) or not (no-access).
• Object-switch costs (time difference for updating a non-updated digit vs. an updated digit) and overall equation completion times were measured.

Main Results:

• Object-switch costs were equivalent in both memory-access and no-access conditions, indicating that WM retrieval time does not contribute to these costs.
• Equation completion times were similar across access and no-access conditions, suggesting WM access is rapid.
• Mismatches in the no-access condition slowed performance, demonstrating automatic WM access even when unnecessary.

Conclusions:

• Working memory contents and their associated contexts are bound together and accessed as a composite unit.
• Information in working memory is automatically accessed, even when not explicitly required for a task.
• The findings challenge models where WM updating is solely dependent on retrieval speed from a buffer.