Catastrophe Amnesia

"Those who cannot remember the past are condemned to repeat it." – George Santayana

Large flat stones, some as tall as 10 feet, can be seen along Japan's northeastern shore, which was heavily affected by a magnitude-9.0 earthquake and tsunami on March 11, 2011, resulting in nearly 29,000 deaths or disappearances.

“The tsunami stones are warnings across generations, telling descendants to avoid the same suffering of their ancestors,” said Itoko Kitahara, a specialist in the history of natural disasters at Ritsumeikan University in Kyoto. “Some places heeded these lessons of the past, but many didn’t.”

Japan has a long history of devastating earthquakes and tsunamis. In 1707, a tsunami caused by the Hōei earthquake killed more than 5,000 people. The Great Yaeyama Tsunami of 1771 killed 8,439 people on Ishigaki Island and 2,548 more on Miyako. In 1896, the Sanriku earthquake sent two tsunamis crashing into coastal settlements, destroying some 9,000 homes and killing at least 22,000. This list doesn’t include several other major earthquakes that have occurred inland without triggering a tsunami. Yet, the country remained vulnerable to the devastating 2011 earthquake and tsunami, not only leading to 29,000 deaths, but also a nuclear plant disaster, that could have easily gotten much worse. If this experience can’t be put down to collective amnesia, what is it?

Previously, in our series on Cognitive Biases in Risky Decisions , we explored Disaster Myopia . In this article, we dive into Amnesia, a bias that is closely intertwined with Myopia. Where the Myopia bias reveals a tendency to underestimate the future consequences of high impact events, Amnesia biases the judgment by causing us to forget the experiences and lessons of historical events. It is possibly the reason for the truth in the old adage that “history often repeats itself”. Amnesia bias is caused because individuals, decision makers and institutions tend to have short memories.

Amnesia is the reason why individuals continue to build coastal property in towns that have already been flooded by storm surge. For instance, as this article points out, the threat of rising sea levels and flooding, even after recent near-misses, like Hurricane Ike, is not slowing life or development in Galveston. Not only is there rampant development in disaster prone areas, but also a lack of long-term memory of the financial risk it entails. Studies have also shown that shortly after being hit by a hurricane, flood insurance demand increases temporarily (7%) only to return to the previously low baseline within three years.

Bubbles and Market Cycles

Amnesia also impacts risk seeking behavior in the broader financial markets and insurance underwriting. In financial markets, people continue to invest in bull markets even though there may be ample historical evidence for stock market bubbles and crashes every so often. Similarly, it is possibly one of the main reasons for the “hard” and “soft” cycle phenomena in the insurance industry. Shortly after a significant catastrophe event, insurance rates harden and policy terms and conditions tighten. However, hard markets are notorious for their short half-life - most don’t last for more than a couple of years. Very quickly, new capital flows into the insurance industry in search of attractive returns, increases competition, and, consequently, insurance rates soften and policy terms and conditions relax to pre-event levels. And, the cycle continues.

Why is Amnesia so prevalent?

The issue of Amnesia when it comes to rare events and risky decisions has perplexed behavioral psychologists and economists for a long time. We dive into several factors that are in play in this section.

Blame Evolution

Any serious introspection into a deep-seated habit or bias eventually points to evolution! Human beings are naturally wired to have a short-term memory of pain. When a child is learning to walk, every time she falls, it causes a little bit of pain. More importantly, the pain provides valuable feedback on how to do it better in the next try. This creates a virtuous cycle and increases the amount of positive experience that comes with the gradual success in slowly being able to walk. If we were naturally predisposed to long term memory of pain, we wouldn’t push through the pain episodes towards a successful outcome.

Unfortunately, when it comes to rare (low probability) events, this tendency to have a short-term memory towards pain is counterproductive, since it doesn’t spur us into preventative actions. By nature, these events are few and far between, and often may occur once in a lifetime or generation. In contrast to our natural tendency, what we need is the ability to have long term memory of the potential pain that these events may cause, if we don’t proactively mitigate our risk. ?

