FACULTY FOCUS

Spike W.S. Lee

they don’t just WHEN PEOPLE ASSESS RISKS, apply facts and data to the target risk. Research shows that they also apply their feelings. Therefore as emotions run high during a crisis like a global pandemic, people tend to perceive higher health risks and overgeneralize them, which exerts predictable influence on socio-political attitudes.

COVID-19 provides a powerful illustration of this phenomenon because a) it is a salient, enormous threat; b) it is unfamiliar; and c) it is intertwined with xenophobia.

Last summer Dr. Neil M. Ferguson, a British epidemiologist who is regarded as one of the best disease modelers in the world, produced a sophisticated model with a worst case scenario of 2.2 million deaths from COVID-19 in the United States alone. In the face of such a salient, enormous threat to humanity, fear knows no limits.

Consider my research findings from the 2009 H1N1 (‘swine flu’) pandemic. Within a matter of two months, the World Health Organization declared a public health emergency of international concern as the virus spread to more than 70 countries and all 50 states in the U.S. Media coverage was extensive, highlighting the risk of contagion and the importance of frequent hand washing and avoiding physical contact.

I conducted two experiments with Julie Huang of Stonybrook University’s College of Business and Norbert Schwartz of the University of Southern California. The first took place three weeks after the first documented case of human infection in the U.S., while H1N1 remained the primary focus of media attention. Passers-by were approached on a university campus and asked to estimate the probability that the average American may experience the following events within the next 12 months:

• Contract a serious disease;

• have a heart attack before age 50; or

• die from crime or an accident.

These questions tapped into three perceptions: the health risk directly related to H1N1, a health risk unrelated to H1N1, and a non-health risk. Respondents were also asked to express their overall view of the U.S. healthcare system by completing an item adapted from a New York Times/ CBS News poll.

To test the influence of disease salience on these measures, a research participant would walk by and sneeze or

cough loudly (in the experimental condition) or just walk by without sneezing or coughing (the control condition) before respondents answered the questions.

Our perception of risk is often over-generalized to unrelated domains.

Risk estimates were higher by an average of 13.7 OUR FINDING: percentage points in the experimental condition than in the control condition. And this increase was observed regardless of whether the risk was related to H1N1, unrelated to it, or even unrelated to health. Disease salience also led respondents to express a less-favourable view of the healthcare system and to consider it more in need of complete rebuilding.

Our second experiment was conducted three weeks later and involved a similar procedure. This time, passers-by in the downtown area of a college town were asked if they would prefer to spend a $1.3 billion federal investment (a) to facilitate the production of flu vaccines, or (b) to create green jobs. Without any disease cues (i.e. no sneezing or coughing), 16.7 per cent of respondents favoured the flu vaccine production, while with disease salience, 47.8 per cent favoured it.

These findings have important implications. The most obvious is that people’s risk perception can be both heightened and over-generalized to unrelated domains. Risk overgeneralization is likely to contribute to risk-averse behaviours across the board, as exhibited in the present pandemic — from hoarding products at the individual level to market sell-offs at the collective level.

Furthermore, the public’s attitude towards the healthcare system and preference for federal spending (e.g., on vaccine production) can change dramatically in the midst of a palpable crisis. Public opinions on other policies are likely to ride on the wave as well — as seen in the current rising support for paid sick leave and governmental payouts, among other institutional changes.

In motivating these public opinion changes, policymakers would therefore want to ensure that the disease threat is salient. As the sneezing experiments illustrate, making a health threat salient in the moment exerts a significant influence even when the threat already has extensive media coverage.

The Effects of Unfamiliarity

Research by others shows that new and unfamiliar threats often feel riskier than old and familiar ones, and that this feeling is adaptive insofar as caution around unknowns is often warranted. As the late psychologist Robert Zajonc liked to say, ‘If it’s familiar, it hasn’t eaten you yet’, whereas new and unfamiliar threats just might. Indeed, feelings of unfamiliarity with an entity can heighten people’s perception of its riskiness (e.g., the perceived health hazards of food additives), resulting in reduced investment in stocks and reduced trust in others.

Coronavirus — which has ‘novel’ right in its name — was first found in a Chinese city that many people had never heard of. Facts about it, including infection and mortality rates, geographical reach, time course of spread, treatment options and long-term effects, were either unfamiliar or unknown — all of which have reinforced the perception of its riskiness.

As the pandemic unfolds, however, facts are emerging. Insights are being gained and information is becoming widely available. Transparency is critical here, because any perceived delay or suppression of information undermines public trust and sustains public fear. This was evident in China, where the government’s handling of coronavirus information caused a wave of widespread, publicly-expressed skepticism about its intentions and competence — which is rare under the Communist Party’s rule. In more open, democratic societies, it is crucial that the public be continuously informed, without delay, in ways that they can comprehend. Failing to do so will prolong heightened risk perception and with it, risk-averse behaviour.

The Dire Result: Xenophobia

Xenophobia can be defined as ‘fear and hatred of strangers or foreigners or of anything that is deemed to be strange or foreign’. Research on the behavioural immune system — the psychological and behavioural mechanisms that we use to avoid disease and that both activate and are activated by our biological immune system — suggests that contagious disease threats sensitize us not only to personal risks (as

described above), but also to inter-group risks, resulting in increased prejudice against ‘outgroups’ — people unlike ourselves.

