Can AI better predict why children struggle at school?

Researchers from the Medical Research Council (MRC) Cognition and Brain Sciences Unit at the University of Cambridge have used machine learning – a type of computer algorithm – to identify clusters of learning difficulties which did not match the previous diagnosis children had been given.

The study, published in Developmental Science, used a sample of 530 children (ages 5­–18) who were referred to the Centre for Attention Learning and Memory (CALM) by health and education professionals because they were struggling in school. All the children in the sample completed a number of cognitive and learning assessments, they underwent a structural MRI scan, and their parents completed behaviour questionnaires.

Based on the data collected from these tests, the computer algorithm identified four groups that the children could be matched to: (1) broad cognitive difficulties and severe reading, spelling and maths problems; (2) age-typical cognitive abilities and learning profiles; (3) difficulties with working memory skills; and (4) difficulties with processing sounds in words.

While these groups aligned closely with other data on the children, such as the parents’ reports of their communication difficulties and educational data on reading and maths, there was no correspondence with their previous diagnoses. To check if these groupings corresponded to biological differences, the groups were checked against MRI brain scans from 184 of the children. The groupings mirrored patterns in connectivity within parts of the children’s brains, suggesting that the machine learning was identifying differences that partly reflect underlying biology.

The researchers conclude that these findings reinforce the need for children to receive detailed assessments of their cognitive skills to identify the best type of support.

Source: Remapping the cognitive and neural profiles of children who struggle at school (September 2018), Developmental Science

Early years childcare and cognitive and socioemotional development at age three

From September 2013, two-year-old children living in the 20% most disadvantaged households in the UK became eligible for 15 hours of funded early childhood education and care (ECEC) per week. This was extended in September 2014 to two-year-old children living in the 40% most disadvantaged households. The longitudinal Study of Early Education and Development (SEED) followed the progress of children from approximately 6,000 families, from ages two to seven, to help provide the Department for Education (DfE) with evidence on the effectiveness of early years education.

This latest DfE impact study presents findings for 4,583 children from the SEED longitudinal study and focuses on the relationship between the amount and type of ECEC between age two and three and children’s cognitive and socioemotional development at age three. After controlling for home environment and demographic factors, the amount of ECEC received between the ages of two and three was found to be associated with cognitive and socioemotional developmental benefits. There was also evidence that ECEC is associated with higher cognitive verbal ability (naming vocabulary). The study also found that children who participated in more than 35 hours of ECEC per week between two- and three-years-old had higher levels of conduct problems and lower levels of emotional self-regulation than children receiving less than two hours a week, although this group of children comprised only 3.25% of the sample (149 children). The researchers note that the children who received more than 35 hours of ECEC between two- and three-years-old were also more likely to have started early (ie, during the first year of life), and that this early start combined with high use when aged two to three is a significant factor behind these effects.

Source: Study of early education and development (SEED): Impact study on early education use and child outcomes up to age three. Research report (July 2017), Department for Education. Reference: DFE-RR706

Bookworms benefit

New research published in the British Educational Research Journal has found that reading for pleasure is more strongly linked to cognitive progress in adolescence than parents’ education.

Data on 3,583 16-year-olds was taken from the 1970 British Cohort Study. This study follows the lives of people born in England, Scotland, and Wales in a single week of 1970, collecting information on health, physical, educational, and social development, and economic circumstances among other factors.

The authors set out to explore the relative importance of economic and cultural resources in determining class differentials in educational outcomes. They found that the home reading culture (including reading to the child and parents reading books and newspapers) was linked to children’s test scores, and this had a role in mediating the influence of parents’ education and also to some extent in mediating parents’ social class.

Childhood reading was linked to substantial cognitive progress between the ages of 10 and 16. Reading was most strongly linked to progress in vocabulary, with a weaker, but still substantial link to progress in mathematics.

