Educational games, custom-built around what your students need to learn.

I taught for 6 years, from technical school to MBA programs, and earned a postgraduate degree in Education. That path left me with a habit: asking for the source before believing in a new method, including my own. This section is for people who decide based on evidence. It covers what research supports about games in education, what it does not yet guarantee, and where Nextborn stands. Every claim has a full reference at the end of the page.

Attention in class has a number

PISA 2022, the OECD's assessment of 15-year-old students, measured what every teacher sees from the front of the room. Across OECD countries, on average, 30% of students said they get distracted by digital devices in most math lessons, and 25% said they get distracted by other students' devices. Students who report that second kind of distraction in at least some lessons score 15 points lower in math, after accounting for socioeconomic status (OECD, 2023). UNESCO's report on technology in education adds that the mere presence of a phone is distracting and harms learning, in studies from 14 countries (UNESCO, 2023).

A popular talking point deserves to be set aside here: the idea that students can only pay attention for 10 or 15 minutes. Wilson and Korn (2007) reviewed the studies behind that idea and found little support for it. Bunce, Flens and Neiles (2010) measured attention in class and saw something else: it drifts in short lapses, a minute or less, in cycles that get shorter as the lesson goes on, and each interactive activity reduces those lapses. What decides attention is what happens inside the lesson, far more than how long it lasts.

What holds attention has a name

Phones hold attention for reasons psychology has known for decades. In 1981, Thomas Malone studied computer games to find out what made them fun and identified 3 sources of motivation: challenge, fantasy and curiosity. Challenge works when the outcome is uncertain and the difficulty adjusts. Csikszentmihalyi (1990) described the state of total absorption, flow, which appears when the challenge matches the person's skill. Ryan, Rigby and Przybylski (2006) showed that the appeal of games comes from 3 basic needs: feeling competent, deciding with autonomy and belonging to a group.

This holds up in educational games. Hamari and colleagues (2016) followed 173 students in 2 educational games and found that challenge at the right level was the factor that best predicted perceived learning, both directly and through engagement. The design lesson: the challenge has to grow along with the student's skill.

4 decades of research and 3 different names

Clark Abt coined the term serious games in 1970 for games made with an educational purpose. Malone published his theory in 1981. James Paul Gee argued in 2003 that good games apply learning principles recognized by cognitive science. In 2005, Shaffer, Squire, Halverson and Gee argued that games matter because they place the player inside simulated worlds, and warned that most educational games up to then had been made without any theory of learning behind them. In 2011, Deterding and colleagues gave gamification its academic definition. From 2011 to 2020, a series of meta-analyses measured what all of this produces in the classroom.

The literature separates what the market tends to mix together. Gamification is the use of game elements outside a game: points, badges and leaderboards on top of an activity that already exists (Deterding et al., 2011). Landers (2014) explains the mechanism: the badge acts on a behavior, such as doing more exercises or coming back more often, and it is that behavior that can raise the grade. The content stays outside the badge. A serious game is a complete game designed with a purpose beyond entertainment (Abt, 1970). Here the subject lives inside the mechanics: moving forward requires using the content. Research on the two is separate, and this section keeps them separate.

What the layer delivers and where it stops

The layer of points and badges works, and the numbers say how much. Sailer and Homner (2020) pooled the studies and found a consistent effect of gamification on learning (g = 0.49, across 19 studies with 1,686 participants), on motivation (g = 0.36) and on behavior (g = 0.25). In classroom terms: the average student in the gamified class ends up ahead of about two thirds of the students in the regular class. The first review of the field had already warned that the effect depends heavily on context and on who is using it (Hamari, Koivisto and Sarsa, 2014).

