A thought-provoking working paper suggests that installing air filters into classrooms could help students perform better by a significant margin.

In October 2015, workers at a gas well in Aliso Canyon’s underground storage facility in Los Angeles, California, reported a massive gas leak. It was an environmental disaster, comparable to the Deepwater Horizon oil spill. Local communities felt the impact.

Residents reported headaches, nausea, and severe nosebleeds. Children were particularly affected — dozens of children went to school nurses every day with severe nosebleeds. Faced with public and political pressure, the company responsible for the leak installed air filter in every affected classroom. But here’s the thing: they did this three months after the leak was first reported.

Thankfully, with methane being lighter than air and the leak plugged after a couple of months, the level of pollutants had fallen to pre-leak levels by the time the filters were installed. In other words, the filters didn’t help clear up leaked methane because the methane was already gone. They did, however, clear the regular levels of pollution — and that’s when something weird started happening.

Test scores were going up, by a lot: 0.20 standard deviations for math and 0.18 standard deviations for English. This is the equivalent of cutting down class sizes by a third (or, similarly, adding a third more teachers) for the cost of an air filter (around $700).

The changes appear to be sustained over the next year, and cannot be readily explained by any other factor (there was no reform or substantial change). In addition, while Los Angeles has some relatively high pollution levels, they were similar to those of other schools — the effect only happened where the air filters were installed. There’s also some underlying science which could explain why this happens.

Study after study has shown that pollution, even pollution considered within ‘normal’ levels, can affect the cognitive abilities of both children and adults. A recent working paper found that pollution “harms chess players’ performance in cognitive tasks” — and for children, who are generally more sensitive to the effects of pollution, the effects may be even stronger.

Of course, we shouldn’t look too much into one single study (which was not peer-reviewed). However, the effects are so impressive and the costs of installing air filters would be so small that we would really love to see a few schools and cities experiment with this. This opens up some very interesting research avenues, and potentially, offers some very concrete in helping students’ cognitive abilities with low costs.