← Studies Suggest 🧠 Psychology

Everyone Seeks Quiet to Think Clearly. Five Experiments Found That Coffee-Shop-Level Background Noise Makes You More Creative Than Silence.

Five experiments totaling 330 participants found that moderate ambient noise at 70 decibels — roughly the hum of a busy café — enhanced creative performance over both quiet and loud environments. The mechanism: noise-induced processing difficulty forced the brain into abstract thinking, which is exactly the cognitive mode creativity requires.

By Daniel Voss, Psychology & Cognitive Science · July 27, 2026

Listen to this article
Loading...
A warm busy café interior seen through a rain-dotted window, golden ambient light spilling across wooden tables with scattered notebooks and coffee cups

📋 The Study

Title
Is Noise Always Bad? Exploring the Effects of Ambient Noise on Creative Cognition
Authors
Mehta, Zhu & Cheema, 2012
Institution
University of Illinois at Urbana-Champaign; University of British Columbia; University of Virginia
Journal
Journal of Consumer Research, 39(4), 784–799
DOI
10.1086/665048
Sample
n≈330 across five experiments (Exp 1: n=65; Exp 2: n=60; Exp 3: n=95; Exp 4: n=42; Exp 5: n=68)
Method
Five between-subjects laboratory experiments using ambient noise at 50 dB (low), 70 dB (moderate), and 85 dB (high); plus one natural-setting field study
Key Finding
Moderate ambient noise (70 dB) enhanced creative performance over both quiet (50 dB) and loud (85 dB) conditions; the effect was mediated by processing disfluency and abstract construal
Effect Size
Exp 1 RAT: M=5.80 (moderate) vs. 4.29 (low), F(3,61)=3.21, p<.05; Exp 3 creativity: F(1,93)=14.48, p<.001; disfluency: F(1,93)=20.26, p<.001
Counterintuition
⚡⚡⚡ 3/5
Replication
Partially replicated. Awada et al. (2022, Scientific Reports) found low white noise (45 dB) boosted creativity over quiet and loud conditions, supporting the inverted-U pattern with shifted optimal levels. Multiple citing studies confirm moderate noise benefits for divergent thinking; no contradictory meta-analysis exists.

The Quiet Myth

When people sit down to think, their first instinct is to eliminate noise. Close the door. Put in earplugs. Find the quietest room in the building. The assumption runs so deep it barely registers as an assumption: creative work requires silence, and anything that breaks that silence is a distraction to be neutralized.

Libraries enforce it with signs. Offices build soundproof pods for it. Noise-canceling headphones became a $5 billion industry partly on the promise that blocking sound would unlock focus. The logic feels airtight. Noise intrudes on thinking, thinking requires concentration, and concentration collapses when interrupted. Except that creativity is not concentration. And the research on what actually helps people think original thoughts tells a story that runs directly against the instinct to seek silence.

Five Experiments in the Background Hum

In 2012, Ravi Mehta at the University of Illinois and colleagues Rui Zhu and Amar Cheema set up a series of experiments that manipulated the one environmental variable most people take for granted: ambient sound. They didn't pipe in white noise or pink noise, the artificial signals most prior research had used. They recorded the ambient sounds of a roadside restaurant and played them through speakers at three calibrated levels: 50 decibels (roughly the quiet hum of a residential street), 70 decibels (a busy café or a passenger car on a highway), and 85 decibels (the roar of a major traffic artery).

In the first experiment, 65 participants worked on the Remote Associates Test, a standard creativity measure that gives you three words and asks for the one word that connects them. At 70 dB, participants averaged 5.80 correct answers. At 50 dB, they averaged 4.29. At 85 dB, just 3.88. The pattern formed an inverted U, with the moderate noise group outperforming both extremes.

The second experiment replicated this with a different task. Sixty participants were asked to generate creative ideas for a new kind of mattress. Independent judges rated the ideas, blind to condition. Again, the moderate-noise group produced ideas rated significantly more creative than both the quiet and loud groups. The difference wasn't subtle: judges scored moderate-noise ideas at 4.01 on a 7-point creativity scale, compared with 3.57 for quiet and 3.58 for loud.

A third experiment, with 95 participants, tested the time course. If noise boosts creativity through increased arousal, the benefit should fade once people habituate to the sound. If it works through something more persistent, the benefit should remain. Mehta's team measured heart rate and blood pressure alongside creative output. Heart rate spiked early in the moderate-noise condition, then normalized. But the creative benefit held steady whether participants tackled the task immediately or after a 15-minute delay. Arousal faded. The creativity did not.

Why Noise Makes You Think Bigger

The mechanism turns out to be counterintuitive in its own right. Moderate noise doesn't help by energizing you or putting you in a better mood. Mood measures showed no difference across conditions. Instead, it helps by making thinking slightly harder.

Mehta's team measured this directly. At 70 dB, participants reported higher "processing disfluency," a technical term for the subjective experience that thinking required more effort. That disfluency pushed participants into what psychologists call a higher construal level: they processed information more abstractly, thinking about the forest rather than individual trees. Abstract processing is precisely the cognitive mode that supports creative connections, because it loosens the tight links between concepts and lets distant ideas bump into each other.

