In this session, we'll explore an important way that empathy influences how we can help others deal with risk, which is how we can better communicate information about risk. We've talked about some ways that limits on human judgment can also put limits on our ability to evaluate information about risk. Understanding these limits can help us present information in a way that's useful and impactful to our audience. I'll start with a few examples that should be salient. Regardless of your opinions about the COVID-19 pandemic and measures taken to mitigate the pandemic, I believe it's reasonable to say that the pandemic provided many examples of poor risk communication from political leaders, commentators, public health officials, and others. Regardless of your opinions, these instances of poor communication sometimes overstated risks and sometimes understated risk. No matter which way you slice it, people need to understand the risks in order to evaluate the costs and benefits of responding to that risk. One example that comes to mind is a quote by the director of the CDC, Rochelle Walensky, from July 2021. She said, of every 20 vaccinated people, one or two of them could get a breakthrough infection. In other words, she was saying there's a five to 10 percent risk. Now, at the time, data suggested the actual risk for vaccinated people was between 0.05 percent and 0.1 percent, or about 100 times less than the CDC director claimed. Now, everyone makes mistakes. We're all human. But it's one of many examples of overstating or understating risk during the pandemic. One of the more interesting examples is that people tend to do both at the same time. One example of this is that the public, on average, underestimated COVID risk for older people and overestimated COVID risk for younger people. The risk was real for both groups, but we weren't very well calibrated. What follows are a handful of principles to consider when trying to communicate information about risk to others. The first two principles apply to situations other than risk as well, but are worth mentioning here. First, use visuals whenever possible. It's a cliche that a picture is worth a thousand words. But it's true. Visuals, condensed information and allow us to store information as an image and verbally through our interpretation of the image. This makes the information in a visual easier for us to retrieve from memory later. Second, use plain language that is understandable to your audience. A good rule of thumb is to seek to communicate at a grade level that matches the reading level of your typical audience. If you're communicating to the general public then an eighth grade reading level is probably high enough. In a lot of cases you may be communicating to an audience with a higher reading level, so tailor your communication to what your audience understands. Communicate to inform, not to impress. A second rule of thumb about plain language is to minimize technical or domain specific jargon. You may understand financial terms like diminishing marginal returns or statistical terms like multicollinearity, but your audience probably doesn't. People only give you so much attention before they shut down. Do you want them to spend all of their attention defining all of your terms or spend a few moments coming up with an easier to understand message? Third, order of presentation matters, particularly due to the recency effect and the primacy effect. This means that when evaluating a sequence of information, people tend to remember information that comes first and information that comes last. In one classic psychology study, a researcher asked participants to learn a list of words and then recall the words afterwards. Each word was presented for 1-2 seconds. People were more likely to remember the words that came first and the words that came last. This held whether the list was short, 10 words, or long, 40 words. A possible explanation is that words early in the list get put into long-term memory, which is the primacy effect because the person has time to rehearse each word acoustically. Words from the end of the list went into short-term memory, which is the recency effect. Working memory can typically hold about seven items. In an example of risk communication, a medical study found that when the risks of a breast cancer drug were presented after the benefits, participants perceive the risks is more worrisome and more common, a classic recency effect. What helps when presenting a long sequence of effects, try summarizing the risks and benefits when you're done. Fourth, emphasize the incremental risks of a given event or choice distinctly from baseline risks that a person would confront in daily life, no matter what. If you tell someone that people taking a given drug have a 50 percent risk of dizziness, it's natural for the person to believe that the drug is responsible for the entire 50 percent risk. It's important to separate the incremental risk from the baseline risk, otherwise, people will naturally overestimate risks and avoid making decisions that expose themselves to that risk. They may avoid medical treatments, they may avoid sensible organizational decisions. We could say that an organization has 50 percent chance of decreased sales this quarter if it implements a new training system. The wisdom of implementing the training is quite different if the organization has 45 percent chance of decreased sales, even if it doesn't implement the new training. Finally, when presenting information about the likelihood of an event, it's often more understandable to others if you present the information as a frequency rather than a percentage. I'll go into more detail on this point in the next session and discuss it briefly here. This point means that you should consider saying five out of 10 people instead of 50 percent of people. Yes, these figures are mathematically equivalent. Five is 50 percent of 10, but they're not computationally equivalent, one is easier for people to grasp. Two researchers gave the extreme example of dividing one number by another. It's easier to do if you present the numbers in Arabic numerals like we're accustomed to, than in Roman numerals. Sure, the two systems represent the same thing, but one is easier to understand. In one medical study, patients evaluated the risk of getting a rash as a side effect to a medicine, either as 10 percent or 10 out of 100 people. Patients with lower numerical skills perceive them medicine to be less risky when they viewed a percentage instead of a frequency. This isn't just because people had difficulty understanding percentages, they also believed that percentages are more abstract and less severe, less concrete than literally thinking about a group of 100 other people experiencing side effects. In conclusion, this session provided some guidance about communicating risk information in ways that are easier for people to understand. Use visuals whenever possible, avoid jargon, pay attention to order effects, especially presenting risks last, separate incremental and baseline risks, finally, consider using frequencies instead of percentages when communicating likelihood.