(Inside Science) -- Do you see your neighborhood as safe, or riddled with thieves and cutthroats? Either way, there's a good chance you're wrong, according to past research. Now, a new mathematical model may help explain why people's fear of crime often swings so wide of the facts.
Rafael Prieto Curiel, a mathematician at University College London in the United Kingdom, saw the disconnect between fear and facts firsthand when he worked for the Mexico City police force from 2009-2013. Police officers dramatically reduced crime rates during that time, but their achievement did hardly anything to improve Mexico City's reputation as dangerous, said Prieto Curiel. The experience inspired him to study how fear spreads via social interactions.
For the new study, published today in Proceedings of the Royal Society A, Prieto Curiel and his colleagues built a mathematical model that simulated a population of 100,000 people over six years. The "people" in the model had different levels of fear about crime, and they updated those levels each week based on their recent experiences.
For example, if a person in the model suffered a crime, their fear rose sharply. If nothing happened, their fear gradually faded, as it would if someone forgot past experiences or assumed that things had gotten better. And if they interacted with someone who had a different level of fear, they adjusted their own fear to be closer to the other person's.
Importantly, social interactions didn't cause both people to shift their views by equal amounts. Instead, the formerly secure person became much more fearful, while the frightened person was only slightly reassured. Prieto Curiel believes this imbalance is realistic because of past research involving interviews with real people, which suggests that people are strongly influenced by other people's crime stories and perceptions of danger. Fear, he said, is contagious.
According to the model, even low crime rates were enough to make fear spread through the population. For example, if just one in every 200 people was the victim of a crime every year, the model suggests that 40 percent of people would end up with significant levels of fear, said Prieto Curiel. The actual robbery rate in Mexico is about 20 times that level, he added.
Strikingly, when the researchers changed the crime rates in the model, people's fears only weakly reflected the difference. Even doubling the crime rate or cutting it in half had a "tiny, negligible impact on the perception of security," said Prieto Curiel.
Social dynamics, on the other hand, could profoundly influence fear levels when crime rates were above a certain threshold. In the real world, certain subgroups are relatively safe from crime, whereas others, such as young men, are disproportionately victimized. The researchers included such inequalities in their model, and manipulated how much the different groups talked to each other.
They found that when people avoided socializing with members of other groups, fear stayed largely confined to the groups suffering the most crime. However, when people from different groups talked to each other, fear spread to those who were relatively safe, increasing the average levels of fear in the population. The finding might help explain why regions with similar crime rates sometimes have vastly different levels of fear.
"You are not creating a safer society. But just by controlling the interactions between groups, you might create a society that feels more secure," said Prieto Curiel. "For me, that was a little surprising."
Of course, the model is much simpler than the real world, and it doesn't include many factors that would influence the fear of crime, said Andromachi Tseloni, a quantitative criminologist at Nottingham Trent University in the United Kingdom who was not involved in the study. Moreover, Tseloni said she is not convinced that fears always spread easily through social interaction. For example, if someone hears about a type of crime that can't possibly affect them, it likely won't make them afraid.
"If someone has a Lamborghini, and the Lamborghini's stolen, I'm not going to care, because I will never own a Lamborghini," she said.
Nevertheless, the model helps make sense of a wealth of real-world data that researchers have gathered in the past.
"It is a big step forward in understanding why fear of crime most of the time does not match with the criminal victimization risk," said Tseloni.
Fear of crime is not necessarily a bad thing, noted Prieto Curiel. It can prompt people to take wise precautions, such as locking doors and keeping track of belongings on the subway. But unfounded fears can also lead to unwise actions. For example, if a police force pays attention to people's fears rather than actual crime rates, it might invest its resources in the wrong neighborhoods, and thus fail to protect people who are actually in danger.
While the study focused on crime rates, Prieto Curiel believes the findings may also apply to other scenarios, such as immigration and terrorism. Perhaps to manage these issues wisely, we need to focus on facts, not on fears that can spread like a disease.