Is Trust is Overrated in Risk Perception? Some New Thinking.

Is it time to throw out the 'Trust Deficit Model' when we consider human responses to risk?

A thoughtful article on the way we think about trust in risk management was just published in Risk Analysis by Angela Bearth and Michael Siegrist at ETH Zurich. It is a refreshing perspective on some go-to concepts in risk perception, including the 1) 'Information Deficit Model' , 2) 'Trust Deficit Model' , and 3) the role 'Echo Chambers' on social media shaping risk (and benefit) perception.

Let's take a look at each concept in turn, and then explore what's new. And I will also try to slowly walk back the clickbait title of this newsletter, so I have at least a little bit of credibility left at the end of this newsletter.

This post is a little nerdier, but bear with me - you will like it. (Angela, Michael: If you read this, I do beg your forgiveness for what I am doing to the arguments you are making in your really interesting article).

"But but but: Just look at my Excel sheet!!!" - Every risk expert at every board meeting or public hearing, everywhere, always

There is as fundamental and fascinating problem - maybe better: class of problems - in risk management: Why doesn't everyone just agree with the experts (that is usually: you and me), and move on with their lives?

Risk Analysis just published a fantastic Special Issue on the 30 year anniversary of the Social Amplification of Risk Framework (SARF), pioneered by Roger Kaplan in the 1980ies. SARF is one of the more prominent sense making frameworks to explain why people's reaction to risk are not confined to carefully studying expert publications and then calmly making up their mind (you laugh - but we technocrats believe that is how the world ought to work!). There is a lot going on in SARF (and it surely deserves its own newsletter), but I want to pick on a few specific aspects that Angela and Michael discussed in their article: Information, Trust, and Social Media.

What could possibly go wrong???

1) The Information Deficit Model: People are just too dumb.

'Information Deficit Model' is a polite way of saying: The reason anyone disagrees with my opinion is because they are ill informed, lack education and experience, and/or are just generally a little too dumb for this.

This is many expert's default attitude when communicating their expert findings: If I only present all the complicated factors well enough that have allowed me to reach my conclusion, you have no other option but to agree with my risk assessment. We drag a ton of PowerPoint slides into the meeting, or better even: hundreds of pages of reports and Excel sheets. We show probability distributions, talk about P80, and brag about the number of iterations and independent variables in our Monte Carlo Simulation.

I like to think that sometimes, this works. But good luck to you. Part of the motivation for the Social Amplification of Risk Framework was the insight that really, what matters more in a stakeholders risk perception is their level of trust in the expert - not the number of rows in their Excel table. Even if presented with "correct" information, many studies have shown that people's emotional response - and their personal decision making - will not necessarily change accordingly.

2) Trust Deficit Model: People just trust the wrong people.

This brings us to the 'Trust Deficit Model'. As you may have guessed correctly, the problem of anyone disagreeing with my expert risk assessment is now no longer that they are dumb and ill-informed, but that they do not trust me.

Trust, in all its forms, is generally considered a moderator in risk perception. Over the years, we have done a number of studies that strengthen this point. If I trust a person or organization, and they tell me the risk is acceptable and worth it, I go along with them. If I do not trust them, then I will assume the opposite - they are trying to trick me, and I should most definitely oppose their recommendations.

Trust has also a strong effect on our perceived fairness of a process or a decision, and by extension, on the legitimacy of other stakeholders and the decision makers. All of those factors, in turn, also influence risk and benefit perception (the link between perceived risk and perceived fairness is particularly strong).

Sometimes, the effect is quite dramatic: I do not trust the health authorities, so I will not get vaccinated. I trust my internet forums, so I will panic over a nuclear accident somewhere on the other side of the planet.

In the every day life of a risk manager, it is a little less dramatic: When you present your report, your CEO will look at you and ask: "What do you think Josef, should we do it?" - They care about your analysis, because they trust you. Or not: "This is all fine and well Josef, but it's just too risky." - Same report, different levels of trust.

