What if training ourselves on one task yielded improvements in all other tasks we perform? This is the promise of the cognitive training movement, which is increasingly showing that such “far transfer” of training is indeed possible, while short of being “universal transfer.” Interestingly, this phenomenon might be most likely to occur for some of the most abstract and challenging cognitive functions.
New evidence for this claim comes from an in-press article at Psychological Science, by Persson & Reuter-Lorenz. The authors used several tasks which have been shown to engage the left inferior frontal gyrus (lIFG) of the prefrontal cortex. The best theoretical interpretation of this activity is that the lIFG is “biasing” the activity of particular representations in posterior cortex. For example, you may need more of such “biasing” when trying to come up with a verb that’s related to “giraffe” than one that’s related to “lion” (lion has some obvious associates [roar! eat! hunt!] and is therefore less likely to require any help from lIFG). This is an example of a verb generation task; like many other similar tasks, it engages the left IFG.
Persson & Reuter-Lorenz used verb generation as a test for transfer – that is, subjects were never trained on this task, or the others used to assess transfer (which included tests of recognition and recall for items where lIFG is engaged). Instead, subjects were trained for 8 hours each on one of three versions of other tasks: all of the trained tasks required that subjects remember the groupings of recent stimuli (whether letters, words, or faces) but for some subjects these groupings were easily confusable (known as a “recent probes” version). It is those versions of the tests which are thought to involve lIFG, and on which training was predicted to have the largest effects.
Consistent with this prediction, only the group trained on the easily-confusable memory tasks improved on the untrained tasks. In addition, the observed improvement was roughly equal on all three of the untrained tasks, indicating impressive generalization of the trained ability. A critical additional analysis showed that the improvement showed by each individual on the trained tasks predicted each individual’s benefits on the untrained tasks. This analysis is an important compliment to the group level analyses which are more typically done in experimental psychology, because it shows a one-to-one correspondence between training and transfer of that training.
A similar effect has been observed for training of a different brain region – the striatum – which is thought to be involved in “updating” the representations in prefrontal regions (like lIFG). Thus it appears that at least two critical components of this executive control system (active maintenance/biasing and updating) can be affected by training.
An interesting message from the Persson & Reuter-Lorenz work is that transfer of training may occur only when the same brain regions are activated across the training & transfer tasks. This was not observed in the wider fronto-parietal network in a study training the capacity for “cognitive updating,” but they also did not perform the critical analyses at the individual level – so it’s a little bit difficult to know what to make of that work, as discussed previously.
Another component of the executive control system is “switching” – as when we must rapidly change from performing one task to performing another. New work from Minear & Shah seems to show that some aspects of this capacity can be improved, but that paper is not yet available… I’ll post an update when it is.