Mental Operations on Long-term Memory do not Require a Sustained Increase in Working Memory Engagement

The human mind is often described as having two distinct storage systems: a vast library of past experiences and a small, active workspace for immediate thoughts. The library, known as long-term memory, holds the facts, skills, and associations we have accumulated over a lifetime. The workspace, called working memory, is the mental scratchpad we use to hold a few pieces of information in our mind right now, such as a phone number we are about to dial or a route we are navigating. For decades, scientists have believed that when we perform mental tasks, like calculating a midpoint or solving a puzzle, our brain must keep the relevant information locked tightly in this active workspace. The prevailing view was that thinking harder requires holding more things in this workspace for longer periods, essentially demanding a sustained increase in mental effort.

This assumption, however, relied heavily on experiments where people had to remember brand-new, unfamiliar information. In those cases, the brain had to work hard to keep the new data from fading away. But much of our daily thinking involves manipulating information we already know well, such as recalling the location of a familiar object or using a known rule to solve a problem. It remained unclear whether the brain treats these familiar tasks the same way as the novel ones. If we are operating on information retrieved from our vast internal library, does our active workspace still need to stay fully engaged the entire time, or does it step in only when absolutely necessary?

A team of researchers set out to answer this question by observing the brain's electrical activity while people performed mental operations on familiar information. They began by teaching participants a simple set of associations: specific colors were linked to specific locations on a screen. Once these connections were learned and stored in long-term memory, the participants entered the testing phase. On each trial, a color would appear, prompting the person to retrieve the location they had learned. Immediately after, a new, different location appeared on the screen. Across different blocks of trials, participants either performed a mental operation—needing to imagine the exact spot halfway between the remembered location and the new one—or they simply had to judge whether a final probe matched one of the locations they had seen.

To see what was happening inside the brain during these moments, the researchers used a technique that measures electrical signals on the scalp. They focused on a specific pattern of brain waves known as alpha-band power, which acts as a window into how the brain manages information. By analyzing these signals with advanced computer methods, they could track two things: which task the person was doing and where the person was mentally focusing their attention. The results showed that the brain's representation of the task itself—whether the person was calculating a midpoint or making a comparison—remained steady and clear throughout the entire trial. This indicated that the person knew what they were supposed to do and kept that goal in mind.

However, the picture changed when the researchers looked at how the brain held the actual location information. When participants were asked to calculate the midpoint, the brain showed a stronger signal for the remembered location early in the process, suggesting a brief, intense use of the active workspace to perform the calculation. But this heightened engagement did not last. As the trial progressed, the difference in brain activity between the calculation task and the simple comparison task faded away. The brain did not maintain a high level of active storage for the duration of the task. Instead, it relied on the information stored in long-term memory, bringing the active workspace online only for a short burst to handle the specific operation.

These findings challenge the long-held idea that mental operations inherently require a continuous, heavy load on our active working memory. The study suggests that when information is already available in our long-term memory, our brain does not need to keep it under constant, intense scrutiny. Instead, the active workspace acts more like a temporary tool, stepping in for a moment to perform a specific operation before stepping back, allowing the vast library of long-term memory to do the heavy lifting. This reveals a more efficient and flexible way the mind handles the complex calculations of everyday life.