Memory and Imagination in Understanding and Managing Risk

By Peter Ho


The COVID-19 pandemic belongs to a category of risk that is called “dread risk”. Professor Gerd Gigerenzer, a German psychologist who studies risk, describes dread risk as a low probability, high-damage event. Dread risk is often used in reference to natural hazards and threats to the environment or health, such as nuclear power.

The Fukushima nuclear disaster demonstrates aspects of dread risk. The huge Tohoku earthquake in Japan generated a massive tsunami that breached the defences of the Fukushima nuclear power plant, leading to meltdowns that leaked radiation into the atmosphere. Although the earthquake and tsunami combined killed around 18,500 people, the Fukushima nuclear disaster itself produced no direct fatalities.[1] Instead, about 600 died as a result of evacuation procedures and stress-induced factors. But Fukushima was perceived to have the dreadful potential to kill a lot of people. It was this accident rather than the earthquake that captured the imagination and dominated the news around the world for weeks, creating widespread anxiety. For example, news that potassium iodide pills could help prevent radiation-induced thyroid cancer sparked a run on pharmacy supplies in the United States, thousands of miles away from the disaster.[2] More significantly, the incident led to Germany foreswearing the use of nuclear power.[3]

Even though dread risks like the Fukushima nuclear disaster might actually cause fewer deaths than other risks that we happily live with — such as obesity, heart disease or even crossing the road — they capture the attention of the media, stoke anxiety and make people excessively fearful. That fear can change our behaviour in ways that actually increase our risk of injury or death. For example, people are now avoiding visits to hospitals because they fear catching COVID-19. According to a survey by the World Stroke Organization of a hundred countries, hospital admissions for stroke symptoms had a median decrease of 50 to 70 per cent in the first months of the pandemic, compared with the same period in 2019.[4]

Fear can change our behaviour in ways that actually increase our risk of injury or death.


What makes a dread risk different from other risks? A key difference is the perception that a dread risk can cause many people to die at the same time. When threats are new and dramatic, such as the 9/11 attacks with their visceral images of passenger jets crashing into the Twin Towers of the World Trade Center in New York and the Pentagon in Washington DC, they create a strong emotional impact that skews our perception of how dangerous they are. People will also misjudge the likelihood of that catastrophe recurring. This is the availability heuristic, the tendency to over-estimate the likelihood of events with greater “availability” in memory, which can be influenced by how recent the memories are, or how unusual or emotionally charged they may be. So, after a shocking event like 9/11, we will think that another terrorist attack is a more probable risk than something else, simply because it is fresh in our minds.

In an oft-cited study, Professor Gigerenzer found that in the months after 9/11, more people chose to drive rather than to fly, feeling it was safer. Passenger miles on the main US airlines plunged between 12 per cent and 20 per cent over a three month period, while road use jumped.[5] The change is widely believed to have been caused by worried people opting to drive rather than fly. But the reality is that travelling long distances by car is actually more dangerous than travelling the same distance by aeroplane. Gigerenzer estimated that an extra 1,600 Americans died in car accidents in the year after the 9/11 attacks — indirect victims of the tragedy.[6]


Since the outbreak of the COVID-19 pandemic, we have been deluged by alarming reports in the media. For the layman, it is hard to make sense of the risk that COVID-19 poses. This is in comparison to single events like plane crashes and natural disasters, for which we have learnt to weigh the risks, and even to become inured to them. Professor David Spiegelhalter, the well-known British statistician, suggests that to meaningfully understand the risk of COVID-19, we should compare it to the risk of dying the next year — which is our annual death risk. Catching COVID-19 doubles your annual death risk, which is very low if you are young. But because our annual death risk rises exponentially from the age of ten, the COVID-19 risk factor increases significantly the older you are.[7]

Furthermore, even if the personal risk is low, you may spread the infection to other more vulnerable people, such as seniors and those who cannot be vaccinated due to medical reasons. That is why we need to look beyond individual risk, which actually can be low if you are young. Instead, we should think about the collective risks, which are high. This is because infectious diseases like COVID-19 spread and multiply in ways that other individual risks, like those of car crashes or heart attacks, do not.