The Trust Deficit Model is a convenient, intuitively convincing explanation of why people follow expert recommendations - or not.

3) Social Media Echo Chambers: We Live in a Post-Truth and Post-Trust Society

With the Information and Trust Deficit Models in mind, it becomes obvious why we should be so concerned about the current use and abuse of social media: If we do live in a world of Post-Truth and Post-Trust, how are we ever going to come together as a society and talk about fun subjects such as, say, maybe, just maybe, nuclear playing a role in saving the climate (and species and economies and electric cars and heating)? Or maybe a little closer to home: What are appropriate pharmaceutical (i.e. vaccinations) and non-pharmaceutical interventions (i.e. masks and lock-downs) in a public health crisis?

Given the reality we live in today, the outlook seems rather bleak: We have fragmented into our personal social media echo chambers. We no longer share neither a common information base, nor do we trust one another.

Or, do we?

Spoiler Alert: Angela and Michael think there is a way out (and that things are not as bad as they may seem).

Rethinking Information, Trust and the Role of Social Media for Risk Communication

So getting back to the paper , it raises a number of interesting observations. I am grabbing a few (go read the paper for more):

• First, the 'trust deficit model' also suffers from some problematic notions - the notion that it is clear what the 'correct' level of risk perception is, and that it is clear that the person with that answer can be trusted. As a Technocrat, my personal answer is of course: "What please is problematic here? (As long as you ask me.)"
• Secondly, they describe how the 'Trust Deficit Model' also works the other way around - after Fukushima, decision and policy makers adjusted their public messaging around their support for nuclear power. Not because the risk of tsunamis had suddenly increased in Germany, but because a continued support would have led to an erosion of the public's trust in them (and an erosion of their chances in the upcoming general election).

As Angela and Michael put it in their paper: "[...] the mere perception of a potential loss in trust might lead to important implications for decision making [...]"

• Third, they summarize that there is no empirical evidence that we are actually facing a "distrust spiral" of continuously eroding trust in science and official organizations (or the media, for that matter). Fluctuations, yes, but no continuous decline (some politicians individual fate notwithstanding).
• Fourth, instead of fearing that people trust the "wrong" people and subsequently worry about the "wrong" risks, we should move risk communication into the 21st century and - believe it or not - start using social media. Let me spend a little more time on this point and the reasons behind it (even if LinkedIn does not allow me second level bullet points):
• 90% of people consume social media without creating (1%) or editing / modifying (9%) content. That means that whatever you see is likely to be based on the most extreme and motivated 1% of the population. Just because that dominates your online news stream is not an excuse to refuse to offer your viewpoint. Also, it is not evidence that everyone has lost touch with reality (information deficit) or trust in science (trust deficit).
• Sharing your information does NOT mean engaging with the loudest 10% that are already there. It means offering the other 90% that passively consume content an opportunity to understand your viewpoint. Those are the 90% that likely trust experts and institutions, and are looking for information.
• Additionally, your absence from social media can easily be interpreted as being "out of touch" - so just like in Kindergarden, there are points being awarded for showing up.

So, should we throw out the 'Trust Deficit Model'?

Coming back to my original question, the answer is a clear "Yes and No". Trust is not overrated, but it is sometimes misunderstood.

Yes, we should throw out the notion that we live in a post-trust (or post-truth) society. Some few percent do, but the majority of people still trust experts and institutions.

No, because we should keep building trust. But that means that we do need to communicate more openly (or better: "more accessible") on social media.

And last but not least: Trust arrives on foot and leaves on horseback. Earning it is hard, and losing it is easy. That may explain why, as discussed above, we also make decisions based on expected loss of trust, not on perceived risks.

If you made it this far - thank you - for your trust in me that reading the newsletter would be a good use of your time! ;-)

Liked it? Share it with your friends, click 'Like' to help others discover it, and leave a comment below to let me know what you think!

Until next month,

Josef.