Mark Power / Magnum

College students are being invited back onto campus after a summer of isolation and confinement, with strict instructions to stay apart. School officials, pleading with students to treat their new autonomy gingerly, are leaning hard on shame to get compliance. The president of Penn State recently admonished students, “Do you want to be the person responsible for sending everyone home?”

Meanwhile, elementary-school students are resuming the school year on Zoom, and districts are laying down new rules: Keep laptop cameras on; dress properly, as if attending school in person; stay seated for the lesson—even when, best-case scenario, sitting at the kitchen table instead of the math table.

While the demands vary by age—don’t space out on camera; don’t congregate on the quad—many share a faulty premise: If young people would just be careful and pay attention, everybody could face the school year as though the disruption were cosmetic rather than structural.

Willpower is not going to be enough. The life of the American student, from kindergarten through college, is fundamentally embedded in a collective enterprise, one whose success depends on where teaching takes place, who is present, and how everyone involved is used to behaving.

The analogy between remote and in-person learning is more fragile than it looks. The districts expecting second graders to lock into their laptops for six hours a day, like the universities expecting college students to lock into their dorm rooms week after week, are not reckoning with the psychological burdens that remote learning imposes. And for a society accustomed to the norms and rhythms of a system designed for humans in rooms together, the tax on students can be hard to fathom.

For American students at all levels, the school day offers a social and physical setting to reinforce academic progress. Classrooms and lecture halls announce their purpose, and the ritual of moving from room to room buttresses the connection between the physical and the intellectual contexts. This semester, many students will take part in online school from a single chair, where they must try to focus on a screen while what happens around them is unrelated, if not at cross-purposes, to the learning they are doing. This stasis is good for public health but hard on students. Moving from classroom to recess or dorm to dining hall is not merely an incidental aspect of the educational experience. Learning is more effective, cognitive research shows, when embedded in the material world. Indeed, just changing physical locations during studying increases subsequent recall of the material in question. Animal research suggests that nonspatial information—a category that for humans includes faces, events, and facts—can be easier to remember when tied to a specific place.

Perhaps more important is the social context of learning. That context is extraordinarily hard to replicate online for reasons that are not always obvious.

Humans are powerfully attuned to social feedback, whether they are aware of it or not. Social rewards, the little affective fillips you get from human interaction, flow from social connection. People get validation, measurable in neuroimaging studies, from eye contact. Children watching funny videos smile and laugh more when other people are around. Even casual social bonding releases the love hormone oxytocin. Normally, these boosts push the day along, easing difficult tasks and smoothing out moments of boredom or discomfort.

Videoconferencing seems like a low-cost approximation of normalcy. Some instructors, for example, prefer that students keep their camera on because doing so fosters a community environment—and because teaching into a sea of black screens is not easy. In a physical classroom, of course, students are visible to one another. But putting your home or dorm room on virtual display invites a much higher level of scrutiny that, for reasons of privacy, some students do not want.

Normally, humans can sense that they might have a moment of privacy in a crowded place. Video calls disrupt that expectation. Students who think that they’re merely watching risk awkward moments, like when unsuspecting sports fans are featured on the jumbotron but don’t realize it yet.

To teach or learn under these circumstances requires mental exertion. English has a number of idioms that evoke the feeling of effortless, involuntary focus—caught up, carried away, in a groove, going with the flow. Since March, flow has been hard to come by. If a busy day in January or February felt like getting on a luge course, where social, psychological, and organizational urgencies whisk you along from one task to the next, pandemic life is more like snowshoeing: Every step takes conscious care, and the environment actively hinders forward progress.

A line of psychological research on a subject known as “ego depletion” has tried to document the costs of exercising self-control under adverse circumstances. The study protocols ask subjects to engage in an annoying cognitive-override task and then to turn to something else that is otherwise difficult. The first task is often something like this: Please think of zoo animals, but every time you think of a white bear, suppress the thought, stop thinking of the white bear, and move on. Ego-depletion research suggests that immediately frustrating cognitive tasks—such as ignoring a family dog, a crying sibling, or a comfortable bed—will make people give up sooner on subsequent hard tasks and give in more readily to temptation. (That all of this is now occurring amid the loneliness, grief, and fear caused by the pandemic makes every task all the more difficult.)

One study asked a question pertinent for universities: How does ego depletion affect risk-taking behavior? After subjects suppressed thoughts of white bears, they performed a variety of tasks meant to measure risk-taking behavior, including a simulated driving test. Relative to a control group, people who engaged in the depleting activity later sought more risk, and not just behind the wheel.

Many K–12 schools and colleges have declined to hold classes in person for entirely sensible safety reasons, and the psychological challenges of remote learning are nowhere close to the most serious problems facing many students this fall. Without physical school—already a reflection of America’s deeply inequitable society—the most vulnerable children have less access not just to the tools of learning but to basic necessities, including food, shelter, and reliable supervision. Indeed, the gravest threats to education this fall are driven by long-standing, unconscionable disparities. Yet a lack of attention to the psychological hurdles to online learning will only amplify those inequities.

For decades, scholars of behavioral economics have been exhorting those in power: If you want people to make choices that are costly to them in the short term but beneficial in the long term, you have to structure the environment to make those choices easier, more palatable, and more habitual. That same principle applies even during a pandemic. When young students cannot pay attention to a videoconference, they are not misbehaving; they are responding to the limitations of an electronic medium. Likewise, when college students meet up at bars or indoor parties in ways that are likely to spread a deadly virus, schools need detailed plans to promote social behavior in other forms. Create opportunities for outdoor interactions; incentivize choices that support common goals; ask students at every level what they need to succeed, and take them seriously.

The plan cannot just be to keep telling them to stay in their dorm room or their seat. Institutions should recognize that, even in the best version of this fall, students will be paying a steep price. The system does not work without their cooperation, and educators who want to meet students halfway need to understand what is happening to them.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.