Growth mindset, the belief that one's abilities can improve through cognitive effort, is an important psychological construct with broad implications for enabling children to reach their highest potential. However, surprisingly little is known about malleability of growth mindset in response to cognitive interventions in children and its neurobiological underpinnings. Here we address critical gaps in our knowledge by investigating behavioral and brain changes in growth mindset associated with a four-week training program designed to enhance foundational, academically relevant, cognitive skills in 7-10-year-old children. Cognitive training significantly enhanced children's growth mindset. Cross-lagged panel analysis of longitudinal pre- and post-training data revealed that growth mindset prior to training predicted cognitive abilities after training, providing support for the positive role of growth mindset in fostering academic achievement. We then examined training-induced changes in brain response and connectivity associated with problem solving in relation to changes in growth mindset. Children's gains in growth mindset were associated with increased neural response and functional connectivity of the dorsal anterior cingulate cortex, striatum, and hippocampus, brain regions crucial for cognitive control, motivation, and memory. Plasticity of cortico-striatal circuitry emerged as the strongest predictor of growth mindset gains. Taken together, our study demonstrates that children's growth mindset can be enhanced by cognitive training, and elucidates the potential neurobiological mechanisms underlying its malleability. Findings provide important insights into effective interventions that simultaneously promote growth mindset and learning during the early stages of cognitive development.

Gaining insight into course choices holds significant value for universities, especially those who aim for flexibility in their programs and wish to adapt quickly to changing demands of the job market. However, little emphasis has been put on utilizing the large amount of educational data to understand these course choices. Here, we use network analysis of the course selection of all students who enrolled in an undergraduate program in engineering, psychology, business or computer science at a Nordic university over a five year period. With these methods, we have explored student choices to identify their distinct fields of interest. This was done by applying community detection to a network of courses, where two courses were connected if a student had taken both. We compared our community detection results to actual major specializations within the computer science department and found strong similarities. To compliment this analysis, we also used directed networks to identify the ”typical” student, by looking at students’ general course choices by semester. We found that course choices diversify as programs progress, meaning that attempting to understand course choices by identifying a ”typical” student gives less insight than understanding what characterizes course choice diversity. Analysis with our proposed methodology can be used to offer more tailored education, which in turn allows students to follow their interests and adapt to the ever-changing career market.

Namsval nemenda felur i ser verðmaetar upplýsingar fyrir haskola, serstaklega þa sem miða að þvi að gera namið sitt sveigjanlegt og vilja aðlagast hratt að breytilegum vinnumarkaði. Það hefur hinsvegar ekki verið logð mikil ahersla a að nýta þau stafraenu gogn sem skolar safna til að skilja namsval nemenda enn frekar. Við notum gagnanam með aherslu a menntun (e. Educational data mining, EDM) og einbeitum okkur serstaklega að netagreiningu og hvernig samfelog innan neta myndast. Gagnasafn okkar naði yfir alla þa nemendur (N = 11207) sem innrituðu sig i nam a siðustu fimm arum við Haskolann i Reykjavik (HR). Með þessum aðferðum viljum við skoða namsval nemenda og bera kennsl a ahugasvið þeirra. Það var gert með þvi að beita Louvain samfelagsgreiningu (e. community detection) a net namskeiða. Netið var þannig sett upp að tengingar a milli noða taknuðu fjolda nemenda sem hofðu lokið baðum namskeiðum. Við byrjuðum a að fjarlaegja mest tengdu noðurnar ur netinu þar sem þaer taknuðu yfirleitt skyldufog frekar en frjalst val, en við vildum skoða frjalst val til að skilja ahugasvið nemenda betur. Niðurstoðurnar voru svo bornar saman við þaer aherslulinur sem eru i boði innan Tolvunarfraeðideildar HR. I ljos kom að samfelogin sem við fundum endurspegluðu helst þaer aherslulinur sem flestir nemendur taka, sem ýtir undir að tekist hafi að bera kennsl a raunveruleg ahugasvið. Til samanburðar við ahugasviðin þa notuðum við einnig attuð net til að skoða “meðalnemandann” við HR ut fra algengustu samsetningu afanga fyrir hverja onn. Meðalnemandinn var mun skýrari a fyrri en seinni onnum namsins, þar sem skipulag namsleiða verður opnara með timanum. Ut fra þessum niðurstoðum vaeri mogulega haegt að innleiða nýjar aherslulinur og skoða viðbrogð nemenda til að sannreyna hagnýtingu þessarar aðferðar. Vonandi baetist þessi aðferð i hop annarra EDM aðferða, sem saman verða nýttar til að bjoða upp a serhaefðara og sveigjanlegra nam sem haegt er að sniða betur að þorfum nemenda. Þetta myndi hjalpa nemendum að fylgja sinu ahugasviði og geta brugðist hraðar við breytingum a krofum vinnumarkaðsins.; Gaining insight into student course choices can hold enormous value for universities, especially those who aim for flexibility in their programs and those who wish to adapt quickly to changing demands of the job market. However, little emphasis has been put on utilizing the large amount of administrative educational data to understand these student choices. Here, we use educational data mining, focusing on network analysis with community detection, on data from all students who have enrolled at Reykjavik University in the past five years (N = 11207). With these methods, we aimed to explore student choices and identify their distinct fields of interest. This was done by applying the Louvain community detection algorithm to a network of courses, where two courses were connected if a student had taken both. We first removed hubs from the network to give higher weight to students’ direct choices rather than mandatory courses. We compared our community detection…