Journal of Economic Behavior & Organization
Volume 241, January 2026, 107358
Research Paper
The cognitive benefit of a window view
Author links open overlay panelXuan Li a, Xiang Zhou b
Show more
Share
Cite
RedirectingGet rights and content
Abstract
This paper examines whether sitting by a window can influence cognitive performance in a high-stakes setting. Leveraging unique administrative data from Chinese college entrance exams with randomized seating assignments, we find that a seat by a window with an outside view significantly enhances cognitive performance, resulting in 8.9 percent of a standard deviation increase in exam scores. Further evidence suggests that this finding aligns with Attention Restoration Theory. This study highlights the value of restorative environments in enhancing cognitive performance.
Introduction
Americans spend about 87 % of their time indoors-at home, work, school, and other enclosed spaces (Klepeis et al., 2001). In these spaces, windows serve as the primary connection to the natural world, providing natural light, views, and a sense of spaciousness. Indeed, access to natural light and outdoor views ranks among the most valued workplace amenities among North American employees and corner offices with desirable views are typically reserved for senior executives (Future Workplace, 2018). Yet, despite their recognized appeal and potential influence on human experience, rigorous empirical evidence of windows’ cognitive benefits remains limited. In this paper, we investigate how sitting by a window impacts cognitive performance in a high-stakes testing environment.
We leverage the Chinese college entrance examination, a high-stakes test that determines college admissions in China. To ensure test integrity, county-level education authorities randomly assign students to test rooms, each accommodating 30 students in a five-column by six-row seating grid. This randomization generates exogenous variation in window proximity: some students are assigned seats next to windows with outside views, while others are placed in interior locations. As a result, students from the same high school class, despite sharing similar academic backgrounds and taking identical exams, encounter different testing environments based on their assigned seating positions.
Using administrative data from over 3700 students who took the college entrance examination in a county in 2019, we first validate the random assignment mechanism. Balance tests confirm that seating arrangements are uncorrelated with student characteristics, including gender, ethnicity, age, and urban/rural status. Having established the validity of randomization, we estimate the effect of window seating on academic performance. Our analysis reveals that students seated next to windows with outside views score 8.9 percent of a standard deviation higher than their peers, while those next to windows with blocked views show no significant performance differences. These results remain robust to the inclusion of individual controls, as well as fixed effects for seat row, high school class, and test room.
A seat-level decomposition analysis strengthens our findings: every seat adjacent to a window with an outside view is associated with improved performance, with effect sizes ranging from 7.4 percent of a standard deviation upward-exceeding the effects of 22 out of 24 non-window seats. This consistent pattern across window seats rules out the possibility that our main results are driven by some outliers. The positive effect persists across both science and liberal arts tracks, with particularly strong effects among male students. Importantly, these cognitive gains translate into meaningful college admission outcomes: students seated by windows with unobstructed views are 2.8 percentage points more likely to be admitted to prestigious first-tier universities.
We further explore the underlying mechanisms and present suggestive evidence aligned with Attention Restoration Theory (Kaplan, 1995, Berto, 2005). Attention Restoration Theory suggests that individuals can concentrate better after spending time in nature or even by simply gazing out a window. According to this theory, students in our context who struggle with maintaining attention can greatly benefit from sitting by a window, especially during high-stakes tests where focusing on solving challenging problems is paramount. The registration process for college entrance exams requires teachers to evaluate students’ personality and behavior in school. Through textual analysis of these evaluations, we identify students who have difficulty maintaining attention, and find that these students are the primary drivers of our main results, which is consistent with Attention Restoration Theory.