The psychology of motivation explains the limit. Deci, Koestner and Ryan (1999) brought together 128 experiments and showed that expected rewards reduce interest in the activity itself, while positive feedback increases it. A badge for a finished task belongs to the first group. This has been measured in the classroom: Hanus and Fox (2015) followed a 16-week university course with badges, coins and a leaderboard, compared with the same course without those elements, and the gamified group ended the semester with less motivation and less satisfaction. Sailer and colleagues (2017) showed that each element has its own effect: badges and leaderboards feed the sense of competence; story, characters and teammates feed the sense of belonging. And there is the novelty effect, measured in Brazil: Rodrigues and colleagues (2022) followed 756 Brazilian students for 14 weeks and saw the effect of gamification drop after the fourth week and only rise again between the sixth and the tenth.

The badge changes the behavior around the task. The task stays the same.

What changes when the subject becomes the mechanics

As early as 1981, Malone separated extrinsic fantasy, which depends little on the skill used in the game, from intrinsic fantasy, which is tied to using that skill. 30 years later, Habgood and Ainsworth (2011) turned the idea into an experiment. They built 2 versions of the same math game for children aged 7 to 11. In one, division was the game's mechanic; in the other, the same math appeared on screens separate from the action. Children learned more in the version where the subject was inside the mechanic and, when they could choose what to play, spent 7 times as long in it. Students play because they want to play, and they learn along the way.

Meta-analyses of complete games point the same way. Wouters and colleagues (2013), with 77 comparisons and 5,547 participants, found more learning (d = 0.29) and more retention (d = 0.36) with serious games than with conventional instruction, with larger gains when the game is part of the lesson, used over more than one session and played in groups. Clark, Tanner-Smith and Killingsworth (2016), with 57 studies, showed that games outperform non-game conditions (g = 0.33) and that a well-designed version outperforms the basic version of the same game by an effect of the same size. In their words, the result highlights the central role of design, beyond the medium itself. Sailer and Homner (2020) found that including story, characters and a world was one of the 2 factors that most helped change student behavior. The other was combining competition with collaboration.

The oldest and most studied application of this is the business simulation, used in business schools for more than 40 years (Faria et al., 2009). Sitzmann (2011), in a meta-analysis of 65 studies with 6,476 participants, measured gains of 11% in declarative knowledge, 14% in procedural knowledge and 9% in retention for those who learned with simulations, and found that simulations work better when they complement instruction than when they replace it. In Brazil, the Management Laboratory at FEA-USP (the School of Economics, Business and Accounting of the University of São Paulo) has combined business simulation with research for years (Ribeiro et al., 2015). The same field acknowledges its own gap: Anderson and Lawton (2009) showed that students and instructors perceive a great deal of learning in simulations, but that for 25 years few people rigorously measured what students actually learned.

This is what Nextborn builds on. The game comes first, with story, challenge and consequence, and the subject is what makes the student move forward.

What we do with what the science says

3 method decisions come straight from the studies.

The game is part of the lesson, with materials for the teacher to lead it, because that is where it works best (Wouters et al., 2013; Sitzmann, 2011). The challenge of each level is tuned in testing with real students before the project is finalized, because a well-calibrated challenge is what sustains learning (Hamari et al., 2016). Every student decision is recorded, and the way learning is measured is designed together with the institution, because the field has been asking for this for 25 years (Anderson and Lawton, 2009).

In Brazil, the path is open. Brazil's national curriculum, the Base Nacional Comum Curricular (BNCC), calls in general competency 5 for students to "understand, use and create digital information and communication technologies in a critical, meaningful, reflective and ethical way" (our translation) in order to "solve problems and exercise protagonism and authorship" (our translation), and in general competency 2 to "develop and test hypotheses, formulate and solve problems and create solutions" (our translation) (Brasil, 2018). A game where students decide, make mistakes and try again is one of the most direct ways to practice both. Law No. 14,852 of May 3, 2024, known as Brazil's Legal Framework for Games, authorizes the use of electronic games "in school settings, for teaching purposes, in accordance with the Base Nacional Comum Curricular (BNCC)" (our translation) and provides for incentives for research and development of games aimed at education (art. 10, item II; art. 13, § 1, item IV).

The science says games teach, with a consistent effect, when they are well designed and part of the lesson. That is exactly the work Nextborn sets out to do with your institution. If you want to see this with your students, tell us what you need to teach in the form below. The conversation starts there.