Evaluating a neuroscience-inspired learning activity to improve science and maths in primary-school age children by training inhibitory control skills (the "UnLocke" project)

INTERVENTION: In each school, years 3 and 5 (approximate ages 8 and 10) will be randomly assigned to one of three conditions by the external evaluators (National Foundation for Educational Research; NFER). Each school in the main trial will have one year group in the main intervention (Stop and Think: interference control training in mathematics and science) group and the other in either the active control (See+: socio‐emotional cognition training) or the passive control group (lessons as usual). All classes of a given year group within a school will be in the same condition. The main intervention (Stop and Think) will run for 12 minutes at the start of a maths or science lesson, three times per week, for 10 weeks. The active control group will run See+ for 12 minutes at the start of any lesson which is not maths or science, three times per week, for 10 weeks. Both Stop and Think and See+ run as a whole‐class interactive whiteboard activities, led by the class teacher. Training will be provided to all teachers in the main intervention and active control groups by the UnLocke team before the 10‐week intervention period, and monitoring and support will be in place by a designated UnLocke field officer throughout the intervention. CONDITION: Inhibitory control and maths and science performance in primary school education ; Not Applicable PRIMARY OUTCOME: Cognitive and neuroimaging measures will be collected 0‐2 months prior to the start of the intervention (T1) and 0‐2 months after the end of the intervention (T2). In addition, maths and science performance will be assessed with standardised tests by the external evaluators 0‐4 months after the end of the intervention (T2); ; 1. Change in science and maths misconception accuracy from T1 to T2 (Hypothesis 1a). Science and maths misconception performance will be measured using a computerised task in the cognitive sample (N=184) and with an additional computerised task in the neuroimaging subsample (N=80); 2. Performance on standardised maths or science maths test at T2 (Hypothesis 1b). Each child in the cognitive sample (N=184) will be randomly allocated by NFER to complete either a science or a maths standardised tests, which will be administered by NFER; 3. Change in brain activation during maths and science misconceptions reasoning from T1 to T2 (Hypothesis 2). Whole‐brain activation during the completion of maths and science misconception trials will be measured using a 1.5T magnetic resonance imaging (MRI) scanner in the neuroimaging subsample (N=80). In addition to whole‐brain analyses corrected for multiple comparisons, region of interest (ROI) analyses will be performed in dorsolateral prefrontal cortex, inferior frontal gyrus, and anterior cingulate bilateral clusters. Analyses will be performed with and without RT as a covariate of no interest. INCLUSION CRITERIA: 1. In Years 3 and 5 (approximate ages 8 and 10) of one of our participating UnLocke schools 2. Male or female 3. Informed parental consent (opt‐in) SECONDARY OUTCOME: 1. Change in science and maths misconception reaction times (RT) from T1 to T2 (Hypothesis 1a). Science and maths misconception performance will be measured using a computerised task in the cognitive sample (N=184) and with an additional computerised task in the neuroimaging subsample (N=80); 2. Change in inhibitory control tasks performance from T1 to T2 (Hypothesis 3a). Two domain‐general computerised inhibitory control tasks (Fish Flankers task, Whack‐a‐mole) and one pen and paper task (Chimeric animals) will be assessed in the cognitive sample (N=184). Key measures will be the accuracy and RT congruency costs (incongruent vs. congruent trials) and reversing costs (reverse vs. congruent trials) in the Fish Flankers task, the percentage No‐Go errors in the Whack‐a‐mole task, and the mixing cost in the Chimeric animals task (difference in accuracy scores between sheets of congruent‐only trials and mixed sheets of incongruent and congruent trials); 3. Change in brain activation during domain‐general inhibitory control tasks from T1 to T2 (Hypothesis 3b). Whole‐brain activation in mixed (congruent + incongruent) vs. congruent blocks during the Animal size Stroop task and whole‐brain activation in Go/No‐Go vs. Go blocks during the Pokemon Go/No‐Go will be measured using a 1.5T magnetic resonance imaging (MRI) scanner in the neuroimaging subsample (N=80); 4. Change in brain structure from T1 to T2 (Hypothesis 4). Changes in local grey matter and white matter volumes (analysed with VBM or Freesurfer) will be analysed in the whole‐brain using family wise error correction