Consider illustrative evidence from the height of the H1N1 pandemic. An online experiment in the U.S. during fall 2009 found that participants reported more racist attitudes towards immigrants if they had previously read a news article regarding the swine flu’s health risks and limited vaccine supply than if they had not read that article. This effect, however, only emerged among participants who had not been vaccinated; it did not emerge among those who had already been vaccinated, nor did it emerge for participants who had not read the article.

Converging evidence came from a lab experiment among U.S. undergraduates, which showed that more germ-aversive participants expressed more unfavourable attitudes towards outgroups after reading a news article about the seasonal flu. However, if they had been given an opportunity to clean their hands with an antiseptic wipe, they did not express unfavourable attitudes.

These findings are consistent with the notion that people’s proclivities toward avoiding disease risk can produce over-generalized avoidance of foreign or unfamiliar entities (e.g., immigrants and other out-groups) even when they are not actual disease vectors. Vaccination and hand hygiene — two protective mechanisms against disease — have the potential to weaken perceived disease risks and subsequently, the corresponding xenophobia.

Similar conclusions can be drawn from related research. Survey data from a nationally-representative American sample during the 2014 Ebola outbreak showed that the more vulnerable to Ebola participants perceived themselves to be, the more they exhibited generalized xenophobia. Meta-analytic data across 24 studies also found that disease avoidance tendencies (e.g., fear of contamination) are associated with ethnocentrism and other forms of social conservatism motivated by exclusivity and negativity towards outgroups.

Furthermore, longitudinal analyses of U.S. polling data before and after the Ebola outbreak found that the amount of Internet searches for ‘Ebola’ was associated with increased inclinations to vote for Republican candidates in Senate and House of Representatives elections — but only in states with norms that already favoured Republican candidates.

Such dynamics of the ‘behavioural immune system’ can either tear apart or mend the fabrics of a diverse society, and they are likely to be particularly powerful in the current environment for two reasons.

First, coronavirus appears considerably more severe than the 2009 H1N1 and the seasonal flu. Second, a vaccine is not yet available. Until it is, we must depend on non-pharmaceutical interventions like wearing masks, physical distancing and temporarily closing public spaces. The success of such interventions, however, may be limited by ‘us versus them’ thinking, since they require the cooperation of multiple actors, from individuals to governmental institutions.

When the success of public health responses depends on all parties comprehending the seriousness of the health risk and jointly complying with behavioural restrictions, illness-ignited xenophobia and prejudices constitute a particularly insidious social threat.

Consequently, it behooves governments and organizations to stay mindful of the link between disease risk and xenophobia, even as they seek ways to keep the public’s attention on coronavirus. To that end, the World Health Organization explicitly cautions against naming or referring to human diseases by culture or geography, and a media advocacy group is urging fellow journalists to avoid using images of East Asians in face masks and U.S. Chinatowns when describing coronavirus.

By calling coronavirus the ‘Chinese virus’, President Trump directed blame to an outgroup and away from his government’s response, while fueling anti-chinese and anti-Asian racism and hate crimes. These dynamics make actions such as border-closures particularly appealing, which is likely to reinforce the division between ‘us and them’ whether the line is drawn between Americans and non-americans, Caucasians and Asians, or New Yorkers and Floridians.

The fact is, when a pandemic is already spreading through local communities, and race/ethnicity fails to provide a valid predictor of who is or is not a carrier, the false sense of security and psychological satisfaction provided

New and unfamiliar threats often feel riskier than old, familiar ones.

by such policies can backfire — especially at a time when massive collaborative efforts are needed.

In closing

COVID-19 is a salient, enormous threat that cuts across national and racial/ethnic boundaries. As a contagious disease threat, many facts about it remain unfamiliar to most people, and its global origins predispose its spread to worsen xenophobia.

Consequences such as risk over-generalization and risk-averse behaviours are hard to curb unless and until trustworthy information and effective prevention and treatment become available, which will reduce the enormity and unfamiliarity of the threat.

By highlighting the consequences of powerful feelings invoked by COVID-19, my colleagues and I hope to convey the pros and cons for practitioners and policymakers to consider. On the one hand, leaders have to emphasize the magnitude of the population-level risk from coronavirus and how urgent action by everyone is needed. On the other hand, the same urgency they wish to communicate may drive up riskaverse behaviour, xenophobic attitudes, and potential harm against already-marginalized populations. As a result, concrete solutions and guidelines for reducing these undesirable consequences should be in place alongside other public health recommendations.

Spike W.S. Lee is an Associate Professor of Marketing at the Rotman School of Management and is cross-appointed to the University of Toronto’s Department of Psychology. This is a summary of his coauthored paper, “Risk Overgeneralization in Times of a Contagious Disease Threat”, which was published in Frontiers in Psychology in June, 2020. Co-author Julie Y. Huang is an Assistant Professor of Marketing at Stonybrook University’s College of Business. Co-author Norbert Schwartz is the Provost Professor of Psychology and Marketing at the University of Southern California’s Dornsife College of Letters, Arts and Sciences.