Laura Bliss is a staff writer at CityLab, covering transportation, infrastructure, and the environment. She also authors MapLab, a biweekly newsletter about maps that reveal and shape urban spaces (subscribe here). Her work has appeared in the New York Times, The Atlantic, Los Angeles, GOOD, L.A. Review of Books, and beyond.
Do visual signs of neglect encourage criminality? The long-controversial idea may have just gotten a little stronger.
First devised by a pair of criminologists writing in The Atlantic some 34 years ago, the theory of “broken windows” swept U.S. cities during the crime-ridden 1980s and 90s. Former New York City Mayor Rudy Giuliani and Police Commissioner Bill Bratton famously put the idea that a neighborhood in visual disarray begets crime into practice, by targeting law enforcement in neighborhoods with abandoned houses, litter-strewn sidewalks, and graffitied walls. And they proudly (though perhaps wrongly) credited the principle as the key reason crime later dropped. Police departments nationwide continue to deploy similar tactics, under the stated goal of pre-empting more serious crimes.
Yet despite its lasting impact on criminology, the science behind the broken-windows theory was always a bit thin. The idea was that people “read” neighborhood disorder as a sign that no authorities are watching, allowing them to infer that they, too, can break the rules. But little data has ever proved if, and how, that really happens. Making such an interpretation would be full of subjectivities, assumptions, and complex reasoning about poverty, policing, and social norms around class and race—which hits on one major criticism of the theory, that it gives cops an pathway to penalize poor neighborhoods of color. And all those entangled strands of influence make the psychological mechanics of the theory exceptionally difficult to nail down.
Is there anything scientific about the way certain environments make us behave? Are there are any objective features that make an environment feel “disorderly,” and do those cues trigger deviant behavior of any sort? A new paper published in the Journal of Experimental Psychology by University of Chicago scientists shows that there may be a research-backed explanation for why some environments bring out poor behavior. It teases out some of the visual cues that make a space seem chaotic—and finds that those cues do, in fact, encourage a degree of rule-breaking.
The spatial psychology researchers Hiroki Kotabe, Omid Kardan, and Marc Berman recruited 105 participants from the online labor market Mechanical Turk and showed them a series of digital images of both natural and urban environments—placid nature scenes, quiet neighborhood streets, tangled forests, abandoned warehouses, and so on (some examples are shown below). Participants were asked to rate how disorderly or orderly a given scene appeared (without being given a definition of “disorderly,” since the goal was to first define this based on their subjective impressions). Kotabe, Kardan and Berman found that environments containing lots of curvy, broken edges and asymmetrical lines strongly predicted a “disorderly” rating.
In a separate experiment, the researchers exposed study subjects to a mix of orderly and disorderly images. Then they administered a challenging math exam, which the participants had to grade themselves on afterwards. The higher they scored, the participants were told, the larger the cash reward they received. Without saying as much, that grading model essentially cleared a path for cheating, if the subjects were so inlined.
And cheat many of them did—and more so if they’d looked at a certain kind of visual cue. In fact, the researchers found that subjects who were exposed to more disorderly images prior to the exam were 35 percent more likely to cheat, and cheated by a much greater magnitude, than did participants who saw more orderly images.
In other words, the researchers were able to isolate a specific visual feature—non-straight, asymmetrical lines and edges—and show how it seems to be connected to rule-flouting behavior. Kotabe says that these results resonate with prior research that’s shown disorderly environmental cues to be more tiring for people to process psychologically; essentially, they have to expend more mental energy reading the environment. And there are research-backed ties between cognitive fatigue and breaking more rules. “If you’re attempted to cheat, you might be more likely to do those behaviors after you’re fatigued,” Kotabe says.
This was, of course, just a single experiment that used digital images. But based on previous comparisons between psychological reactions to real-world and image-based environments, Kotabe thinks that a similar study conducted in various sites would produce “probably the same effects, except stronger.”
What the work suggests, then, is that rule-breaking might be less about complicated processes of social reasoning—I can get away with robbery in this neighborhood because that house is boarded up and no one’s watching—and more about simple perceptual properties of an environment—there are so many asymmetrical lines, I can barely think straight. All of which calls into question how much our actions really are in our control. “The brain stops exerting top-down control on what people are feeling or tempted to do,” Kotabe says.
The work may also suggest that planners and architects might start thinking more closely about the psychological effects of the spaces they design—especially insofar as it relates to the potential for crime. (Some builders, intentionally and otherwise, are already exploring the coercive character of “unpleasant design,” such as the sloping “Camden bench” that deflects homeless sleepers.) More research is needed, of course. The researchers want to try out other scenarios that test different forms of rule-breaking, such as littering, and look more closely at how people define disorder in both natural and urban environments.
Yet this paper is itself a valuable contribution to the conversation on broken windows, which has always been marked by questions and criticisms about the theory’s own window for bias. In their original Atlantic article, criminologists George L. Kelling and James Q. Wilson openly wonder, "How do we ensure ... that the police do not become the agents of neighborhood bigotry? We can offer no wholly satisfactory answer to this important question." And subsequent research has shown that people are more likely to call black neighborhoods “disorderly” than non-black neighborhoods, even when they have the same amount of graffiti, loitering, and litter. It’s an inference that has almost certain affected the neighborhoods targeted most heavily by broken-windows policing strategies. But if there really is something objective about the way “broken” environments make residents behave, perhaps there would be less room for discrimination.