SEL Works Best When the Environment Supports It: Two Evaluations of an Interactive Game

This blog summarizes Testing the efficacy of the ZooU social emotional learning program published in the Social and Emotional Learning: Research, Practice, and Policy. It is published here in honor of its coauthor, Jeffrey A. Rosen, Ph.D., who passed away in 2026. Read Dr. Rosen’s obituary.

It is well established that school-based social and emotional learning (SEL) programs improve students’ skills, attitudes, behavior, academic achievement, peer relationships, and perceptions of school climate and safety (Cipriano et al., 2023; Durlak et al., 2011; Greenberg, 2023). However, many school-based programs require substantial teacher training, ongoing monitoring of implementation quality, and renewed support as staff turnover. They also typically deliver the same instruction to groups of students rather than adapting practice to individual needs.

Individualized computer games represent a promising but relatively understudied approach to SEL instruction. They can provide automated, repeatable, and individualized practice with limited teacher training; reduce variation in program delivery; and offer a potentially low-cost, sustainable complement to classroom-based instruction.

We evaluated ZooU because it was designed to address these opportunities. ZooU is an individualized digital SEL game for elementary students that presents realistic social situations in which students make choices, receive feedback, and practice impulse control, communication, cooperation, empathy, social initiation, and emotion regulation.

Across two randomized trials, schools consistently made time for gameplay, students generally navigated and completed scenes with minimal support, and delivery costs were low. Average effects were mixed, but outcomes varied systematically with classroom climate and the amount of content students completed. These findings suggest that individualized games can be a feasible component of school-based SEL while also highlighting the conditions that may shape their effectiveness.

What Our Study Found

We conducted two cluster randomized controlled trials involving more than 1,100 students in Grades 2 through 4. Classrooms were assigned to use either ZooU or Reflex, an individualized math-fact game. This active comparison helped distinguish ZooU’s SEL content from features shared by both conditions, such as computer use, individualized gameplay, and time during the school day.

The first cohort participated immediately before COVID-related school closures and received about 300 minutes of gameplay. The second participated after schools reopened and received about 600 minutes. Because the cohorts differed in geography, dosage, and historical context, we treated them as independent evaluations.

Implementation results supported the feasibility of using an individualized SEL game during the school day. Average gameplay in each cohort closely approximated the intended targets, indicating strong school compliance and buy-in. Scene completion was also reasonably high and increased when students had more time, suggesting that most students could navigate the game with minimal assistance and remain engaged with its instructional content. Delivery costs were low relative to approaches requiring extensive training and ongoing teacher-led implementation.

Effects on student outcomes were more nuanced. In the first cohort, ZooU produced small to moderate improvements on several performance-based measures of SEL knowledge and teacher-rated behaviors. In the second, we did not observe the same positive average effects across all students even with more gameplay. Across both cohorts, however, conditional effects involving classroom climate and program completion were consistent enough to warrant closer attention.

These findings move the discussion beyond a simple question of whether the game worked. A more useful question is: Under what conditions can individualized computer games make a meaningful contribution to school-based SEL?

Classroom Climate Shapes the Conditions for Learning

ZooU’s effects varied according to students’ collective perceptions of classroom climate, including whether teachers cared and listened, students felt safe and liked the class, and rules were fair. The two cohorts differed in their overall climate: students in the first cohort, assessed before COVID-related school closures, generally reported more supportive classrooms, whereas students in the second, assessed after schools reopened, reported less supportive classrooms.

Within those different contexts, a clearer pattern emerged. In the generally high-climate first cohort, ZooU’s benefits were often strongest in classrooms that students perceived as relatively less supportive. In the generally low-climate second cohort, outcomes were more favorable in classrooms perceived as relatively more supportive and poorer in the least supportive classrooms. One possible implication is a “sweet spot”: Individualized games may add value where classroom supports leave room for improvement, but severely disrupted or unsupportive environments may limit students’ ability to engage with and apply what they learn.

These results require caution because classroom climate was not randomly assigned and may reflect other differences, such as classroom management, attendance, school organization, or post-pandemic disruption. Even so, they align with broader evidence that SEL is more likely to succeed in environments with supportive relationships, predictable routines, fair expectations, safety, and authentic opportunities to practice (La Salle-Finley et al., 2024; Wong et al., 2021). A digital game can introduce and reinforce skills, but it cannot establish those conditions by itself.

Engagement Is More Than Time Logged In

Students differed substantially in how many ZooU scenes they completed, even when given similar amounts of time. This distinction matters. A student may be logged in without progressing through the instructional content. Schools should monitor completion, not only minutes, and identify students who need technical help, additional time, or support engaging productively.

Our dosage analyses were not experimental, so completion differences cannot be interpreted causally. Still, students who completed more scenes generally demonstrated stronger performance-based SEL knowledge. In the second cohort, greater completion was also associated with more favorable teacher ratings of self-control, bullying, and hyperactivity. Together, these patterns are consistent with the idea that gameplay can provide both instruction and repeated practice. Students encounter SEL concepts, apply them across varied social situations, receive immediate feedback, and try again—features that may help strengthen both understanding and skill use.

The alignment between performance-based outcomes and teacher ratings also suggests that learning within the game may extend beyond the digital environment to behaviors teachers observe in the classroom. This potential combination of individualized instruction, active practice, and transfer to everyday behavior is an important reason digital games may be useful to schools as a complement to broader SEL supports.

Four Considerations for Using SEL Games in Schools

Schools can strengthen the use of individualized SEL games by attending to four practical decisions:

Ensure schedules permit consistent gameplay. Protect regular time during the school day so students receive an adequate opportunity to engage with the program.
Monitor progress, not only minutes. Track scene completion and provide additional time, technical assistance, or instructional support to students making slower progress.
Connect gameplay to classroom life. Reinforce and extend the skills students practice in the game during group work, conflict resolution, academic persistence, and peer interactions.
Attend to the broader classroom climate. Pair gameplay with trusting relationships, predictable routines, fair expectations, safety, and opportunities for student voice so students can apply new skills.

Moving Beyond Program Adoption

Digital games can offer a feasible way to reinforce SEL by giving students repeated, low-stakes practice with decision-making, empathy, communication, and emotion regulation. Teachers can then connect that practice to classroom discussions, peer interactions, and everyday problem-solving.

But technology should never be separated from the people and settings in which it is used. The promise of SEL lies not only in selecting a strong program, but also in creating environments where students experience safety, belonging, caring relationships, clear expectations, and frequent opportunities to practice the skills the program is designed to teach.

The views in this blog are those of the author and do not necessarily reflect the views of CASEL.

Related Posts

Write for Us

Are you interested in writing for CASEL’s blog, Constellations? Learn more about what we’re looking for and how to pitch your idea!

Learn With Us!

With Belonging & Engagement in the Classroom, you can provide your team with the foundational SEL knowledge to create a learning environment where all students thrive academically and personally. This comprehensive, self-paced virtual training series can be taken individually or as a group PD series. Group rates available!

CASEL Websites