Emotions Rule Facts

An important aspect of amnesia bias is that it is not a result of “factual” memories of past events being erased from our minds. The historical facts of past events may well be codified, and there are often great reminders of the same, especially in the form of monuments, literature, and other media. The Tsunami stones from Japan being a classic example. Unfortunately, factual memory is not enough to stir action. It is the “emotional” part of the experience that is lost with time - the emotional intensity is what spurs the action. The hedonic impact of past losses, the acute sense of tragedy one feels when one’s house is destroyed, fades with time. This in turn relaxes the attitude towards investing in protection. This is the reason why, just three years after Hurricane Katrina, flood insurance take up rates in New Orleans were back to pre-Katrina levels . It is also the reason why, in the summer of 2001, many investors lined up to invest in the World Trade Center, and many insurers provided ample Terrorism risk capacity to the towers within their regular property ‘all–risk’ policies, even though, less than a decade earlier (1993 ), the same building had already been bombed by middle eastern terrorists with a massive explosion creating a 100 feet crater several stories deep.

A Negative Feedback Loop

Another reason why past events lose their influence on decisions is the lack of positive reinforcement when protective action is taken. Since these are rare events, it is highly likely that often, protective action, whether an insurance or a loss prevention investment, will be perceived as a wasted expense. Especially so when the investment is compared to other uses for the same funds that result in more immediate gratification. After all, it is difficult to convince someone that the best outcome from an insurance policy is not to have a claim.

The cost and effort of engaging in preventative action during a near miss scenario or a false alarm scenario can also sometimes result in risk seeking behavior. For instance, an individual having put up storm shutters considering a hurricane warning which eventually doesn’t impact his house, will tend to remember the effort spent in putting up and removing the shutters. This memory will then discourage him from following the same course of action in the next warning.

This concept also applies to a professional risk manager. Given the inherent uncertainty, disaster warnings cover a much larger area than the ultimately impacted area. A risk manager is faced with a decision to call for preventive action, knows well that it is more than likely to be a false alarm, and they might be blamed for ‘crying wolf’, especially in the wake of a recent false alarm. This might create hesitation which indirectly increases the risk.

Money, Kisses and Electric Shocks

Psychologists at the University of Chicago published an article with an attractive title of “Money, Kisses and Electric Shocks: On Affective Psychology of Risk ”. Their finding was that the valuation of gambles was much less sensitive to probability when the outcomes were emotional (“meeting and kissing your favorite movie star” or “getting a painful or dangerous electric shock”) than when the outcomes were just loss or gains of cash. Further studies have since refined this finding to conclude that adding a vivid description to choose disrupts the inherent short-term probability assessment – which myopia might bias towards neglect.

A related and relevant distinction is also made between ‘choice from experience’ vs ‘choice from description’. Ralph Hertwing, a pioneer in the Description-Experience framework , distinguishes between two powerful but imperfect teachers of risk. First, people may learn by consulting symbolic and descriptive material, such as warnings, statistics, and images. Second, people may learn about risks through personal experience. Responses to risk can differ systematically depending on whether people learn through one mode, both, or neither. One reason for these differences is the discrepancy in the cognitive impact that rare events have on the decision maker. It so happens that ‘experience’ dominates ‘description’ – the degree of this disconnect is so prevalent that it even has a name, the ‘description-experience gap’! For ‘rare’ events, this gap is largest in favor of ‘choice of experience’, which typically favors underestimation of risk (or neglect). In fact, it is commonly thought that descriptive warnings are often ignored and may even backfire.

Why does ‘choice from experience’ underweight risk? Well, one of the reasons is that most individuals have not directly experienced a ‘rare’ event. Under weighting is most pronounced when one has limited experience because a rare event is unlikely to arise in a small sample. However, even if the rare event has been experienced, the emotional intensity of their experience fades with time, particularly in relation to other cognitive choices/emotions one must negotiate at the decision time.

Mitigating (or Overcoming) Amnesia

Overcoming amnesia may seem like a losing battle, especially given the evolutionary and competing cognitive tendencies. However, behavioral scientists have been exploring several interesting ideas to counter amnesia, and we will go over them below.

Counter Evolutionary Amnesia

To balance our evolutionary predisposition for short-term memory of pain, modern societies could integrate continuous educational programs and regular drills that simulate disaster scenarios without causing real harm. This continuous exposure helps reinforce the memory of potential consequences, akin to muscle memory in athletes, ensuring that preparedness remains top of mind even in the absence of immediate threats.

