Learning While Playing: A Randomized Trial of Serious Games as a Tool for Word Mastery

Objectives: The aim of this study was to determine whether college students build their vocabularies by playing a serious game that incorporates principles of learning and memory. Design: This study used a single-site, prospective, randomized trial with 2 branches: immediate and delayed treatment. Outcome measures were obtained after a 1-month intervention and, for the immediate group only, after a 1-month retention interval. Setting: College students from the University of Iowa were recruited via mass e-mail. The intervention took place in the participants' homes; the testing took place in a university laboratory. Participants: Forty undergraduates (32 women, 8 men) who planned to take the Graduate Record Examinations (GRE; Educational Testing Service, 2018) within the year following the treatment participated in the study. Participants were allocated to 1 of the 2 treatment branches via biased coin randomization. Intervention: The treatment consisted of vocabulary training on Vocabulary.com for a minimum of 20 min (dose) 4 times per week (frequency) for 4 weeks (treatment duration), for a total cumulative treatment intensity of 320 min. Outcome Measures: The prespecified primary outcome measures of word learning were the number of words mastered during Vocabulary.com play and the percentage of mastered words accurately defined. The depth of information included in the definitions was a secondary outcome. The prespecified primary outcome measure of functional impact was change in the verbal GRE practice test scores from pre- to posttreatment. The examiners scoring the definitions were blind to treatment assignment. All other outcomes were measured automatically by the relevant computer program so blinding was moot. Results: The participants averaged 5.33 hr of play during the treatment interval and mastered 124 words. Amount of play and mastery were highly correlated. Accuracy of definitions was 55% or 59% (depending on treatment branch) after treatment and 55% 1 month later, an insignificant decline. The game itself did not result in GRE gains, but participants who mastered more words per minute of play had higher GRE scores than other participants. Conclusion: Vocabulary.com, an evidence-based game, showed promise as a way to build vocabulary knowledge, but at the intensity of treatment provided, it did not result in gains in high-stakes test performance.