Teaching programming trains executive functions in low socioeconomic status preschoolers

HERMIDA M.J.; GOLDIN, A.; PEREZ SANTANGELO, AGUSTÍN; LIPINA S. J.; SCHAPACHNIK, FP

Belem

Congreso; 3rd Congress of the Federation of Latin American and Caribbean Neuroscience Societies FALAN; 2022

FALAN

Introduction Children’s executive functions are crucial to success in school; they develop rapidly in preschool years and can be trained through different activities. Few studies to date have analyzed whether teaching programming (recently incorporated to most curricula worldwide) also trains executive functions. ObjectivesHere, we addressed whether teaching programming can improve executive functions in low socioeconomic status preschoolers from the province of Buenos Aires, Argentina Methods A cluster randomized controlled trial (n = 45) was carried out in one low socioeconomic status school (study authorized by CEMIC Ethical Committee protocol N° 1060). Five-year-old classrooms were assigned to an experimental group (who received programming instruction with the most widely used tool: ScratchJr) or a control group (who worked on art activities). Activities in both groups lasted 12 one-hour sessions, included tablets and were carried out by the regular classroom teachers. Before and after these activities, all children were administered a wide battery of computerized executive functions tests; Also, their parents were administered a socioeconomic level survey to check the socioeconomic status of their homes. We used generalized linear mixed models to analyze children’s responses to each trial of all cognitive tests.Results Teaching programming with ScratchJr increased achievements in inhibitory control, working memory and non-verbal intelligence tests. We observed ~90% higher OR in inhibitory control [OR median: 1.92, p(rOR > 1) = 0.991]; ~75% higher OR in logic reasoning [OR median: 1.74, p(rOR > 1) = 0.935]; ~65% higher OR in working memory [OR median: 1.65, p(rOR > 1) = 0.950]; and ~29% higher OR in attention [OR median: 1.29, p(rOR > 1) = 0.823]. No relevant differences were found in planning.ConclusionsOur results showed that teaching programming to 5-years-olds, besides its intrinsic value, would train cognitive processes crucial for success in school in the most vulnerable children.