Don’t Neglect Emotions

As discussed above, the “emotional intensity” of the experience is almost as important, if not more, as the factual record. Of course, it is not possible to capture and preserve the emotional intensity of someone else’s experience over the decades. However, it would help to cast the past disaster in the current setting through relevant experiential stories for the decision makers, so that they can visualize the tangible and feel the intangible impacts of the experience.

One recent study , conducted after Japan’s 2011 Tohoku earthquake, aims at understanding how the transmission of local knowledge and experiences of past disasters contributes to increasing community resilience?to tsunami risks. It posits that disaster memorials are more likely to influence?community resilience if passed down using a combination of oral transmission and interpretation together with the use of backup memory archives and physical objects. Such practices are recommended as they allow both the communicability and durability of lessons-learned. Transmitting and maintaining a constant interaction between these two forms of memory is the responsibility of those institutions that oversee the keeping alive of memories, by continuously repurposing and renewing them while engaging with exposed communities.

Simulated Experience

As we have learned above, ‘descriptive’ knowledge and warnings are not as effective as experience. A vivid emotional experience is one of the ways to influence ‘choice from experience’ to be more aligned with the actual risk of rare events. Experience can make things feel more real or foster better insight than description, and it may be worth harnessing this quality in risk communication. Indeed, inspired by work on the description–experience gap, researchers have recently begun to investigate the potential benefits of “simulated experience” for communicating statistical information about risks. For example, financial institutions that offer investment products are by law required to provide information about the key properties of those products (e.g., risks, costs, past performance history). This information can be presented to clients in different ways. The default approach is in terms of numerical and graphic descriptions (e.g., historical returns in fact sheets). A very different approach is for clients to interactively sample possible outcomes, that is, possible returns on an investment—what?Kaufmann et al. (2013) ?called “experience sampling.”

Similarly, for natural disasters, there are various avenues to use virtual reality to create an immersive experience of living through a disaster. Some examples, not an exhaustive list, include:

1. Virtual Flood
2. The Quake Museum in Lisbon, Portugal
3. The Shake House at the California Academy of Sciences

The potential of simulated experiences to enhance disaster preparedness is vast. For instance, virtual reality simulations of earthquakes or floods can provide an immersive learning environment that emotionally engages the participant. Studies have shown that individuals who undergo such simulations are more likely to take proactive measures, such as preparing emergency kits or learning evacuation routes. Expanding access to these technologies in schools, communities, and through mobile applications could mark a significant step forward in our collective resilience against disasters.

Positive reinforcement

One way to counter the natural negative feedback loop involved in preventative actions is to introduce a positive reinforcement to encourage smarter decisions considering the amnesia bias. Some examples include,

1. Communities can institute recognition or commendations for individuals and risk managers who demonstrate proactive risk-averse behavior. This can be done both post disaster, but also, post a false alarm or a near miss.
2. Insurers, regulators, and policymakers can create a recognition system to provide consumers and businesses with a required positive boost in claim-free years. Subject to any applicable regulatory considerations, perhaps a “no claims” bonus or credit could also be considered.
3. In financial firms, internal incentive and risk management structures need to align with risk averse behavior. Risk professionals and underwriters need to manage their profit margins not by the top line, but to a “normalized” objective risk adjusted bottom line.

While the strategies outlined above offer promising paths to mitigate catastrophe amnesia, they are not without challenges. Implementing widespread simulated experiences, for instance, requires significant investment in technology and content development. Additionally, positive reinforcement mechanisms must be carefully designed to avoid unintended consequences, such as incentivizing risk-taking behavior. Addressing these concerns requires a collaborative approach, involving stakeholders from various sectors to share insights, resources, and responsibilities.

Conclusion

As we navigate the complexities of preparing for the unexpected, it's crucial that each of us—policymakers, educators, and individuals—reflect on our role in combating catastrophe amnesia. By fostering an environment where lessons from past disasters are continuously learned and re-learned, we can build a more resilient society. Let us commit to adopting and advocating for strategies that ensure the hard-won lessons of yesterday do not become the forgotten warnings of tomorrow.