That pandemics pose huge collective risks ought to surprise no one. Throughout history, humanity has confronted pandemics with devastating consequences. The Black Death in the 14th century is thought to have killed between 30 per cent to 60 per cent of Europe’s population. Scaled up as a percentage of the world population today, that would reach nearly four billion deaths. The Spanish Flu of 1918 to 1920 infected half a billion people and killed perhaps a tenth of them. Between one and four million people died as a result of the Hong Kong flu pandemic that lasted between 1968 and 1969.[8] A novel outbreak like SARS in 2003 can subside fairly quickly. But others like COVID-19 could rapidly get out of hand and turn into a real existential threat. Although our response to such dread risks might seem like an overreaction on an individual level, it can make sense from the perspective of survival of a city, a country and even the human species. It is not unreasonable to speculate that dread risk may be an evolutionary impulse. Early man lived in small hunter-gatherer tribes. The simultaneous death of a large number of members in a tribe would have threatened its very existence — so actions that entailed that risk, however small, were especially dreaded and to be avoided.

Most people are more concerned with problems which appear to impact daily life rather than long-term problems that may affect future generations.


All risk, and not just dread risk, is very much a psychological construct. It is the subjective judgement that people and even whole societies make about the characteristics and severity of a risk. However, most people are more concerned with problems which appear to impact daily life rather than long-term problems that may affect future generations, such as climate change. In his book Collapse, the American polymath, Jared Diamond, explains the process by which a major change can be accepted as normal and acceptable if it happens slowly through small, often unnoticeable, changes, not unlike the proverbial frog in boiling water. Diamond calls this creeping normality.[9] Most people have no direct experience of climate change, or do not notice its impact because the changes appear imperceptible and occur over a long period. As a result, they adopt a wait-and-see attitude, unwilling to change environmentally destructive behaviours even when experts provide detailed and clear risks caused by climate change.


Risk is also a social construct. Perceptions of risk are not just held by individuals. They are also constructed by society through its institutions, cultural values, and experience. The tighter the binding of society, the more individuals are connected by feelings of belonging or solidarity. This includes attitudes to risk. The basic idea is that collectivism, putting the common good ahead of individual freedoms, may make people more conscientious about avoiding the behaviours that could cause dread risk. This is one reason why countries in Asia, where collectivism has deeper roots, or New Zealand where community ethos is strong, have seen greater success in tackling COVID-19.[10] In contrast, the emphasis on individual rights and freedoms has seen people in many western countries push back when their governments require them to wear masks or stay at home.

Not surprisingly, the spirit of collectivism and the social bonds that exist within small tribes are much stronger than those among residents in larger societies. In 2004, I had the privilege of visiting Port Blair, the capital of the Andaman and Nicobar Islands, an island chain in the Indian Ocean with unique flora and fauna. Within that ecosystem is North Sentinel Island. I learnt that it is a tiny island occupied by a small Stone Age tribe that is supposed to indulge in cannibalism. The authorities, for good reason, steer clear of the island.

Not long after my visit, the 2004 Boxing Day tsunami washed through the Andamans, and swept over North Sentinel Island. An Indian friend told me that some days later, the military flew a helicopter over the island without landing — a prudent precaution against a close encounter with a hungry tribe — and discovered to their surprise that the tribe had survived the catastrophe that had killed two thousand in the Andamans.

It turns out that they were not alone. Another tribe on Little Andaman also fled to higher ground before the tsunami struck. These small tribes appear to have retained a collective memory from their ancestors of natural disasters, developed an ability to read the warning signs of an impending tsunami — perhaps the waters turning a different colour, or the birds and animals getting disturbed — and then to flee to the high point at the centre of the island. The knowledge of these warning signs seems to be incorporated into an oral tradition. Importantly, the tribal system provided the social structure to transmit these oral traditions from generation to generation.[11] In today’s modern world, many of these traditions have either been abandoned or ignored, dismissed as either irrelevant or archaic. But they developed for a reason — to avoid dread risk — and endured for so long in memory because they serve an existential purpose. Such traditions are therefore time-tested heuristics that rely on deeply embedded societal memory in order to survive in a sometimes dangerous and uncertain world.

Traditions are time-tested heuristics that rely on deeply embedded societal memory in order to survive in a sometimes dangerous and uncertain world.


The importance of memory in managing risk is illustrated — once again — in the Fukushima nuclear disaster. In seismically active Japan, earthquakes are a very real and significant risk to nuclear power plants. In 2007, an offshore earthquake caused a fire to break out at the world’s largest nuclear power plant, Kashiwazaki-Kariwa. With this recent event fresh in mind, Japan’s Nuclear and Industrial Safety Agency, or NISA, met to discuss the safety needs of each of the country’s 17 nuclear power plants. In reviewing the safety guidelines for Fukushima, the panel used data from the largest earthquake recorded in the area, one that dated back to 1938. That earthquake had caused a small tsunami. As Fukushima is near the sea, it was decided to build a seawall high enough to stop a tsunami from flooding the reactor complex. The height of the seawall was determined by the tsunami of 1938.

