But it is a wild alternative to despair and fatigue, despair and optimism, such as a scene of change in the last fall, when Americans returned from a holiday trip in the worst of the epidemic. And now, despite the summer peaks that are as bad as ever, business is back to normal in many parts of the country’s society. “People change their behavior dramatically during the ongoing epidemic,” Bergstrom said. “We are constantly updating our beliefs about how serious this is.”
In some ways, this could mean creating more experience with the epidemic More Uncertainty for modelers, no less. Beliefs and behaviors are now increasingly heterogeneous, changing from state to state and, in some cases, from city to city. Delta has arrived at a time when people are becoming more and more polarized in terms of vaccinations, and confused about how they should behave. “One month’s mask order is okay, and the next month’s protest. It’s really hard to predict in advance, ”Gakidu said.
Joshua Weisz, a professor of complex biological systems at the Georgia Institute of Technology, said: “The theme that is making it difficult right now is the state of the disease, how people react, and how people react over time.” Within 18 months of the epidemic it is a completely intuitive idea that our personal perception of the risk and the behaviors that follow from it will have a cumulative effect on the course of the virus. But in the beginning it was not a universal understanding, Witz noted, when some believed the epidemic would go away quickly. In the case of modeling, the term for it (a symbol of 19th-century epidemiological theory) is Fur’s law: transitions must be maximal and then decrease at a relatively equal rate, creating an hourly curve.
This curvature is not going to be tolerated. Last spring, Witz and others saw it coming back in the second round. The first wave was not completely crushed, and many people remained sensitive. The cases peak, then get stuck in the “shoulder” of the curve, slow down at a slower rate than many of the proposed estimates, and then become plates at a stubborn rate of infection. The behavior, Weitz speculated, was not consistent with how the models predicted the intervention, such as the stay order. Studying reports of dynamics pulled from cellphone data, a proxy of how much people are experiencing social interaction, he found that risky behavior decreased as it decreased, but then resumed before turning into a corner. “People look around, see the local situation and they change their behavior,” Weitz said.
One consequence of this reactive behavior is that helpful principles such as the order of masks and vaccines can be difficult to analyze. There is an ambiguity between cause and effect – and both government action and what the public is already doing are responding to the rise and fall of the infection rate. For example, he said, if you look at the timing of the mask mandate introduced in Georgia last year and compare the case rate before and after, you can determine that it has had very little impact. But what if it was because people realized that the rate of lawsuits was increasing and had already donated their masks? What if they start living more at home? Or what if it were different: the requirements were met and very few people followed the rules, so the masks never got a chance to do their job? “Obviously there’s a relationship there,” he says. “I can’t claim we’ve reached the bottom of it.”
For models, this uncertainty presents a challenge. To assess when the Delta Geu could end, one might look at places where it has already grown and like the UK. But will it die fast, or slow, or perhaps take the plateau at a steady rate of infection? These scenarios, Witz argues, will rely most on how people perceive and behave at risk. The low-vaccination Delta variant is expected to hit the high-vaccine Vermont compared to Alabama and eventually decline differently. Different policies for schools and businesses will determine how many different groups of people will blend in, and how people will respond independently will increase or decrease it.