I had been given my instructions: "Three small strawberries, two with chocolate sauce, one with strawberry sauce, two vanillas both with chocolate sauce and one can of lemonade." But, of course, by the time I reached the front of the queue, I no longer felt confident - I was convinced the order had somehow transformed itself in my mind. Hand signals were made across the beach and a runner had to be dispatched to remind me of the details.
This was a classic failure of my working memory - my ability to temporarily maintain information in my consciousness.
Our working memory system is often described in terms of two "slave" systems: a visuospatial sketch pad for visual and spatial information and a phonological loop where we continually rehearse language information (such as an ice-cream order) to maintain it in consciousness.
The boss of these two slaves, and in charge of co-ordinating their contents, is the central executive. Working memory is vital for holding on to information just received or, just as importantly, a place to work on information pulled from our long-term memory. It's not unusual, in everyday distracting environments, for the central executive to fail in its managerial duties and lose control of what its slaves are doing. But even with the best management, slaves have their limits.
The average limit for working memory is about seven chunks of information, and this limit affects our educational potential. It is important when we are learning how to do something for the first time. When the brains of adults learning long multiplication were scanned, increases in frontal areas were seen when their training first began. These reflected the load on working memory these adults were experiencing as they consciously went through the steps they had been taught.
However, after practising for several weeks, this activity reduced and was replaced by activity in more posterior regions (away from the front of the brain). This shift reflected how rehearsing the processes of multiplication had turned a consciously held series of steps into an automatic skill. Practising thinking processes helps make them perfect, but it also develops them into skills, simultaneously liberating our working memory so that we can learn more. That's one reason why skills might be more highly valued than knowledge. However, beyond monitoring and encouraging pupils to rehearse what has been taught, whether we can "teach to skills" is a moot point. Skills, by psychological definition at least, belong to the class of memory known as non-declarative.
Once something has become a skill, it isn't necessary, or sometimes even possible, to explain fully how we can achieve it. We may know how to ride a bike or deliver a good presentation, but if we were asked how we can do it, we may find it difficult to provide a complete set of instructions. Such abilities have become automatic, non-declarative and unconscious - they are skills that have developed through rehearsal of what, once, we had to consciously recapitulate.
Developing skills liberates our working memory and this can be empowering, but several attempts have been made to increase the size of working memory itself through training it. There's a number of technological gadgets aimed at developing brain function, but there has been little solid research to suggest training of any kind can achieve this - until now.
Susanne Jaeggi and colleagues at the University of Bern in Switzerland trained the working memories of students for up to 25 minutes every week day using an "n-back" task. This required students to watch a computer screen present a new stimulus every three seconds and say whether it matched a stimulus presented "n" trials ago. If they got it correct, n got larger, so that the students had to maintain a larger number of previous stimuli in their working memory in order to answer correctly. If they got it wrong, n got smaller and the task got easier. This way, students were always thinking at the edge of their working memory ability.
Jaeggi and her colleagues showed that the students improved their performance with the number of sessions they received, as if their working memory was increasing. What is more startling is what happened to their "fluid intelligence". This is a measure of how well we can adapt our thinking to a new situation and relates to professional and educational success.
In their study, published this year, Jaeggi and her colleagues showed that increases in fluid intelligence followed those in working memory. So, it would appear, new types of "brain training" may soon become available. But if, like me, your working memory is groaning under the pressures of modern life and you haven't got the time for this, there is another solution.
"Showing your working" isn't just useful for pupils wishing to convince their teacher they were on the right track. It also relieves the burden on working memory. That is why I'll be taking paper and a pencil to the beach
Dr Paul Howard-Jones is a senior lecturer in education and co-ordinator of the Centre for Psychology and Learning in Context (CPLiC) at the Graduate School of Education, University of Bristol
Delazer, M., Domahs, F., et al (2003) Learning Complex Arithmetic: An fMRI study. Cognitive Brain Research, 18, 76-88
Jaeggi, S.M., Buschkuehl, M., et al (2008) Improving Fluid Intelligence with Training on Working Memory. Proceedings of the National Academy of Sciences (USA), 105:19, 6829-33.