Cognitive training is a hot topic in educational neuroscience. Can training a certain cognitive function lead to gains in academic performance? This is an exciting question for researchers who (a) want to see real-world impact of their research, and (b) aim to use training as a tool to further inform their theories. But what makes a good training study, and what are the key aspects to be considered throughout the design process? This is the first in a series of posts that examines the key aspects of designing a cognitive training study.
An important consideration is the type of training programme. Will the programme provide practise of difficult tasks (process-based training), or will it train a new strategy to bring to the task (strategy-based training)? Repetition of a task through process-based training may lead to increased automaticity and efficiency, while a new method learnt through strategy-based training may enable a toolkit approach where students can choose the best tool for each problem.
Taking one example, a training programme might aim to improve working memory, since this is known to be important for many academic outcomes. Process-based training would see the student practise working memory tasks, perhaps in an adaptive programme that gets harder or easier depending on performance. On the other hand, strategy-based training would provide explicit explanations of how to perform in the task.
A third approach, that can be considered a type of strategy-based training, is to train metacognitive knowledge. This time, the student might be given a mechanistic explanation of why working memory is so important in their academic studies. They might be explicitly told when to use working memory. Here the aim is not necessarily to train working memory, but rather to train the use of working memory within a certain context. Perhaps a student has adequate working memory but has not previously considered its use in this subject domain. Metacognitive training might allow the student to identify when working memory is needed, and to implement an appropriate strategy.
In the process- and strategy-based approaches we would expect to see an improvement in working memory. This in turn might lead to improvements in academic performance. Conversely we might not expect any working memory improvement through a metacognitive approach, but we might nonetheless see an academic improvement.
Perhaps then, the most effective approach would be to combine all three of the above: Provide repetition of the task, train specific strategies, and increase metacognitive awareness of the cognitive functions involved in a task or subject domain. In terms of educational outcomes, this might be the most likely to show an impact. The challenge for the researcher is that in providing all three, we are no closer to discovering what the “active ingredient” causing change is.
The final consideration in choosing the type of training study is that different methods may be effective for different learners. Perhaps some students require the process- and strategy-based training to improve their baseline working memory ability, while other students already have very good working memory but might benefit from metacognitive training to help them identify when to use this ability. Therefore the type of training programme might depend on the population that the programme targets.
These considerations highlight the importance of designing the training programme from a cognitive theory. While the ultimate aim of educational neuroscience is of course to improve education, as scientists, researchers must use their theory to choose the best type of training programme to answer a particular question. Simply providing training and hoping for a positive outcome is not enough. The outcome must drive theory forward, and the training programme must be carefully designed to enable this. For the scientist, an important result is one that can tell us about the mechanism behind change, rather than one that shows improvement without a good theory about why this change occurred.
This post was informed by this highly recommended article: