Professor Explores Relationship Between Working Memory and Intelligence
Curators’ Distinguished Professor of Psychological Sciences Nelson Cowan has been studying different kinds of childhood language problems with colleagues from Arizona State University. Their grant focuses on the use of working memory as an important part of language use and may provide insights into differences in intelligence. Cowan is a co-author of the research that will be published by Professor Shelley Gray et al. in 2017 in the Journal of Memory and Language and is now available electronically.
According to Cowan, working memory is the small amount of information you can hold in your mind at one time. For example, if you are listening to language, you have to hold in your mind the beginning of the sentence or paragraph until it continues to the point where you can make sense of it. Cowan says people with language problems such as dyslexia may have a working memory deficit or problem.
“Our approach was to, first of all, understand the structure of working memory, and second, applying that to see whether the structure is different in some of these children with different kinds of language-processing problems,” Cowan says. To understand the structure of working memory, Cowan compared and contrasted different models—one that dates to 1974 by Alan Baddeley and Graham Hitch, a revised version by Baddeley published in 2000, and Cowan’s model, first published in 1988. Cowan says the Baddeley models contend the brain stores information in specific modules—one module stores information about speech sounds and another stores information presented visually and spatially.
“In my conception, those modules are replaced with what’s called the ‘activated part of long-term memory,’ so long-term memory holds everything we know, but only certain parts of it are activated at the moment, and that activated information is not necessarily in our awareness, but it’s readily available,” Cowan says. “The other thing that’s different about my model is that within that activated part of long-term memory, there is a subset of it that is said to be the ‘focus of attention,’ and that is limited to just a few, well-thought-out items.” Cowan says researchers used to believe that a person could hold up to seven items (words, phrases, numbers, etc.) in their focus of attention, but his research shows that number actually to be three to four.
Cowan says 168 English-speaking, typically developing 7–9 year olds from three states participated in a battery of tests to determine which model best assessed the relationship between working memory and intelligence. Cowan says his “embedded processes model” fit the data slightly better than the other two models tested.
“Once we got those models, we tried to use them to see how well they could predict intelligence, because we had done so many tests with the kids that we had a lot of good information,” Cowan says. “So we could then ask which components (of the models) correlate with intelligence—one was fluid intelligence, which is the ability to do new tasks you have never seen before, and the other is a measure of visual-type intelligence. Both of them correlated most strongly with the focus of attention, so it’s the ability to keep information in the focus of attention that helped determine intelligence-test scores.”
Going forward, Cowan says he and his colleagues will study what they can do to help improve the focus of attention in children. “Since these are children we are talking about, one could improve education materials by being aware there is a certain complexity that can overwhelm someone’s focus of attention, and then simplify the materials so they don’t overwhelm the person.” For example, he says a teacher could break a work problem into parts and have the student tackle each part individually before gradually linking them all together.