Kevin at IQ's Corner has blogged about a new paper in PNAS showing that "working memory" training can improve measures of fluid intelligence - a capacity long thought to be relatively insensitive to experience, and intricately tied to the most complex human cognitions like reasoning, planning, and abstraction in novel contexts.
Jaeggi et al., posit that no empirical evidence shows "computer games enhance anything beyond task-specific performance and selective visuospatial attention" (which must bother our friends at Lumosity & SharpBrains - sigh), and highlight the concern that by "practicing to the test" of fluid intelligence, one might reduce the novelty of those tests and thereby also reduce their power to measure fluid intelligence.
Nonetheless, Jaeggi et al endorse a view that "far transfer" - in which the benefits of training can be manifest in tasks which were not practiced or bear little apparent similarity to the trained tasks - is theoretically possible, and perhaps even probable in the case of the most complex cognitive functions which are "capacity limited." That is, limitations in attention may be the causal path through which working memory capacity and fluid intelligence are linked, since attention is critically required for the "temporary binding processes" which are used to associate particular items with a particular context. Special emphasis is given to the idea that this temporary binding is in fact "temporal binding": items must be associated with a particular moment in time for the optimal use of capacity-limited resources like attention.
To test this idea, Jaeggi et al trained subjects on a challenging working memory task in 4 separate experiments. The task involved the serial presentation of both visual stimuli (squares in one of eight spatial locations) and auditory stimuli (one of eight possible letters) at a rate of one item every 3s. Subjects were required to identify whether the current stimulus matched the stimulus presented n trials ago in either the visual or auditory modality - known as an "n-back" task. Critically, the value of n was adaptively determined by individual subject performance, such that if subjects made fewer than 3 mistakes for a given n on a block of 20+n trials, n was incremented by 1 for the following block; this culminated into 20 blocks of trials, requiring about 25 minutes of training per session. The experiments differed in the number of sessions provided (anywhere from 8 to 15) but performance was always compared against a control group, who took the same pre- and posttests assessing fluid intelligence (via Raven's Advanced Progressive Matrices or BOMAT [a harder version of Raven's]).
This training task involves a variety of demands on higher-level cognitive processes: "dual-tasking" in terms of divided attention between modalities, commonly thought to tax so-called "executive" processing; a reduction in task- or stimulus-specific processing due to the variability in n and the use of varied stimuli; the rapid binding of particular items to their temporal order; and the maintenance, manipulation, and monitoring requirements for remembering a list of items, iteratively updating that list with the presentation of new items, and the iterative comparison of new stimuli with those in memory, respectively.
The results showed that the trained group showed larger gains in fluid intelligence than the control group (partial eta squared of .07; even the control group showed some improvement, probably via practice effects). Using pre-test scores as a covariate (an important thing to do, since subjects may differ in their gains based on their pre-existing abilities), training had statistically significant effects after 17 days. In fact, subjects with lower pre-test fluid intelligence scores showed greater gains on the post-test (this was not specific to the trained group, but could reflect regression to the mean).
While the magnitude of training effects did not interact with pre-test measures of working memory (assessed via digit span and reading span), training did improve digit span scores. However, the training-induced enhancement of fluid intelligence (and it's relation to the number of sessions) was not fully explained by these gains, nor by the maximum n reached in training, as confirmed through analyses of covariance.
In summary, the authors advocate the view that transfer of training to fluid intelligence occurred through the enhancement of controlled attention, as opposed to the more specific cognitive processes involved in their dual n-back task.
One critical note about this and most other training studies showing far transfer: they tend to implement relatively massive manipulations of executive demands. It is not clear whether such breadth of executive demands is required for training to show transfer, or whether some particular aspect of these demands is sufficient for enhancing intelligence. This methodological shortcoming is most likely due to the large amount of time that needs to be invested in a training study; more targeted manipulations might work, but might show a null effect (publish or perish!). I'll continue to review studies of training working memory and intelligence throughout the week, with an eye towards identifying the necessary and sufficient characteristics for a training regime to show transfer to intelligence tests.
Updating Training Shows only Near Transfer (At the Group-Level)
Dramatic Play and Executive Function Training (with Far-Transfer)
New Approaches to Training Working Memory [A presentation I gave to LearningRX]
Pharmacological Enhancement of Working Memory
Enhancing Memory with Visual Flicker
Book Review: The Future of the Brain
Caffeine: A User's Guide to Getting Optimally Wired
Filtering Perception to Save Memory
Cytoskeletal Enhancement of Memory
"Deprogramming" Through Meditation and Hypnosis
Traumatic Brain Injury: Interventions and Treatment
Yes, it bothers us so much that we published an interview with Martin Buschkuehl...
Now seriously, this is far from being the first study where transfer has been shown. The problem, in my view, is that we seem to be looking for magic pills that work for everyone and everything, vs. identifying what specific cognitive skills may be acting as "bottleneck" in particular groups. individuals and addressing those (as has been shown in several contexts: ADHD, dyslexia, stroke/ TBI rehab, driving, flying).
