It’s not me — it’s you: we believe we’re less likely than another to fall for phishing hooks
New cybersecurity study has significant implications as we increasingly work remotely during COVID-19 pandemic
We believe we are less likely than others are to fall for phishing scams, thereby underestimating our own exposure to risk, a new cybersecurity study has found. The research, led by New York University’s Department of Psychology and co-authored by Quanyan Zhu, a professor of electrical and computer engineering at the NYU Tandon School of Engineering, also reports that this occurs, in part, because we overlook data, or “base rate information,” that could help us recognize risk when assessing our own behavior yet use it to predict that of others.
COVID-19 has had a devastating impact on the physical and mental health of people around the globe. Now, with so many more working online during the pandemic, the virus threatens to wreak havoc on the world’s “cyber health,” the researchers note.
“This study shows people ‘self-enhance’ when assessing risk, believing they are less likely than others to engage in actions that pose a threat to their cybersecurity — a perception that, in fact, may make us more susceptible to online attacks because it creates a false sense of security,” says Emily Balcetis, an associate professor in New York University’s Department of Psychology, who authored the study, which appears in the journal Comprehensive Results in Social Psychology.
Added Zhu, “Human engagement is the weakest link in many cyber security problems. Understanding human incentives and behaviors is a critical step toward developing technical cyber defense solutions.”
Through March, more than two million U.S. federal employees had been directed to work from home — in addition to the millions working in the private sector and for state and local governments. This overhaul of working conditions has created significantly more vulnerabilities to criminal activity — a development recognized by the Department of Homeland Security. Its Cybersecurity and Infrastructure Security Agency issued an alert early last month that foreshadowed the specific cyber vulnerabilities that arise when working from home rather than in the office.
In their study, the researchers sought to capture how people perceive their own vulnerabilities in relation to others’.
To do so, they conducted a series of experiments on computer screens in which subjects were shown emails that were phishing scams and were told these requests, which asked people to click links, update passwords, and download files, were illegitimate. To tempt the study’s subjects, college undergraduates, they were told complying with the requests would give them a chance to win an iPad in a raffle, allow them to have their access restored to an online account, or other outcomes they wanted or needed.
Half of the subjects were asked how likely they were to take the requested action while the other half was asked how likely another, specifically, “someone like them,” would do so.
On the screen that posed these questions, the researchers also provided the subjects with “base rate information”: The actual percentage of people at other large American universities who actually did the requested behavior (One, for instance, read: “37.3% of undergraduate students at a large American university clicked on a link to sign an illegal movie downloading pledge because they thought they must in order to register for classes.”)
The researchers then deployed an innovative methodology to determine if the subjects used this “base rate information” in reporting the likelihood that they and “someone like them” would comply with the requested phishing action. Using eye-tracking technology, they could determine when the subjects actually read the provided information when reporting their own likelihood of falling for phishing attempts and when reporting the likelihood of others doing the same.
Overall, they found that the subjects thought they were less likely than are others to fall for phishing scams — evidence of “self-enhancement.” But the researchers also discovered that the subjects were less likely to rely on “base rate information” when answering the question about their own behavior yet more likely to use it when answering the question about how others would act.
The study was supported by the National Science Foundation’s directorates for Computer and Information Science and Engineering and Social, Behavioral, and Economic Sciences (1720230).