The research also found that parental education was much more strongly linked than parental social class to both vocabulary and mathematics scores, broadly supporting the idea that cultural resources matter more to cognitive outcomes than economic resources.

Source: Reading for Pleasure and Progress in Vocabulary and Mathematics (2015), British Educational Research Journal, 41(6).

The varying impact of an earlier years programme

A paper from MDRC analyses variation in the effects of the Head Start programme in the United States using data from the Head Start Impact Study.

Head Start is the largest US federal programme for early years development of disadvantaged children and has served more than 30 million children since 1965.

The MDRC paper confirms previous studies that suggested substantial variation in the effects of Head Start in relation to the individual, subgroup, and between Head Start Centers.

The main findings were:

  • Head Start improved cognitive outcomes in children with the lowest cognitive skills and tended to reduce disparities between children in key cognitive outcomes.
  • Dual-language and Spanish-speaking children with low pretest scores gained the most from Head Start.
  • Much of the positive effect of Head Start came from mitigating for limited prior English; the positive effect on children with limited English persisted for at least three years.

The added value of Head Start compared with local alternatives varied substantially between Centers and reflected differences in provision (such as hours of care, teacher education, and classroom quality).

Some Head Start Centers were much more effective than alternatives (including parental care) and others were much less effective than alternatives.

Source: Quantifying Variation in Head Start Effects on Young Children’s Cognitive and Socio-Emotional Skills Using Data from the National Head Start Impact Study (2015), MDRC.

Preschool fade out

A working paper from the Curry School of Education at the University of Virginia investigated the persistence of effects of early childhood interventions and asked whether these effects fade systematically. The study found that “by the end of kindergarten [Year 1] children who attended preschool are no longer outperforming on measures of reading and math relative to their peers.”

The study used two datasets: a 2010 kindergarten cohort, which the authors used to explore the relationship between attending preschool and cognitive and behavioural skills at school entry and through first grade; and a 1998 kindergarten cohort, which enabled comparisons of the effects of preschool attendance in 1998 and 2010.

In 1998 and 2010, children who attended preschool showed benefits in cognitive skills when they started school. Analysis of the 1998 cohort showed that these benefits persisted through kindergarten and first grade (equivalent to the end of Key Stage 1). Analysis of the 2010 cohort showed that, despite their advantage when first attending kindergarten, preschoolers had no cognitive advantage over other children by the end of the kindergarten year.

The fade-out effect was similar whether children attended preschool full-time or part-time and irrespective of kindergarten class size, length of the kindergarten day, and other quality measures.

The results are similar to a recent report in Best Evidence in Brief, which compared children in England who went to preschool in the late 1990s with those from 2005 onwards. Again, the older children’s benefits persisted, whereas the younger ones’ faded.

Source: Working Paper: Preschool Fade Out (2015), Curry School of Education

FITKids programme benefits body and mind

Results of a randomised study that compared pupils who attended FITKids (a daily after-school fitness programme) to those who did not showed benefits for the FITKids group in attention, memory, and task-switching.

The study involved 221 eight- to nine-year olds matched by age, gender, ethnicity, socio-economic status, and aerobic fitness during the school years 2009-2013. The experimental groups participated in the FITKids programme for two hours a day after school for nine months. Each day they spent 30 minutes at activity stations, followed by a rest/education period then about 45 minutes of organised games. The control groups were put on a waiting list for the FITKids programme.

All groups were pre- and post-tested on fitness and cognitive measures. Both groups demonstrated post-test gains in aerobic fitness, but these were significant only in the experimental group. The experimental group demonstrated twice the accuracy in cognitive tasks at post-test compared with the control group.

The authors concluded that a daily after-school fitness programme improves brain health. They warned that policies that seek to increase academic achievement by replacing physical education and break times with academic classes may inadvertently do more harm than good.

Source: Effects of the FITKids Randomized Controlled Trial on Executive Control and Brain Function (2014), Pediatrics 134(4)