The fourth experiment, with 42 participants, confirmed the complete causal chain. Noise increased processing disfluency, disfluency increased abstract construal, and abstract construal increased both the originality and the appropriateness of creative ideas. The multi-step mediation was statistically significant at every link. Importantly, the high-noise condition also induced disfluency and abstract thinking, but it simultaneously crushed processing capacity, meaning the brain was thinking abstractly but couldn't sustain the effort long enough to produce anything useful.

The fifth experiment moved into the real world. Sixty-eight participants completed tasks in a student lounge where noise levels varied naturally throughout the day. A hidden sound-level meter recorded the actual decibel level each person experienced. The pattern held: those exposed to moderate ambient noise were more likely to choose innovative products over conventional ones, and the effect was strongest among participants with high baseline creativity scores.

The Strongest Case Against

The most pointed critique targets sample composition and ecological validity. All five experiments used undergraduate students at one Canadian university, a population that skews young, educated, and Western. A 2022 study by Awada and colleagues at the University of Southern California found that 45 dB white noise, not 70 dB, optimized creativity among their participants, while 65 dB white noise increased stress without boosting creative output. That discrepancy matters: the optimal noise level may shift depending on the type of noise, the baseline noise floor, and individual differences in noise sensitivity.

There's also the question of task specificity. Mehta's creativity measures, the Remote Associates Test and open-ended idea generation, capture divergent thinking. But many real creative tasks require sustained convergent thinking, where you systematically refine a single idea rather than generate many. For convergent work, the evidence that noise helps is considerably weaker, and some studies suggest it actively hurts.

Finally, the absolute sample sizes across individual experiments, ranging from 42 to 95, are modest by contemporary standards. The effects are statistically significant and mechanistically coherent, but no independent large-scale replication of the full five-experiment package exists. The Awada study is the closest thing to a conceptual replication, and it only partially converges.

What We Didn't Prove

The optimal noise level of 70 dB was tested with restaurant-style ambient sound, not with music, conversation, or construction noise. Different noise textures carry different informational loads, and a 70 dB jackhammer almost certainly does not produce the same creative benefit as a 70 dB café murmur. The mechanism depends on moderate processing disfluency without semantic distraction, and speech-heavy environments may cross that line.

All laboratory experiments used acute exposure lasting minutes. Whether chronic moderate noise, the kind experienced by people who work in cafés every day, produces the same benefit or triggers habituation is unstudied. The field study in Experiment 5 captured naturalistic noise but still assessed participants only once.

Individual differences remain poorly mapped. Introverts, people with ADHD, and those with high noise sensitivity may respond very differently. The fifth experiment found that baseline creativity moderated the effect, but no study in this line has examined personality traits, neurodivergence, or age as moderating variables. Generalizing the 70 dB sweet spot to all humans doing all creative tasks is a claim the data do not support.

The Bottom Line

Silence is not the friend of creativity that intuition says it is. Across five experiments, a moderate level of ambient noise, about what you'd hear in a reasonably busy coffee shop, consistently outperformed both quiet and loud environments on measures of creative thinking. The mechanism isn't mood or motivation; it's that a manageable level of background noise makes thinking slightly effortful, which nudges the brain toward abstract processing, which is the cognitive mode where original ideas live. Too much noise overwhelms the system. Too little leaves it in a concrete, detail-oriented groove that produces conventional thinking. The sweet spot sits in between, right around where the espresso machine hisses and someone two tables over is having a conversation you can hear but not quite follow.

What You Can Do

When you need to generate ideas, brainstorm, or solve open-ended problems, try working in a café, a co-working space, or a busy common area rather than a silent home office. The research suggests the background hum is helping, not hurting.

If you can't physically relocate, ambient-noise apps like Coffitivity (literally built on this research), Noisli, or myNoise simulate café-level sound at around 70 dB. Set your volume to a level where you can hear the noise but would have to lean in to follow a specific conversation.

Reserve silence for tasks that demand precision and sustained focus, such as proofreading, debugging code, or detailed analysis. Those are convergent tasks where processing disfluency works against you. Match your sound environment to the type of thinking the task requires: ambient noise for generating possibilities, quiet for narrowing them down.

If you manage a team, reconsider the impulse to build quieter offices. The data suggest that some level of ambient sound in collaborative spaces may do more good than harm for creative work. The open-plan backlash is real, and noise does damage focused work, but the answer isn't maximum silence everywhere. It's zoning: quiet rooms for convergent tasks, moderately noisy common areas for divergent ones.

Sources

  1. Mehta, R., Zhu, R. (J.), & Cheema, A. (2012). Is noise always bad? Exploring the effects of ambient noise on creative cognition. Journal of Consumer Research, 39(4), 784–799. doi:10.1086/665048
  2. Awada, M., Becerik-Gerber, B., Hoque, S., O'Neill, Z., Pedrielli, G., Wen, J., & Wu, T. (2022). Cognitive performance, creativity and stress levels of neurotypical young adults under different white noise levels. Scientific Reports, 12, 14391. doi:10.1038/s41598-022-18862-w
  3. Toplyn, G., & Maguire, W. (1991). The differential effect of noise on creative task performance. Creativity Research Journal, 4(4), 337–347.
  4. Martindale, C., & Greenough, J. (1973). The differential effect of increased arousal on creative and intellectual performance. Journal of Genetic Psychology, 123(2), 329–335.
  5. Vallacher, R.R., & Wegner, D.M. (1987). What do people think they're doing? Action identification and human behavior. Psychological Review, 94(1), 3–15.