Promoting sustained effortful mental activity in Indian primary schools markedly improved student cognitive and educational outcomes across a wide range of subjects.
Work in education has traditionally focused on the knowledge and skills imparted through schooling, such as literacy and mathematics. However, the impact of education likely runs deeper: schooling may also change our cognitive capacity at some fundamental level. Our research (Brown, Kaur, Kingdon, and Schofield 2025) explores one such capacity: cognitive endurance, or our ability to sustain mental effort over time. Psychologists have long recognised the importance of this capacity: productive activity often involves sustaining mental effort, for example, over many minutes during a school test, hours during a work shift, or everyday decisions and activities such as voting. We test whether cognitive endurance is malleable, and whether it can be shaped by schooling. This would imply that education may enhance general cognitive capabilities in ways that benefit students across a wide variety of situations throughout their lives.
In our research, we first document that, globally and in the US, lower income individuals exhibit cognitive fatigue more quickly than those with higher income across field settings. They also attend schools that offer fewer opportunities to practice thinking for continuous stretches. We then carry out a field experiment with low-income primary schools in India to test how increasing the amount of time students spend engaged in periods of effortful thought impacts their cognitive and educational outcomes. We find that spending time engaged in cognitively taxing activities–whether academic or non-academic in nature–improves students’ attention spans across a wide range of tasks, while substantially improving their school performance across a wide range of subjects.
Exploring patterns of cognitive fatigue and effortful mental activity on school tests
When individuals engage in a mentally demanding task, their performance tends to decline over time. The effects of this cognitive fatigue have been documented across many settings, from paramedics at work to students taking academic tests (Brachet et al. 2012, Balart et al. 2018).
To lay the foundation for our study, we first explore response trends across two prominent global academic achievement tests: PISA (Program for International Student Assessment), for 15-year-olds across 60 countries, and TIMSS (Trends in International Mathematics and Science Study), for fourth graders across 50 countries. We examine how likely students are to get a randomly placed question correct when it appears earlier in the test (when they are still fresh) versus later in the test (when they may experience cognitive fatigue). Across academic subjects and geographic samples, we find a pattern consistent with prior studies: students are more likely to get a question wrong if they encounter it later in the test (Figure 1). However, while all students experience a decline in performance over time, we find that the decline is considerably more severe for students from more disadvantaged backgrounds. Children in lower-income countries show three times the rate of decline as those in richer countries in TIMSS; similarly, in the US, Black and Hispanic students show 72% steeper decline over time than white students (Figure 1). Moreover, these differences in performance decline align with variations in teaching methods across schools. Students from more privileged backgrounds typically spend more of their school day engaged in independent mental effort, working through challenging material on their own rather than passively receiving instruction.
Figure 1: Performance declines in achievement tests
These correlational patterns motivate our research which aims to improve cognitive endurance by increasing the amount of time students spend on cognitively challenging activities on their own while at school.
A randomised intervention to increase sustained effortful mental activity
Between 2018 and 2019, we carried out a randomised experiment with 1,636 students in grades 1–5 across six private primary schools serving students of a low-income background in Lucknow, India. In this setting, time spent on focused cognitive activity is limited. It is common for teachers to use rote memorisation and recitation during the school day, in crowded classrooms with frequent interruptions. Students typically spend little time doing homework or other cognitively effortful tasks outside of school. We tested an in-school intervention to increase the amount of time students spend on effortful thinking for sustained periods by having students solve cognitively challenging problems on their own for about 20-minute sessions during the school day. Students were assigned to one of three study arms:
- Math treatment: Academic cognitive practice. Students solved cognitively challenging math problems. This arm mimics focused cognitive practice in the context of academics, an element of ‘good’ schooling.
- Games treatment: Non-academic cognitive practice. Students played cognitively demanding games that are free of academic content, such as mazes and tangrams. This arm provides a pure test of our hypothesis; that practicing any intellectually demanding task should improve cognitive endurance, regardless of whether it is academic in nature.
