'Take your bag, put your shoes on, don't forget your lunch and give this note to your teacher.' A minute later the child stands in the hall with one shoe on, trying to remember what came next. This isn't stubbornness or a lack of intelligence: it is simply how human memory works. One of the most practical approaches in the science of intelligence explains why, and what helps.
The mind as an information system
Since the 1960s many psychologists have compared the mind to a computer. Information comes in through the senses, is processed, stored in memory and retrieved when needed, and control processes steer the whole operation. Researchers in this tradition ask not only how clever a child is, but what exactly happens in the mind while a problem is being solved, and at which step things go wrong. The brain is not a computer, of course, but the comparison helps to break thinking into steps that can be studied and taught.
The first step is attention. Out of a flood of sounds, sights and thoughts it lets only a little through for further processing, like a filter. What never gets through cannot be remembered: a child who was looking out of the window during the explanation hasn't forgotten it, because it never came in.
Working memory: a small desk
Working memory is where we hold and juggle information right now: a phone number until we dial it, the start of a sentence while we read the rest, the interim result of a sum. It is a very small desk. In 1956 George Miller wrote about 'the magical number seven, plus or minus two'; later research, summarised by Nelson Cowan in 2001, suggests that we can really keep only about four chunks in mind at once.
A chunk is a unit that means something to us. Ten separate digits are hard to hold, but the same phone number split into a few familiar groups is easy. That is why long instructions get lost: the fourth and fifth steps simply push the first ones off the desk. This is not a sign of poor ability, and in young children the desk is even smaller.
Long-term memory: how experts organise knowledge
Long-term memory is practically unlimited; what matters is how knowledge is organised in it. Experts differ from beginners less in the power of their memory than in the structure of their knowledge. In a well-known 1981 study by Chi and colleagues, experts sorted physics problems by the underlying principle, while beginners sorted them by surface features, such as whether the problem involved a slope or a spring. Well-organised knowledge forms large meaningful chunks that take up little space on the small desk and leave room for thinking. Knowledge and thinking are not rivals: the better connected the knowledge, the easier the thinking.
Speed, planning and learning to learn
Many people believe that a clever child is always a fast child. Robert Sternberg found something different: good problem solvers spend more time at the start understanding the problem, and then get through the rest of it more quickly. Impulsive children grab the first answer that comes to mind; reflective children first look over the whole task. What matters is not raw speed but how time is used.
Sternberg also described the 'executive' steps that steer thinking: recognising and defining the problem, choosing a strategy, planning and allocating time, monitoring progress and using feedback, that is, learning from results and mistakes. The ability to watch and steer one's own thinking is called metacognition.
Can these skills be trained? Studies that trained thinking skills give a clear answer: drilling one type of task improves only that task, while lasting gains come when children learn general strategies and practise them on many different tasks. Feuerstein's Instrumental Enrichment programme (1980) is a well-known example of this approach. In 2013 Dunlosky and colleagues reviewed popular study techniques and found especially strong support for two of them: retrieval practice (recalling material from memory, for example by self-testing) and spacing (spreading practice over several days). Rereading and highlighting, though popular, help much less.
What helps at home
- Give one instruction at a time. For younger children one or two steps are enough; ask them to repeat the steps back.
- Break big tasks into chunks. A short written or picture checklist frees working memory for the task itself.
- Plan, do, check. Before homework, ask: 'What do you need to do? Where will you start?' Afterwards: 'Did it work? What would you change?'
- Praise careful thinking, not just speed: 'I like the way you checked your answer.'
- Choose self-testing over rereading: close the book and retell, use flashcards, have a quick family quiz at dinner.
- A little every day beats a lot the night before, so spread revision over several days.
- Protect sleep. A good night's sleep helps the brain hold on to what was learnt, and late-night cramming at the expense of sleep backfires.
Teachers can use the same ideas: short steps written on the board, a quick quiz on earlier material at the start of a lesson, and time to think before anyone answers.