Good communication practices in times of crisis

When dealing with a crisis, communication takes on a critical role in guiding individual and collective behavior. Case in point: the anxiety that’s currently growing at pace with the global proliferation of the coronavirus. Evidence of the complexity of the problem is the controversial tone coloring the debate on the communication strategies that Italian institutions and media have adopted. Complexity intensified by the public’s constant need for news and the variety of information sources and channels people are tapping to satisfy this need.

In light of the crisis that we’re experiencing, now is a good time to reiterate certain basic notions of behavioral economics. This field of study, leveraging an interdisciplinary approach, analyzes people’s choice architectures and the psychology that underpins decisions and deviations from classical canons of rationality. Let’s start with this last area.

A number of scholars have verified the existence of what’s referred to as the “context effect.” Given a decision problem and a set of options for dealing with it (such as, in our case, different behaviors that would minimize the risk of contagion), this involves assessing the impacts that these options can have on the choices that individuals make. Put another way, a person’s preference for a certain option would depend on the number and type of available options. So for instance by introducing an additional option we can significantly influence people’s choices and behaviors, which defies certain fundamental principles of rationality.

So tying this context effect into the coronavirus emergency, to promote the adoption of more effective measures to curb the spread of the virus, we can specifically leverage the attraction effect and the compromise effect. The first happens when we take a set of options with diverse characteristics, and we add a new alternative (comparable to one of the others, but with fewer benefits and/or higher costs). This drives up the number of subjects who choose the option that’s most similar to the new entry. Second, we see the compromise effect when an option wins the greatest consensus if it’s presented as the middle ground with respect to the other more ‘extreme’ alternatives. In situations of uncertainty, individuals tend to turn toward these ‘compromise’ solutions to minimize the risks associated with the decision at hand.

Another contribution from behavioral economics springs from the demonstration of what’s called ‘systematic distortion’ in the probability distribution of an event, so that the least probable outcomes are overestimated, while the most probable outcomes are underestimated. This concept would explain the level of anxiety (and the behaviors that ensue) triggered by the fear of exposure to the virus and dread of the direst consequences of that exposure.

The function of objective and subjective probabilities also underscores how essential it is to communicate information that is properly contextualized. And with this we come to the third central contribution of behavioral economics, which has to do with the fact that our perceptive apparatus is better at recognizing differences and changes in the state of things, more so than evaluating a given condition in absolute terms.

The takeaway here is that we categorize the outcomes of a decision as either losses or gains in relation to a specific point of reference. What’s more, the (negative) value we attribute to a loss (with respect to the point of reference) is far greater than the (positive) value we associate with a gain of equal absolute value. This is what’s called ‘loss aversion.’ So for the current pandemic, when we hear that two people have died, we are much sadder (actually about twice as sad) than we are happy at the news of the same number of people recovering from the virus.

This illustrates the importance of establishing an appropriate point of reference for each specific crisis situation, and building the right communication frames to effectively activate the desired ‘framing effect’. To explain, when we modify the conceptual frame of a decision problem, while maintaining the same final outcomes, individuals make different choices. For example, the website of the Italian Ministry of Health states that ‘around one out of five people with COVID-19 become seriously ill and have respiratory problems.’ But the same information could be conveyed another way: ‘four out of five people who contract the virus do not become seriously ill.’ Which frame is better? Behavioral economics research shows that the second example with the positive frame will very likely assuage people’s fears about the seriousness of the epidemic (all other conditions being equal). But if the aim is to raise awareness of the problem, negative framing is the better way to go.

Beyond the need to know how to strike a balance between the value attributed to gains and losses, we also have to consider the fact that people’s propensity for risk taking is contingent on whether they are moving in a positive or negative context in relation to the reference point. Specifically, risk-averse behaviors emerge when people are considering and choosing between options that will generate gains; instead when facing possible losses, people typically lean toward risk-seeking. Going back once again to novel coronavirus, many media channels have decided to give the number of infected individuals as the first point of reference, going with a negative frame on the data (the evolution of the rate of contagion). In contrast, the Special Commissioner for the Coronavirus Emergency opens daily press conferences by citing the number of people who have recovered as the initial reference point, utilizing the rate of recovery as the information frame. Again, the choice between the two options depends on the reactions and behaviors that we want to activate on an individual level. To prevent people from adopting risky behavior for themselves and for others (such as going to the ER when they show symptoms of infection), the best option is positive framing, which encourages risk-averse behaviors.

Last of all, both for gains and losses sensitivity to changes decreases over time, which determines diminishing marginal utility/disutility. So here too we need to factor in the duration of the emergency: the longer it lasts, the less sensitive people become. In this scenario, communication needs to take on a broader scope, preventing lower sensitivity to the problem from turning into inappropriate behavior, which would be detrimental to individuals and to society as a whole.