Where Should You Place Your Bet? On Developing Working Memory or Many Mental Mechanisms?

Working memory has long been considered by cognitive psychologists to be the most important cognitive predictor of fluid intelligence. For the longest time, working memory was thought to be a fixed capacity. However, about a decade ago, following the publication of a paper by Jaeggi et al (2008), interest started to mount in the possibility that working memory could be enhanced. No less an intelligence authority than Richard Sternberg commented most favorably to the latter report, making several very audacious claims to the effect that working memory is significantly trainable and has transferable effects on fluid intelligence tasks! Meanwhile neurosplacticity was becoming a hot topic of research and popular discussion and hope. The holy grail was in sight.

The burgeoning cognitive training (“brain training”) software industry, whose business plans were designed to capitalize on the tremendous wealth and fears of aging baby boomers, quickly capitalized on this possibility. Huge numbers of consumers signed on, paid money, and devoted significant portions of their extremely precious time to playing games that promised to make them smarter.

Yet single critical experiments—in any discipline — do not simply overthrow decades of science. (Compare Lakatos, 1980, for instance.) The difficulty of achieving far transfer is well established. All indications had been that working memory is not very malleable. However, experiments can quickly redirect researchers attention, and that they did! A large number of studies and some very heated debates have ensued. Clearly, the Jaeggi et al experiment, and Sternberg’s overenthusiastic commentary led to scientific progress.

Meanwhile, the US Federal Trade Commission has reigned in a cognitive training company, Lumosity, for deceptive advertising[1]. And SharpBrains has been promoting the industry while trying to keep it honest.

I could see the controversy brewing in the early days. I concluded that although R&D in working memory was warranted, prescribing working memory training was premature. I felt (and still believe) that potentiating long-term memory is a surer bet. There was already an enormous body of literature, some of which I reviewed in chapter 7 of Cognitive Productivity, that showed that testing long-term memory significantly improves one’s ability to access and use it. This literature goes under the banners of “test effects”, “distributed recall practice” and “test-enhanced learning”. Meanwhile, the expertise literature had established that experts (in a narrow technical sense of the term), are remarkably quicker at encoding and accessing long-term memory content. The concept of long-term working memory was established.

And yet, this literature is hardly ever cited in the context of “cognitive training”. Cognitive training folk generally hope that a general purpose mechanism like working memory can be improved, with significant transfer effects. Well, for one thing, working memory is not a single construct. It’s a complex theoretical construct with multiple properties, including its interfaces with perception and long-term memory.

In Cognitive Productivity I acknowledged that working memory processes are key bottlenecks and enablers of fluid intelligence and hence problem solving. (Compare for instance Cornoldi & Giofrè 2014.) However, I also cited some research suggesting that the contribution of working memory capacity to the criterion, expert performance, has important bounds. Ackerman, Beier & Boyle (2005) argued that variance in working memory does not fully account for measures of fluid intelligence. Moreover, Mogle, Lovett, Stawski, & Sliwinski (2008) concluded that working memory capacity correlated .42 with fluid intelligence measures, whereas encoding and access to secondary memory (long term memory) correlated at .58. Processing speed only correlated .07. Around the same time, Unsworth & Engle published a pair of experiments that were also suggestive of the importance of access to secondary memory, particularly with cues.

Merlin Donald (2001) also provided strong grounds for cognitive science to balance its obsession with working memory with an equal emphasis on long-term working memory (LTWM) and intermediate governance, a form of consciousness, even outside the domain of expertise in which LTWM was first proposed.

In this general context, I developed a framework based on a large body of literature on expertise, educational psychology and cognitive science more generally. In Cognitive Productivity I wrote:

It is therefore central to the aims of this book to understand how to develop long-term working memory from knowledge resources. This is particularly pertinent firstly because the role of memory and knowledge are often downplayed in discussions of education. There are those who believe that expertise is the result of superior intelligence and experience. Others believe it is a matter of being “highly skilled”. Secondly, even amongst those who know that knowledge is critical to expertise, not everyone appreciates the importance of long-term working memory and the activities and mental processes that develop it.

