The vocabulary learning challenge - How can we improve learning efficiency?

In this project we developed a digital morphology based language intervention for primary school children. We tested the effects of the intervention in a randomized controlled trial with more than 700 second grade students, including almost 200 children with language minority backgrounds. Vocabulary plays a key role in language development, and a limited vocabulary has both immediate and long-reaching negative effects on students' learning to read, their comprehension of disciplinary content, and on their overall success in school and subsequent employment. Thus, a vocabulary intervention leading to more efficient and transferable learning in all children, but especially in children at-risk for language impairment, will have a direct societal impact. Previous studies suggest that knowledge of the smallest meaning-bearing linguistic elements in language, morphemes, has a positive effect on vocabulary development as well as reading and spelling abilities. Moreover, recent studies have shown that systematic variation in input can facilitate generalization of linguistic patterns. Thus, the goal of the project was to develop and test a digital vocabulary intervention based on the following two principles: 1) strengthen the students? knowledge of common Norwegian morphemes (e.g. ?re? as in reuse and ?ist- in ?guitarist?) and 2) use variation in input systematically to facilitate generalization of the linguistic knowledge: If the student works with 24 instances of the morpheme re-, this help the student infer that re- is a part of a word that has its own meaning, and additionally, to understand what this morpheme does to a word?s meaning. The app contains 40 sessions which focus on an affixes (e.g. ?re-?, ?-ist?) or compounding patterns (e.g. ?racehorse? has a different meaning than ?horserace?). The app does not focus on root words (e.g. ?create? in ?recreate?). A pilot version of a morphology-based app was developed in 2016, in collaboration with Engagelab University of Oslo. The target morphemes were selected after assessment of the morphological knowledge of approximately 100 4th graders. The app was piloted in 4 schools with 150 second graders. The complete app (8 week program) was finalized during fall 2017. The randomized-controlled trial was preregistered in the Open Science Framework in January 2018. Pre-tests were carried out from January to March 2018, the intervention from March-May, and post-tests from April to June. 717 students participated in the study, and were individually randomized to the intervention group or active control group. The control group used a research-based math app for the same number of hours and at the same times as the intervention group. Effects of the math intervention were also measured. A delayed post test to examine long-term effects of the intervention was carried out from October to December 2018. The main journal article from the intervention (Torkildsen et al., submitted) was submitted to Journal of Educational Psychology during summer 2020, and will be resubmitted in February 2021 after a minor revision. The results show significant and durable effects of the intervention, for trained and untrained words containing trained morphemes. For example, by working with words like ?reuse? and ?recreate?, students learned to generalize their knowledge about the affix and use it to explain, understand, read and spell other (untrained) words such as ?revive? or ?reproduce?. The intervention effects were larges for word knowledge (the ability to explain and understand words), but also significant for reading fluency and spelling for trained and untrained words. However, there were no effects on general vocabulary, probably due to the focus on derivations and compounding processes, but not on root words. Most of the intervention effects were sustained to the follow-up in third grade. The effects on word knowledge were the most durable. A separate study of intervention effects for the 188 students with language minority backgrounds was pre-registered in 2018. A paper from this study (Bratlie et al, submitted) was submitted during fall 2020 and is currently under revision. Students with language minority backgrounds had considerably lower scores on vocabulary and morphology tests than students with language majority backgrounds (i.e. students who spoke only Norwegian at home, n = 521). Group differences were moderate for spelling and small for reading fluency. There were no significant group differences for verbal short-term memory. The effect of the intervention was comparable for students with language minority and language majority backgrounds. The effect for students with language minority backgrounds were not associated with their use of Norwegian at home or parents? education. These findings show that students with language minority backgrounds benefitted from the intervention and made as much progress as majority language students.

I prosjektet har vi utviklet en app, Kaptein Morf og stjernestøvet, som består av 40 øvingsøkter. Fra 2020 har appen vært gratis tilgjengelig for allmennheten, og er tatt i bruk i mange ulike kommuner i Norge. Appen er presentert på flere åpne seminarer. Prosjektet har ført til økt samarbeid med University of Oxford, Harvard University og Gøteborgs universitet. Prosjektet har også ført til et nytt samarbeid med Malin Wass ved Luleå tekniske universitet. I 2020 fikk hun midler fra svenske Vetenskapsrådet for å utvikle en svensk versjon av Kaptein Morf (2021-2025). Prosjektleder for Vocabulary Learning Challenge (Janne von Koss Torkildsen) er samarbeidspartner. Et annet forskningssamarbeid er startet med Andreas Falck, Lunds Universitet, Sverige og Institut Nicod, Frankrike. Han har søker NFR Forskerprosjekt for fornyelse i 2021 for å utvikle en flerspiller-versjon av Kaptein Morf som kan brukes for å undersøke hvordan felles oppmerksomhet og roller i samspillet påvirker språklæring.

The current application regards a randomized controlled trial (RCT) of a vocabulary intervention program targeting typically developing elementary school children as well as children at risk for language and reading impairment. A main aim of the project is to increase the effectiveness and transfer effects of a vocabulary intervention program by incorporating new evidence from morphological training studies and experimental work in cognitive science. A crucial attribute of the proposed program regards the manipulation ot input variability in order to faciliate transfer to untrained items. Vocabulary plays a key role in language development, and a range of studies show that a limited vocabulary has both immediate and long-reaching negative effects on students' learning to read, their comprehension of disciplinary content, as well as on their overall success both in school and subsequent employment. Thus, a vocabulary intervention leading to more efficient and transferable learning in all children, but especially children at-risk for language and reading impairment, will have a direct societal impact. The proposed project has also introduce elementary school teachers to new evidence-based approaches for supporting children's language and reading comprehension development.