Btw, I asked Martin: A common question we get is, How are computerized programs like the one you used fundamentally different from, say, simply doing many crossword puzzles?
"In terms of why our program worked, I could say that the program has some inherent properties that are at least in this combination unique to our training approach. Our program is:
- Fully adaptive in real-time: The person using the program is truly pushed to his or her peak level all the time, thereby "stretching" the targeted ability.
- Complex: We present a very complex task, mixing different forms of stimuli (auditory, visual) under time pressure.
- Designed for Transferability: The tasks can be designed in a way that do not allow for the development of task-specific "strategies" to beat the game. One needs to truly expand capacity, and this helps ensure the transfer of to non-trained tasks.
This is very different from enhancing task-specific capacities, such as memorizing lists of 100 numbers, which have been shown not to necessarily transfer to related domains."
I interviewed Art Kramer last week, will be publishing the notes late this week/ early next. He provides a great overview of both the cognitive benefits of physical exercise and cognitive training applications.
is there a flash version of this training protocol available? that would be sweet! I could goof off and improve my mind at the same time.
If you're interested to try it out, my wife and I have written an open source application that implements dual n-back. It's pretty fun and pretty challenging. The game is at: http://www.soakyourhead.com/N-Back.aspx and the source code is at http://www.soakyourhead.com/dual-n-back-open-source.aspx .
Also, I enjoyed your caffeine post. Really interesting stuff.
Here is another implementation of the dual-n-back task that includes some fairly novel "levels" as the difficulty increases: http://cse.ucsd.edu/~ckanan/Game.html
It runs in Linux, Windows, and OSX for all of the mainstream browsers.
"...varied stimuli; the rapid binding of particular items to their temporal order; and the maintenance, manipulation, and monitoring requirements for remembering a list of items, iteratively updating that list with the presentation of new items, and the iterative comparison of new stimuli with those in memory..."
Sounds as complicated as trying to fly a big airplane!
Mac, I think you raise an interesting point. Shouldn't any task, complicated enough to consider a profession, actually enhance gF through it's varied executive demands? This may be a case where "scientific reductionism" works against scientific discovery - maybe you need the "whole shebang" to actually far transfer.
Mac, Chris: good point, probably it does.
Now, the follow-up question would be, is that the only route? or the most efficient one? Not necessarily.
You may be familiar with the research by Kramer, Gopher and Shebilske on training, precisely, pilots in complex environments (military).
From my interview with Gopher on cognitive simulations (http://www.sharpbrains.com/blog/2006/11/02/cognitive-simulations-for-ba…):
"Daniel Gopher: My main interest has been how to expand the limits of human attention, information processing and response capabilities which are critical in complex, real-time decision-making, high-demand tasks such as flying a military jet or playing professional basketball. Using a tennis analogy, my goal has been, and is, how to help develop many Wimbledon-like champions. Each with their own styles, but performing to their maximum capacity to succeed in their environments.
What research over the last 15-20 years has shown is that cognition, or what we call thinking and performance, is really a set of skills that we can train systematically. And that computer-based cognitive trainers or cognitive simulations are the most effective and efficient way to do so.
This is an important point, so let me emphasize it. What we have discovered is that a key factor for an effective transfer from training environment to reality is that the training program ensures Cognitive Fidelity, this is, it should faithfully represent the mental demands that happen in the real world. Traditional approaches focus instead on physical fidelity, which may seem more intuitive, but less effective and harder to achieve. They are also less efficient, given costs involved in creating expensive physical simulators that faithfully replicate, lets say, a whole military helicopter or just a significant part of it.
AF: Very interesting. In the Serious Games Summit this week we are seeing a number of simulations for military training that try to be as realistic as possible. Are you saying that they may not be the best approach for training?
DG: The need for physical fidelity is not based on research, at least for the type of high-performance training we are talking about. In fact, a simple environment may be better in that it does not create the illusion of reality. Simulations can be very expensive and complex, sometimes even costing as much as the real thing, which limits the access to training. Not only that, but the whole effort may be futile, given that some important features can not be replicated (such as gravitation free tilted or inverted flight), and even result in negative transfer, because learners pick up on specific training features or sensations that do not exist in the real situation."
Not sure you hear this enough Chris ... but this blog is extremely good. Always interesting ... always relevant.
After reading the full paper and contacting Jaeggi, I was so impressed by this study that I developed a software program using the same progressive dual n-back method so that anyone can achieve these improvements at home.
I've been using the software myself - of course - and the improvements I've noticed are indeed examples of far transfer and far from subtle. It's improved my ability to solve difficult crossword puzzles and even play the piano. This may seem odd, but after going through the training for more than twenty sessions now, it makes sense. For instance, while working on a crossword puzzle I can hold more possible solutions and variations in my mind and juggle them to see what works. And when I'm playing the piano I'm better able to think ahead of what my hands are doing...
If anyone wants a solid program with some nice features (archival of scores and history) you can get it at iqtesttraining dot com.)
mind evolve, llc
Have the Jaeggi-Buschkuehl dual n-back results been replic- ated?