Nevertheless, a seismologist at the NISA meeting felt that the 1938 tsunami was inadequate to serve as a basis for the Fukushima guidelines. He pointed out that there had been a much larger earthquake in the region. It had occurred more than a thousand years earlier, in 869. The tsunami it generated led to extensive flooding of the Sendai Plain. But obviously, because it happened so long ago, the data for this earthquake depended more on archaeological evidence than on the precise information available for the 2007 earthquake. In a strong manifestation of the availability heuristic, the NISA panel decided that the 869 earthquake and tsunami were simply too “historical” and therefore would not be incorporated into the model for updating the Fukushima safety guidelines. This decision to focus on a short time horizon proved to be a costly mistake. In 2011, the tsunami unleashed by the Tohoku earthquake inundated an area that corresponded very closely to the 869 tsunami.[12]

We often think of earthquakes in spatial terms: how far away we are from fault lines, or from the Ring of Fire, or the proximity of our homes to the sea in areas vulnerable to tsunamis. But earthquakes and tsunamis also present us with temporal challenges. Ancient earthquakes and tsunamis are hidden from us because we do not look far enough backwards into the past. Ignoring the occurrence of an earthquake that occurred more than 1000 years ago does not reflect a lack of memory. The earth is 4.5 billion years old, but we are a young species. Kathryn Schulz of The New Yorker elegantly explains that “the brevity of our lives breeds a kind of temporal parochialism — an ignorance of, or an indifference to, those planetary gears which turn more slowly than our own.”[13] In contrast, the tribes of North Sentinel Island and Little Andaman preserved ancestral memory of catastrophe, reflecting a collectivist determination to avoid dread risk. Collectively and across generations, they were able to overcome the temporal parochialism and survive the Boxing Day tsunami.

“The brevity of our lives breeds a kind of temporal parochialism — an ignorance of, or an indifference to, those planetary gears which turn more slowly than our own.” — Kathryn Schulz of The New Yorker


This is actually not a problem of information. We also have a lot of archaeological evidence of when and where ancient earthquakes occurred. Kathryn Schulz also wrote that “we excel at imagining future scenarios, including awful ones. But such apocalyptic visions are a form of escapism, not a moral summons, and still less a plan of action. Where we stumble is in conjuring up grim futures in a way that helps us to avert them.”[14] Indeed, many such grim futures, or dread risks — natural disasters, pandemics, climate change — can often be assigned probabilities. This ought to lead the authorities to take precautionary measures. But they often do not. The author Margaret Heffernan called this wilful blindness.[15]

The reasons why they do not are several. In the first place, leaders often have a hard time properly discounting the present value of events that will take place in the future. This tendency to place less emphasis on future risks and contingencies, and instead to place more weight on present costs and benefits is a common cognitive bias known as hyperbolic discounting. Many, if not all, governments indulge in it. One example of hyperbolic discounting at work is climate change. Governments understand the theoretical need to consider the effects of global warming on future generations. But instead of spending time worrying about a problem that will — hopefully — occur only after they leave office, they tend to place greater emphasis on the current costs of mitigation and adaptation.

The reasons why are deeply embedded in our human nature. This leads me to the black elephant. What is the black elephant? The black elephant is a cross between a black swan and the elephant in the room. The black elephant is a risk that is actually visible to everyone — the proverbial elephant in the room — but no one wants to deal with it, and so they pretend it is not there. When the risk actually happens, we all feign surprise and shock, behaving as if it were a black swan.

The Thai floods of 2011 were an example of a black elephant. Everyone knows that floods occur frequently during the wet season along the low-lying Chao Phraya river system. But few paid much attention to the full downside potential of the flood risk in Thailand. Before 2011, it would have taken a big leap of imagination to connect flooding in Thailand to major disruptions of global supply chains. Perhaps, floods occurred with such regularity that everyone was inured to them. So a little more rain than the historical maximum in the northern region, compounded by man-made factors such deforestation, led to the massive floods in 2011. As a result, major industrial parks located outside Bangkok were covered by floodwaters for weeks. Production ground to a halt, and global supply chains were disrupted. The World Bank estimated the economic cost of the Thai floods at US$46 billion.[16] Ironically, for Japan’s big three property casualty insurers, the Thai floods proved even more costly than the Tohoku earthquake.

