I set up a volunteer Mastermind league in my small village school one winter after a spate of wet lunchtimes. By chance my then current Open University study involved two learning styles, impulsivity and reflectivity. Rational, logical thinkers reflecting on their moves would surely defeat the pot luck brigade incapable of high intellectual reasoning... or so I supposed.

I labelled all 14 players, aged 8 to 11, ”...” or “R” from close knowledge of each. Sometimes I watched them. John and Petra were painstaking, often minutes over each peg. Simon and Roger were the opposite, making instant moves; Roger once set out his pegs in a pattern he’d already tried - and he’d had feedback! I left them to it. After playing each other once, both groups alternated at the top of the league evenly with Simon, in fact, top and Roger, two years younger than Simon, fourth. Petra and John were second and third. In effect I had a league table of raw ability tempered by maturity, virtually an 11-plus selection list had we still had it. The same absence of significant difference had shown in the far more scientific Open University experiment I was studying.

Another OU unit, on creativity, contrasted divergent and convergent thinkers, citing artists as more divergent and scientists convergent, again compartmentalising human ability by style. I remember Peter, an able child whose parents were both computer experts high on logical, mathematical thought. He was a chip off the same block. Our balanced curriculum encouraged art, music, poetry and free writing. He applied reasoned analysis in problem-solving, but became a competent all-rounder, his work reflecting qualities normally assigned to artists. We tapped powers possibly more dormant than missing.

A child who arrived from the 11-plus class in her previous school responded badly to our more open atmosphere. Sitting next to a five-year-old who asked for help when writing under his picture of a cat she responded tartly, “Can’t you spell cat?” She played the piano competently but mechanically, with poor dynamics. Yet in months she was sensitively and richly describing the agonies of a wasp buzzing to death on a windowsill after our precautionary spray with insecticide.

We debate differences in the way children learn, but lapse too easily into polarity. Are there really so many intelligences? Surely human beings are multi-capable and it is life, including home background and teaching, that shapes performance.

Phonics, streaming, real books, testing and learning styles are blind alleys if they are seen as alternatives rather than integrated into existing knowledge and experience. It is clear we need most of them some of the time, but rarely all the time. Phonics do not help a child to read the word “ought”. Graded reading books strengthen the vital role of repetition in learning, but real books encourage parental interest and a love of literature. Children of like ability need the stimulus of similar others, but also the chance to express their abilities within the wider group.

Testing can reveal need but also confine the curriculum. We need to balance and integrate competing ideas.

Despite millions spent on systems, buildings, examinations and structures, research continues to show 50 per cent of educational outcomes still reflect home background, while the rest comes from quality of teaching.

These are better investment targets, and they encompass those early stages where vital work on cognition is needed. Home-school partnership remains an enduring success of the 1967 Plowden report; Americans have found extra money invested in such relationships repays up to ten-fold long-term to the exchequer.

Back to my village school. After three years teaching cooking, a volunteer parent told me she finally understood why her own four children were so different. People are different, but they possess all the latent potential we unwisely assign to particular labels. We need classrooms which allow each pupil to work in the way nature designed them, while letting them also evolve as nature allows.

Mervyn Benford is an education consultant