Our paper adds to the growing literature on environmental factors influencing cognitive performance and productivity (Graff Zivin and Neidell, 2012, Chang et al., 2016, Park et al., 2020, Adhvaryu et al., 2020, Fu et al., 2021, Somanathan et al., 2021, Adhvaryu et al., 2022, Künn et al., 2023, Dean, 2024), particularly within high-stakes examination contexts (Ebenstein et al., 2016, Bensnes, 2016, Graff Zivin et al., 2020a, Graff Zivin et al., 2020b, Park, 2022). While existing studies primarily examine the effects of temperature and air pollution, we are the first to investigate the impact of a widely present yet largely overlooked factor - access to windows, and thus outside views - on cognitive performance. Compared to other environmental factors, we find a significant cognitive benefit associated with sitting near a window: students seated by windows with outside views score 9.1 % of a standard deviation higher than their peers. In a similar context, namely the Chinese college entrance exam, Graff Zivin et al. (2020b) report that a one-standard-deviation increase in temperature during the exam period (3.6∘F) lowers scores by 5.8 % of a standard deviation, while Graff Zivin et al. (2020a) find that a one-standard-deviation increase in agricultural fires reduces scores by 1.4 % of a standard deviation. Furthermore, using administrative data from Norway, Bensnes (2016) finds that a one-standard-deviation increase in ambient pollen levels leads to a 2.5 % standard deviation decrease in exam scores.
Our study also contributes to ongoing policy discussions about the value of school facilities-a major component of public spending on K-12 education (Jackson and Mackevicius, 2024). Several studies document sizable positive effects of school capital spending on student test scores (Neilson and Zimmerman, 2014, Hong and Zimmer, 2016, Lafortune and Schönholzer, 2022). For instance, Neilson and Zimmerman (2014) find that school construction increases reading scores by 0.15 standard deviations, while Lafortune and Schönholzer (2022) show that students who attend a newly constructed school for four or more years score 0.097 standard deviations higher in math. However, some other research finds limited or negligible impacts (Cellini et al., 2010, Martorell et al., 2016). Understanding which specific types of capital investment yield meaningful educational returns may help reconcile these mixed findings (Biasi et al., 2024). Biasi et al. (2024) demonstrate that investments in essential infrastructure-such as HVAC systems, plumbing, roofs, and furnaces-lead to significant improvements in test scores (ranging from 0.1 to 0.2 standard deviations), whereas spending on athletic facilities has no comparable effect. Similarly, Park et al. (2020) show that although hot school years can impair learning, air conditioning can mitigate these negative effects. More closely related to our study, Li and Sullivan (2016) find that exposure to green landscapes through classroom windows helps attention restoration and stress reduction. In the context of Virginia, Bullock (2007) finds that the condition of school facilities-including windows in instructional areas-is positively associated with student achievement. In U.S. schools, windows are among the most frequently cited infrastructure elements in poor condition and have been identified as a priority for replacement (Alexander and Lewis, 2014). Our study provides direct evidence that windows, a specific component of school infrastructure, can substantially influence student academic assessment in a high-stakes exam.
Section snippets
Setting: The national college entrance examination
The national college entrance examination in China is a high-stakes assessment that uniquely determines college admissions through a single annual test. Often characterized as a “life-changing event,” this examination profoundly shapes both educational pathways and subsequent labor market outcomes (Jia and Li, 2021).
The national college entrance examination in China follows a structured assessment framework consisting of four components, with a total score of 750 points: Chinese language (150
Identification
To rigorously quantify the impact of sitting by a window on cognitive performance, we estimate the following specifications:where StdScorei represents the standardized total score of student i in the national college entrance exam, within the specific academic track (such as science or liberal arts). The variable Col 1 is a dummy variable indicating whether a student was seated in Column 1 (next to a window but with a blocked outside view). Col 5 is a dummy
Conclusions
In this paper, we use administrative data from a county that randomly assigned students taking the 2019 college entrance examination to test seats to explore the cognitive benefits of sitting by the window. Our analysis reveals that sitting by a window with an unobstructed outside view significantly boosts cognitive performance, while sitting by a window with a blocked view has no impact. The positive effect of sitting by a window holds across both science and liberal arts tracks and is
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We are grateful to the editor and two anonymous referees for their constructive comments. We also thank Raymond Fisman, Joshua Goodman, Kevin Lang, and Daniele Paserman for their valuable feedback and guidance. Zhou acknowledges the financial support received from the Department of Education of Hunan Province (Project No. 24C0054). Author names are listed in alphabetical order. Li and Zhou contributed equally to this work and are recognized as co-first authors. All remaining errors are our own.
Last edited by @suen 2025-12-22T00:43:36Z