There are many ways in which we can use software to develop our abilities to access and apply specific information in long-term memory in a manner that is coherent with a large body of literature. One of my key points in Cognitive Productivity is that these “ways” had better be both efficient and effective. Time is precious. Many people already find (or ought to realize) they spend too much time with information technology. And yet, there is no silver bullet for cognitive performance. So, some effort is required.

Moreover, my “meta-effectiveness” framework is not purely cognitive. It also deals with many affective processes, such as valenced perception and motive generators.

For many people, I think the best approach is to look at the amount of time they spend working with information. It may be that they spend too much time taking in new information, and not enough time ensuring that they can systematically apply some of the most potentially helpful knowledge gems they encounter. It’s then a matter of adjusting the ratio of time spent surfing, delving and mastering knowledge.

Mastering information is not primarily about recalling information, it’s ultimately a matter of developing new motivators, habits and various mental mechanisms (or “internal mindware”)[2] that use knowledge.


Beaudoin, L. P. (2015), Cognitive Productivity: Using Knowledge to Become Profoundly Effective. BC: CogZest.

Clark, A. (2013, 2nd ed.). Mindware. New York, NY: Oxford University Press.

Cornoldi, C., & Giofrè, D. (2014). The crucial role of working memory in intellectual functioning. European Psychologist, 19(4), 260–268. http://doi.org/10.1027/1016-9040/a000183

Donald, M. (2001). A mind so rare: The evolution of human consciousness. New York, NY: W. W. Norton & Company.

Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. Proceedings of the National Academy of Sciences, 105(19), 6829. http://www.ncbi.nlm.nih.gov/pubmed/18443283

Lakatos, I. (1980). The methodology of scientific research programmes: Philosophical papers. (Vol. 1). Cambridge, UK: Cambridge University Press.

Mogle, J. A., Lovett, B. J., Stawski, R. S., & Sliwinski, M. J. (2008). What’s so special about working memory? An examination of the relationships among working memory, secondary memory, and fluid intelligence. Psychological Science, 19(11), 1071–1077. doi:10.1111/j.1467-9280.2008.02202.x

Perkins, D. N (1995). Outsmarting IQ: The emerging science of learnable intelligence. New York, NY: Free Press.

Sternberg, R. J. (2008). Increasing fluid intelligence is possible after all. Proceedings of the National Academy of Sciences. 105 (6791–6792) http://doi.org/10.1073/pnas.0803396105

Federal Trade Commission (2016). Lumosity to Pay $2 Million to Settle FTC Deceptive Advertising Charges for Its “Brain Training” Program. Retrieved from https://www.ftc.gov/news-events/press-releases/2016/01/lumosity-pay-2-million-settle-ftc-deceptive-advertising-charges

Unsworth, N., & Engle, R. W. (2005). Simple and complex memory spans and their relation to fluid abilities: Evidence from list-length effects. Journal of Memory and Language, 54(1), 68–80. doi:10.1016/j.jml.2005.06.003

Unsworth, N., & Engle, R. W. (2007). The nature of individual differences in working memory capacity: active maintenance in primary memory and controlled search from secondary memory. Psychological Review, 114(1), 104–132. doi:10.1037/0033-295X.114.1.104


[Lumosity]. Federal Trade Commission (2016). Lumosity to Pay $2 Million to Settle FTC Deceptive Advertising Charges for Its “Brain Training” Program.

Mindware. David Perkins introduced the term mindware to refer to the internalization of knowledge, rules, strategies etc. Perkins, Keith Stanovich and I interpret this term as referring to mental mechanisms (in the head). (Our views of personal development also, incidentally, focus on using knowledge as opposed to improving working memory). However Andy Clark published two editions of a book called Mindware that used the term in a very different way. I discuss these differences in Cognitive Productivity.

Published by

Luc P. Beaudoin

Head of CogZest. Author of Cognitive Productivity books. Co-founder of CogSci Apps Corp. Adjunct Professor of Education, Simon Fraser University. Why, Where, and What I Write. See About Me for more information.

3 thoughts on “Where Should You Place Your Bet? On Developing Working Memory or Many Mental Mechanisms?”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.