Working Memory: The Brain's Spam Filter Located
Andrew W. McCollough & Edward K. Vogel
University of Oregon
What makes you so smart? Might be your lizard brain.
Anyone who has tried to find an urgent email amid masses of advertisements for dubious stock opportunities and sexual enhancement drugs understands the critical importance of being able to filter out distracting information. That important email may be in there but it is lost among irrelevant clutter. And while the capacity of our email inbox is limited only by disc space, our mental 'inbox' of working memory
is much more constrained. In fact, several decades of research have indicated that our capacity to hold information "in mind" for immediate use is limited to a mere three or four items
. Moreover, just as people vary in height and eye color, we also vary in the capacity of this memory inbox. Interestingly, these differences in working memory capacity are strongly predictive of a person's ability to perform abstract reasoning, mathematics, and other forms of complex problem solving. This relationship between memory capacity and fluid intelligence has motivated many scientists to try to understand why and how people differ in this important cognitive ability.
Hard drive or spam filter?
There are at least two primary explanations for this severe limitation in working memory capacity. First, it could be that working memory capacity is essentially determined by storage space, and that some people have larger "hard drives" than others do. The alternative explanation is that capacity depends not on the amount of storage space but on how efficiently that space is used. Thus high-capacity individuals might simply be better at keeping irrelevant information out of mind, whereas low capacity individuals may allow more irrelevant information to clutter up the mental inbox. High-capacity individuals may just have better spam filters.
Some of our recent work on differences in controlling access to working memory (abstract
or pdf download
has provided evidence favoring this mental spam filtering idea. In one experiment, measuring electrical signals emitted by the brain enabled us to show that high-capacity people were excellent at controlling what information was represented in working memory: They let in information about relevant objects but completely filtered out irrelevant objects. Low-capacity individuals, by contrast, had much weaker control over what information entered the mental "in box"; they let in information about both relevant and irrelevant objects roughly equally. Surprisingly, these results mean that we found that low capacity people were actually holding more total information in mind than high capacity individuals were -- but much of the information they held was irrelevant to the task.
Where can I get one of those filters?
So the evidence is amassing that your mental spam filter's effectiveness largely determines your working memory capacity. Yet a critical question remains unanswered. Where in the brain does this spam filter reside?
In a recent study published last month in Nature Neuroscience
, Fiona McNab and Torkel Klingberg
at the Stockholm Brain Institute appear to have found the spot. To do so, they had study participants perform a working memory task in which they had to remember the positions of some red and yellow squares on a computer screen. Sometimes they were asked to remember all the items on the screen (red and yellow), and other times they were asked to remember just the red items and to keep the yellow items out of mind -- an act akin to filtering spam. A symbol at the start of each trial told them whether they had to focus on just red squares or let all of the information from the display flow into memory. The researchers recorded the subjects' brain activation during this instruction period as a way of determining what portions of the brain became active as a person started up the "spam filter".
They found that when participants were told that they would need to filter the upcoming trial, portions of the basal ganglia
(an area known to be important in movement, among other things) and prefrontal cortex
(considered the brain's rationalizing, "thinking" part) became much more active than in the nonfiltering trials. Importantly, McNab and Klingberg found that the jump in activity levels in these areas was largest for high capacity individuals and smallest for low capacity individuals. That is, when told that they needed to filter, the high-capacity individuals ramped up activity in these brain regions to keep out irrelevant items. By contrast, the low-capacity individuals showed little additional activity in these areas when they were instructed to ignore the irrelevant items. Thus, a leading candidate for the "mental spam filter" appears to be a cooperative effort between the basal ganglia and the prefrontal cortex.
Higher cognition that lizards can relate to.
This new work ties together several converging lines of evidence indicating that your working memory capacity is closely related to how well you can keep irrelevant information out of your mental inbox. In particular, it suggests that this filtering mechanism is determined by coordinated activity in the basal ganglia and prefrontal cortex -- with the prefrontal cortex providing details about the current task goals and the basal ganglia providing the muscle to block out information that doesn't match these goals.
This role for the basal ganglia in helping to control the flow of information into working memory is quite similar to one of the basal ganglia's other major functions, which is selecting which motor movements to use in a given context and suppressing the movements that we don't want. Particularly intriguing is that the basal ganglia is an evolutionarily ancient brain structure that has been highly conserved across species; even lizards have them. Consequently, what is thought to be our uniquely human ability to engage in abstract reasoning and problem solving appears to be dependent upon brain structures that have been around for far longer than humans have. The ability to filter out irrelevant spam, it appears, is critical for lizards as well as humans.
is a graduate student and Edward Vogel
the principal investigator at the Visual Working Memory & Attention Lab
at the University of Oregon, in Eugene, where McCollough also teaches tango
Let's Not Forget the Memory Size
University of Missouri
According to both the McNab and Klingberg study of working memory and McCollough and Vogel's review of it, the efficiency of mental "spam filters" in our basal ganglia largely determines the capabilities of working memory. Inefficient spam filters permit needless, excess activity in the brain regions that actually store working memory information -- the posterior parietal areas, which are along the top of the brain toward the back. In working memory operations, these parietal areas were said to play the role of the computer's hard drive.
According to both pairs of researchers, the efficiency of the basal ganglia's "spam filters" determines the capabilities of working memory, which vary considerably among individuals. As I note above, these spam filters help determine what information reaches the posterior parietal areas that actually store working memory information. But these parietal areas function here not so much as the permanent storage of a hard drive, but rather as the temporary storage of random access memory
, or RAM, where information is held when it is in use or might soon be used. And although there is valid evidence for the importance of the efficiency of filtering out irrelevant items from working memory -- from the McNab-Klingberg work and other research McCollough and Vogel discussed -- we must be careful not to overlook the possibility that differences in RAM capacity also affect working memory. If RAM size does matter, then RAM size and filtering efficiency may be imperfectly correlated. By analogy, individuals' top sprinting speeds and endurances may be imperfectly correlated, even though both qualities depend on certain common factors like health.
In fact there is evidence that the storage capacity of working memory is important. In 2005, Todd and Marois
showed that brain activity in the posterior parietal areas -- the working memory "RAM" -- correlated with working memory performance. Their test materials did not include irrelevant visual items, so filtering was not difficult.
Behavioral tests in other studies reinforce their conclusion. In a 2006 study of normal and schizophrenic adults
, Gold and colleagues occasionally tested memory for items that subjects were told they could ignore. Compared to normal control subjects, schizophrenic patients remembered fewer items across the board -- that is, they remembered fewer of the items they were told to remember and fewer of those they were told they could ignore. Yet both controls and schizophrenics did far better remembering "attended" items than items they were allowed to ignore. The filtering efficiency, in other words, was about equal in the two groups. Meanwhile, a 2006 study in my own lab
found that the storage capacity and filtering efficiency of working memory were partly related and partly distinct -- something like the sprinting-to-endurance relationship suggested above. Not every relatively high-capacity participant was able to filter out irrelevant items very efficiently, for example.
The panorama of results demonstrates that both storage capacity and filtering efficiency affect an individual's working memory ability. New methods for analyzing the brain do not necessarily replace older ones -- just as automobiles did not replace bicycles, which did not replace walking. To find the whole truth, brain imaging methods must be used along with older, behavioral methods and philosophical reasoning about the mind. Back in 1971, in an essay titled "Art in Bits and Chunks," the perceptual psychologist Rudolf Arnheim suggested that a psychologist's most important tool is the armchair. The statement still rings true for brain research.
Nelson Cowan is director of the Working-Memory Laboratory at the University of Missouri, Columbia, and is the author of Working Memory Capacity (Psychology Press, 2005).