Much of our reluctance to grapple with such risks stems from an unwillingness to face the consequences of an uncertain and unpredictable future. These consequences interfere with long-held mental models — and business or self-interest — creating cognitive dissonance. Cognitive dissonance leads to many organisations — including governments — to underestimate risks, ignoring warning signs of impending crisis, and taking decisive action only when the crisis unfolds. This is the mother lode of black elephants. How can we limit or counter the influence of such bias? Obviously, the occurrence of a crisis that radically alters our mental models is one corrective. The SARS crisis of 2003 forced the Singapore government — as well as governments in China, Taiwan, Hong Kong and Vietnam — to take more deliberate steps to prepare for future pandemics. SARS corrected our biases, making us realise that the risks and costs of a pandemic were not trivial, and increased our alertness to the onset of another pandemic. Without SARS, it is difficult to imagine that our subsequent responses to COVID-19 would have been as aggressive and proactive as they have been. Contrast these responses with the initial lack of urgency in other countries which were largely unaffected by SARS. Identification, management and communication of risk must take into account this human tendency to underestimate or overestimate risk.

Cognitive dissonance leads to many organisations — including governments — to underestimate risks.


Memories of unpleasant or even catastrophic events drive our attitudes to risk. This is a natural defence mechanism. But memory is not perfect, and is often impaired by our cognitive biases, and by limits to our time horizons. While imagination can drive us to construct fanciful risks, our cognitive limitations such as creeping normality prevent us from translating these into proper risk assessments and then converting them into appropriate contingency plans. To understand and manage risks goes well beyond identifying what they are. It also depends on an awareness of the importance as well as limits of memory and imagination, and other psychological and social factors that shape attitudes towards risk.


Peter Ho is Senior Advisor at the Centre for Strategic Futures.

This post is adapted from his speech at the 2020 Understanding Risk Forum on 2 December 2020.

The views expressed in this blog are those of the authors and do not reflect the official position of the Centre for Strategic Futures or any agency of the Government of Singapore.


1. Britannica, “Japan Earthquake and Tsunami of 2011”, accessed 4 December 2021,

2. Larry Greenemeir, “Does Potassium Iodide Protect People from Radiation Leaks?”, Scientific American, 15 March 2011, accessed 4 December 2021,

3. “How Fukushima Triggered Germany’s Nuclear Phaseout”, Deutsche Welle, 10 March 2021, accessed 4 December 2021,,historic%20turning%20point%20for%20environmentalism.&text=On%20March%2011%2C%202011%2C%20one,tsunami%20off%20Japan's%20Pacific%20coast

4. World Stroke Organization, “The Global Impact of COVID-19 on Stroke — Survey Report from Prof. Marc Fischer, WSO President-Elect”, accessed 4 May 2020,

5. Sharon Begley, “Afraid to Fly After 9/11, Some Took a Bigger Risk — In Cars”, The Wall Street Journal, 23 March 2004

6. Max Plank Gesellschaft, “More traffic deaths in wake of 9/11”, accessed 4 December 2021,

7. Jemima Kelly, “Covid kills, but do we overestimate the risk?”, Financial Times, 20 November 2020

8. Michael S. Rosenwald, “History’s deadliest pandemic, from ancient Rome to modern America”, Washington Post, 7 April 2020

9. Jared Diamond, Collapse: How Societies Choose to Fail or Succeed (New York:Viking, 2005)

10. Danson Cheong, “Putting common good ahead of individual freedoms have helped in Asian countries’ fight against Covid-19, says Chan Heng Chee”, The Straits Times, 18 June 2020

11. Subir Bhaumik, “Tsunami folklore ‘save islanders’”, BBC News, 20 January 2005, accessed 4 December 2021,

12. Richard A. Clarke et al., Warnings: Finding Cassandras to stop catastrophes (New York: Ecco, 2017)

13. Kathryn Schulz, “The Really Big One”, The New Yorker, 13 July 2015, accessed 4 December 2021,

14. Ibid.

15. Margaret Heffernan, Willful blindess: Why we ignore the obvious at our peril (New York: Walker, 2011)

16. The World Bank, “The World Bank Supports Thailand’s Post-Floods Recovery Effort”, accessed 4 December 2021,



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Centre for Strategic Futures

Welcome to CSF Singapore’s blog site, a space to share our shorter think-pieces and reflections. Visit our main website at