Tests of verbal and non-verbal reasoning skills are now taken by around 1.5 million pupils every year. They appear similar to IQ tests and raise the spectre of determinism, feared by critics of the old 11-plus.
There are two main testing systems. One is the CAT (cognitive abilities test), developed by nferNelson. A range of CATs cover verbal and non-verbal reasoning, plus mathematical aptitude. Around one million pupils a year take them, mostly in Year 7 or Year 9.
The other system is MidYIS, introduced in 1997 to carry out baseline assessments in Years 7 to 9 and now used in one in three English secondaries.
Supporters of such cognitive assessments insist they are different from the 11-plus straitjacket because they do not measure innate ability. "Cognitive assessment is not a measure of fixed potential," insists Dr Steve Strand of nferNelson. "Good teaching can change scores and outcomes."
Motivation, experience and education all have an impact on test results. Dr Strand says that in the space of a year, 30 per cent of pupils can alter their CAT scores by as much as 10 per cent. This is why, he says, regular testing is helpful.
But CATs can still be a valuable predictor of educational achievement because they measure abilities such as problem-solving, seeing patterns in data, reasoning with symbols and making links between different concepts.
These are all important skills for successful learning.
Predictions of individual pupils' results at GCSE or Scottish Standard Grade, based on their test scores, are now an important part of the assessment service.
Thousands of pupils are tracked from Year 7 to Year 11, showing a close correlation between CAT scores and attainment at GCSE and Standard Grade.