- Control: Study hall. Students attend a status quo study hall period with limited cognitive practice.
The content for each of the two treatment arms was delivered over inexpensive Android tablets. The Math Treatment problems were delivered through Pixatel’s Imagine Math software. The Games Treatment used simple non-animated games available on the Google Play store.
Boosting cognitive endurance and educational outcomes
We tested student performance on listening comprehension, IQ tests, and math problems, described below:
- Listening comprehension: Using headphones, each student listens to a pre-recorded set of short, simple stories. After each story, the student is asked a series of simple factual questions about its content, for example, “What colour was the dolphin?”
- Raven’s progressive matrices (IQ test): This is a nonverbal multiple choice test of reasoning in which the participant is asked to identify the element that completes a pattern in a figure (Raven 1936, 2000). This test is viewed as capturing fluid intelligence and is commonly used as an IQ test.
- Math problems: A standard test of math problems, including a mix of remedial and more grade-appropriate questions. These do not correspond to the specific content or format of the questions in the Math sub-treatment software.
Tests ranged from 12-30 minutes and difficulty was adjusted to grade level. For all tests, we: 1) minimised the scope for treated students to have directly learned content that would help them perform better on the tests, 2) randomised the order of each question in the test, and 3) ensured students had sufficient time to finish the tests without time pressure.
Both treatment approaches (solving math problems or playing non-academic games) markedly improved cognitive endurance: students showed 22% less decline in performance over time when engaged in each of these intellectual activities (Figure 2). These effects persisted 3-5 months later following a summer break.
Figure 2: Performance declines on experimental tests by treatment group
We also found that treated students exhibited increased attentiveness in the classroom and scored higher on psychological measures of sustained attention.
Moreover, each treatment improved students’ school performance by 0.09 standard deviations for end of year grades across a range of subjects: Hindi, English, and Math. This is notable because the intervention content did not include any Hindi or English practice. The fact that each treatment improved grades in these unrelated subjects further supports the view that the interventions changed a fundamental cognitive skill.
These results imply that experiencing (good) schooling which allows students to practice exerting attention will improve their mental stamina. We test for this among 5,300 students across 66 schools in Pakistan, where small changes in students’ date of birth affect which year a child enrols in kindergarten. Leveraging a regression discontinuity design, we find that an additional year of schooling similarly improves cognitive endurance, but only in higher-quality schools with pedagogy which encourages sustained focus by the students.
Policy implications for designing school curricula
Our results offer evidence that the experience of effortful thinking itself, even when devoid of any subject content, improves general cognitive capacity. In doing so, it consequently boosts learning across a wide range of unrelated subjects in school. These findings suggest that disparities in the quality of schooling may further disadvantage poor children by hampering the development of a core mental capacity. We identify a policy lever that may be useful in improving cognitive endurance among lower-income children: incorporating opportunities for them to engage in effortful thinking for sustained periods of time at school or at home. More broadly, our study sheds light on the potential pathways through which schooling may shape human capacities beyond its effects on academic skills. Additional work linking specific elements of schooling to core cognitive capacities can further our understanding of why education has such broad and persistent benefits.
References
Balart, P, M Oosterveen and D Webbink (2018), “Test scores, noncognitive skills and economic growth,” Economics of Education Review, 63: 134–153.
Brachet, T, G David and A M Drechsler (2012), “The effect of shift structure on performance,” American Economic Journal: Applied Economics, 4: 219–246.
Brown, C, S Kaur, G Kingdon and H Schofield (2025), “Cognitive endurance as human capital,” The Quarterly Journal of Economics, 140(2): 943–1002.
Raven, J (1936), “Mental tests used in genetic studies: The performances of related individuals in tests mainly educative and mainly reproductive,” Unpublished manuscript.
Raven, J (2000), “The Raven’s Progressive Matrices: Change and stability over culture and time,” Cognitive Psychology, 41: 1–48.