There is a stretch of restored forest running through the middle of North Carolina State University’s campus. Twenty years ago it was a degraded urban creek. Between 2002 and 2006, crews re-engineered the floodplain, put meanders back into the stream, pulled out invasive plants, and planted native trees. Today the Rocky Branch Greenway has more than 70% canopy cover, and if you sit in the right spot, you can’t see a single building.
A team of researchers led by Nils Peterson at NC State used that spot to ask a deceptively simple question: how much of nature’s psychological benefit comes from what you see, and how much from what you hear? Their answer, published in Environmental Conservation, is that both matter, and that they stack.
The setup
The design was elegant enough to explain in a sentence. Take 106 students, put each of them in two places (the restored forest, and a “grey space” outside the Talley Student Union with more than 70% built structures in view), and at each place, have them listen to two soundtracks through noise-canceling headphones: ten minutes of birdsong, or ten minutes of traffic.
That’s four conditions per person, and because everyone experienced all four, each participant served as their own control. Order was randomized. The whole circuit took about two hours.
The birdsong was a real field recording from a Florida wildlife refuge: blue-grey gnatcatchers, a Carolina wren, a white-eyed vireo, with woodpeckers, kingfishers, mourning doves and catbirds in the background. The traffic was a real recording from a busy intersection outside Paris. Both played at conversational volume, roughly 50–70 decibels.
The researchers even controlled the weather. Participants could pick any day between February and April 2025, provided it wasn’t raining and the temperature was in the 20–25°C range, where temperature itself doesn’t meaningfully affect stress.
After each ten-minute session, participants rated their stress on a 0–10 scale, filled out a twelve-item measure of positive and negative feelings, and rated how restorative the place felt.
What they found
Every measure moved, and every measure moved in the same direction:
- Stress. Forest-plus-birdsong produced a mean of 3.17 on the eleven-point stress index. Grey space plus traffic produced 6.35, roughly double. The two mixed conditions fell between and were statistically indistinguishable: 4.81 for birdsong in the grey space, 4.97 for traffic in the forest.
- Negative feelings. Same shape. Forest-plus-birdsong scored lowest at 9.44 on a 6–30 scale; grey-plus-traffic highest at 15.43.
- Positive feelings. The pattern runs the other way here, because a higher score is the better one. In the forest with birdsong, people averaged 22.96. In the grey space with traffic, 13.53. That is the widest gap anywhere in the study. Researchers size up a gap like this with a statistic called Cohen’s d, where 0.8 is the usual cutoff for calling a difference large. This one came in at 2.38, roughly three times that.
- Perceived restorativeness. Forest-plus-birdsong scored 36.77 out of 55; grey-plus-traffic 18.37.
The pattern that makes this study interesting isn’t the ranking, though. Anyone could have guessed the ranking. It’s that the two ingredients behaved additively. Statistically, the researchers found strong independent main effects for both soundscape and location on every outcome, and essentially no interaction between them. Birdsong helped in the concrete plaza. Trees helped even with traffic in your ears. Put them together, and you got the sum of both, not a bonus and not a redundancy.
The one place the pattern breaks
Perceived restorativeness was the exception, and it’s worth sitting with. There, location did far more work than sound: the setting accounted for about 34% of the variance, the soundtrack only about 5%. Traffic noise in the forest still beat birdsong in the grey space, the only outcome where that flip happened.
The authors offer a candid explanation: the scale might be the problem. Perceived restorativeness questionnaires repeatedly ask people about “this place” and “here,” and place is a word that pulls the mind toward what the eyes are doing. The instrument may simply be built to measure the visual. That’s a limitation in the measurement, not necessarily in the ears.
Why this matters beyond the study
A common claim about birdwatching is that it can be done anytime, anywhere, by anyone. No gear, no fitness requirement, no skill. This study strongly supports the “anybody” part. Participants needed nothing but ten minutes, a chair, and a sound. No binoculars, no walk, no bird identification, no actual birds.
But it complicates the “anywhere.” The forest mattered independently, and it mattered most for the sense that a place could restore you. If canopy cover is doing real work, then access to it becomes a health equity question. Neighborhoods without trees aren’t just less pleasant; on this evidence, they’re offering their residents a measurably smaller share of a mental health resource. The authors note this risk directly, and it lands harder given that birdwatching in the US already skews older and whiter than the population.
There’s a hopeful counterweight, though, and it’s the whole point of the paper’s title. The forest in this experiment was not old growth. It was a restored urban creek corridor barely two decades past construction, wedged inside a university campus with a road nearby. That’s the good news for anyone trying to build this: the benefits didn’t require wilderness. They required about twenty years, some native trees, and thoughtful placement of a place to sit.
The honest caveats
The participants were 106 college students at one university, average age 22.5, a narrow slice of humanity. Everyone heard the same single birdsong track, so we can’t say whether a marsh chorus differs from a desert one, or whether some bird calls are less pleasant than others. And while volume was held in the 50–70 dB range, it wasn’t precisely measured, which limits how exactly a practitioner could reproduce the effect.
None of that undercuts the core finding. It just means the interesting questions are the next ones. How much birdsong? Which birds? For how long? Does the effect hold for a fifty-year-old in a different climate?
For now, the takeaway is unusually actionable for an academic paper. Ten minutes of birdsong measurably lowered stress even in a concrete plaza. Ten minutes among trees did the same even with traffic playing. And the combination, restored forest with real birds, was better than either alone, in a place that didn’t exist in that form twenty-odd years ago.
Restore the forest, and the birds that come with it, and you’re not only doing conservation. You’re building infrastructure for people’s heads.
