Working memory and Learning


I am indebted to Tracy Packiam Alloway for her work on memory and learning. I have summarised here some of her findings. Forgive me if I’ve written these before!
Working memory is the term used by psychologists to refer to the ability we have to hold and manipulate information in the mind over relatively short periods of time. It provides a mental workspace or jotter that is used to store important information in the course of our everyday lives. Working memory is limited in capacity and this capacity varies between individuals and is affected by characteristics of the material that is being stored.

We usually experience mental activities that place significant demands on working memory as a kind of mental juggling in which we try to keep all elements of a task going at the same time. Often, the juggling will fail either because the capacity of our working memory is exceeded, or because we become distracted and the task in hand is displaced by other information.

Working memory is distinguishable from short term memory because it involves manipulating information mentally and holding it in one’s head for a time. Recalling a phone number in order to make an immediate call is a good example of an activity that depends on short term memory. Working memory is a term for more complex tasks, such as following lengthy directions about how to reach a location.

The expression ‘long term memory’ has four components: episodic, autobiographical, semantic and procedural.

  • Episodic memory is about details particular experiences such as what we had for breakfast.
  • Autobiographical details from away back also form part of our long term memory;
  • and semantic knowledge is, for example, knowing that Paris is the capital of France.
  • Procedural memory lasts a life time, once a skill is established, for any information that can be used automatically such as ‘knowing’ how to drive.

Working memory increases from childhood through to adolescence when adult levels are reached, in most people.  Increases in working memory capacity with age relate to improvements in the efficiency of processing and of attention. It is related to, but distinct from, intelligence. Because it is independent of factors relating to the child’s background and learning opportunities it provides a comparatively pure measure of learning ability.

Thus children with learning difficulties in reading and maths typically have poor working memory capacities, and their memory scores predict the severity of their learning problems. Poor working memory does not appear to be due to more general factors such as language difficulties or general cognitive delay. The poor rates of learning in children with low working memory capacities are due in large part to memory overload in structured learning activities which causes them to forget crucial information and so to fail in these tasks.

Memory overload leads to difficulties following instructions, in completing tasks that combine storage and demanding mental processing, and problems in keeping track of their progress in complex tasks. These frequent task failures impair learning in key academic domains – not least because they have enormous impact on the child’s sense of self efficacy.

Many children with poor working memory appear to be inattentive and highly distractible.

Interventions are possible. Teachers need to:

  • be aware of the warning signs of working memory failure,
  • monitor the child, 
  • evaluate working memory load if the warning signs are detected,
  • reduce the working memory load if necessary,
  • repeat important information, use memory aids
  • encourage the child to use strategies for supporting working memory.

 Teachers who are particularly effective at implementing the intervention approach combine a number of principles and strategies in a single activity to provide a strong network of working memory support.


Inchworm or grasshopper. What’s your thinking style?



Maths can cause problems in ther same way as reading and spelling does for learners with dyslexia. This is because there are differences in the way thinking, processing and memory are organised. So you get the pupil who can understand and do higher order maths but who makes ‘careless’ errors  from misreading the problem, or reversing digits or sequences of digits, that make nonsense of the calculations. Confusion about maths symbols and language exacerbates memory and processing difficulties.

A student may be able to understand the maths concepts, but if s/he can’t read the question, or read it sufficiently accurately, s/he will be unable to work to her/his mathematical ability. A talking word processor may help with ‘wordy’ problems, but not when formulae are invovled.

And of course, a learner with dyscalculia will have much more significant problems. Dyscalculia is ‘a specific learning difficulty or difficulty involving innate difficulty in learning or comprehending maths’.

Dyscalculia may be likened to dyslexia in that an individual might suffer difficulties with computation, with no impairment of abstract mathematical reasoning abilities. I don’t know where I heard this, but there is some evidence to show that Einstein was not very good at adding up. Likewise, someone may be more than capable of understanding sophisticated language but struggle to decode or encode individual words.

Steve Chinn writes:

Children with special needs have more problems with aspects of mathematical long-term memory, notably retrieving basic facts from memory. They are slower to retrieve and process information. Also, they often have poor short term and working memories, both of which are key pre-requisite skills for mental arithmetic and indeed mathematics in general. …

Children who do perform badly in maths do not remember facts and processes as effectively as their better performing peers. This problem is exacerbated by their lack of effective compensatory strategies. Their strategy for overcoming the deficit is counting and counting is a low level cognitive skill with many disadvantages. Better performing peers use linking strategies which are based on understanding numbers, operations and how they relate.

 In an interesting article Chinn writes

 A number of researchers over the years have identified two thinking styles. Although they have a good selection of labels, basically one style is formulaic and sequential and the other is intuitive and holistic. In work done with American colleagues, we call the two styles ‘inchworm’ and ‘grasshopper’. Most learners will lie somewhere on a continuum between the two extremes of style and indeed this blend is likely to be the most successful style as success in maths tends to require flexibility in thinking.