Surveillance changes behavior in predictable ways: people become more cautious, more compliant with social norms, and more self-conscious about their choices. This happens whether the watching is real or merely perceived. The effects show up everywhere from workplaces to city streets to online browsing habits, though they’re not always the effects the watchers intended.
The Psychology of Being Watched
The core mechanism is simple. When people believe they might be observed, they shift from acting on impulse or personal preference to acting in ways they think will be judged favorably. The philosopher Jeremy Bentham described this with his concept of the Panopticon, a circular prison where guards could see every cell but prisoners could never tell if they were being watched at any given moment. The uncertainty itself was the point. Inmates would regulate their own behavior continuously because they could never be sure they were unwatched.
This principle extends well beyond prisons. Research in psychology has found that even images of eyes, not real cameras, alter how people act. In experiments using the “watching eyes effect,” people who saw eye-like images became measurably more generous, splitting resources more equally with strangers compared to those who saw neutral images like flowers. They also used less self-serving language. The explanation is that eye cues activate the same social awareness you’d feel with an actual audience: a heightened concern for your reputation and a stronger pull to follow whatever norms apply to the situation.
In contexts where the rules are clear, like paying for drinks in an honor-system coffee station or sorting waste into the right bin, eye cues make people noticeably more compliant. The effect works not because people consciously think “someone is watching me” but because the cue triggers an automatic shift toward behavior that avoids social punishment.
How Surveillance Shapes Online Behavior
The behavioral shift becomes more concerning when it suppresses legitimate activity. Researchers call this the “chilling effect,” where people avoid exercising legal rights because they fear scrutiny. Studies have found that people are less likely to search for certain articles or topics online when they believe the government may be tracking those searches. The behavior being suppressed isn’t illegal. It’s reading, browsing, and asking questions. But the perception of surveillance narrows what people are willing to explore, creating a kind of invisible self-censorship that’s difficult to measure because the suppressed behavior simply never happens.
This extends to speech as well. People become less likely to express controversial or dissenting opinions when they believe their communications are monitored. The result is a flattening of discourse, where public conversation skews toward safer, more conventional positions. Over time, this behavioral uniformity can become self-reinforcing as people mistake the absence of dissent for genuine consensus.
Workplace Monitoring: A Mixed Record
The explosion of remote work accelerated a trend that was already underway. Roughly 78% of companies now use some form of employee monitoring software, up from about 60% in 2021. These tools range from simple time trackers (used by 96% of companies with monitoring) to sophisticated AI systems that log keystrokes, capture screenshots, and even build predictive models of which employees are likely to quit.
Employers and employees see the results very differently. About 68% of managers believe monitoring improves productivity, and some companies have reported productivity gains of up to 28% in the first year. But 72% of workers say monitoring either has no positive impact or actively makes their work worse. One in ten employees reports working less effectively when monitored, and twice as many say it interferes with their ability to focus.
The stress numbers tell a clearer story. Over 56% of employees report anxiety related to workplace surveillance, and 38% say it increases their overall stress levels. Research on AI-powered monitoring tools has found they can actually decrease productivity while increasing quit rates. More than half of employees say they would consider leaving if their employer ramped up surveillance. So while monitoring can produce short-term compliance, it often trades long-term engagement and retention for the appearance of control.
How Long Do the Effects Last?
One of the most important and least discussed aspects of surveillance is that its behavioral effects tend to fade. A systematic review of the Hawthorne effect, the phenomenon where people change behavior simply because they know they’re being studied, found that observed changes are generally short-term. Most studies showed effects lasting three to six months. Only a handful demonstrated behavioral changes persisting beyond six months, and even fewer showed effects at the twelve-month mark.
This has real implications. When a new camera system goes up in a neighborhood or a company rolls out monitoring software, you can expect an initial spike in compliance or productivity. But people adapt. They learn the blind spots, develop workarounds, or simply stop caring as the novelty wears off. Sustained behavioral change from surveillance typically requires either escalating the monitoring or pairing it with other interventions.
Public Surveillance and Crime
CCTV cameras are one of the most studied forms of public surveillance. A 40-year systematic review published by the Office of Justice Programs found that cameras are associated with a modest but statistically significant reduction in crime. The effects vary enormously by location. Parking lots showed the largest and most consistent crime reductions, followed by residential areas. The results in city centers and public transit were less clear-cut.
Two factors made camera systems substantially more effective. First, active monitoring, where someone actually watches the feeds in real time, produced larger effects than passive systems that simply record footage. Second, cameras deployed alongside other interventions like improved lighting, signage, or increased patrols performed better than cameras used alone. In other words, the camera itself matters less than the broader environment it creates.
This aligns with the psychological research. Cameras work partly as deterrents and partly as signals that a space is being managed and maintained. A camera in a well-lit parking garage with visible security staff sends a very different message than a dusty camera mounted on a pole in a neglected lot.
Gamified Surveillance: China’s Social Credit System
Perhaps the most ambitious example of surveillance-driven behavior change is China’s social credit system. One well-studied municipal version scores residents using 389 rules: 124 that reward desirable behavior and 265 that punish undesirable behavior. People are classified into eight tiers, from AAA at the top to D at the bottom. High scorers receive tangible perks like discounts on heating bills. Those rated D face police monitoring.
The system demonstrably changes what people do. Rewarded behaviors fall into five categories: charitable donations, volunteer work, government service, social honors, and reporting problems. When the city promoted civic engagement campaigns, volunteer-related rewards increased accordingly. During the COVID pandemic, new rules were added rapidly, penalizing mask refusal (a 10-point deduction) and rewarding participation in disease control efforts (up to 100 bonus points). Public servants are required to earn bonus points annually through volunteering to remain eligible for promotions.
The system also reveals how surveillance can suppress legally protected behavior. About 67% of the political offenses recorded in one studied city involved petitioning the local government, an activity that is technically legal but discouraged by officials. When a scoring system gives authorities the power to define and penalize “bad” behavior in granular detail, the line between encouraging civic participation and punishing dissent becomes dangerously thin.
The Tradeoff Between Compliance and Autonomy
Surveillance reliably produces more norm-compliant behavior. People litter less, pay more honestly, work more visibly, and commit fewer crimes in monitored spaces. But the same mechanism that increases compliance also reduces autonomy, creativity, and willingness to take risks. Monitored employees stick to safe, measurable tasks rather than experimenting. Monitored citizens avoid controversial but legal activities. Monitored internet users narrow their reading and searching to topics that feel safe.
The behavioral effects also depend heavily on who holds the power to define what “good behavior” looks like. A camera in a parking garage and a scoring system that penalizes petitioning your government both operate through the same psychological lever: the awareness of being watched. The difference lies entirely in what the watcher rewards and punishes. Surveillance is never neutral. It always encodes someone’s idea of how you should behave, and the closer you look at any surveillance system, the more that embedded value system reveals itself.