Eye of the Beholder: Understanding the Psychology of Risk Perception to Improve Risk Management

This post first appeared on Risk Management Magazine. Read the original article.

Individuals perceive risks in markedly different ways. One person may consider a risk to be critical, while another could consider it inconsequential. Often rooted in psychology, these differences in risk perception can create challenges for risk professionals, especially when designing and implementing an effective risk management program. After all, if the program is focused on the wrong risks from the outset, the consequences to the organization could be dire. It is therefore critical for risk professionals to understand the psychological aspects of risk perception and develop techniques to address the resulting challenges.

The Effect of Heuristics and Biases

Our reaction to risks can be traced to early humans, who
were either the hunter or the hunted and responded to danger by fighting or
fleeing. Our ancestors’ survival depended on responding quickly and correctly.
A part of the brain—the amygdala—helped humans’ survival by bypassing cognitive
processes and initiating immediate responses. Today, we generally have the time
to obtain information, analyze risks and develop a reasoned response. Yet we
still seem to let our innate reactive-mode override our cognitive thinking.

Many factors impair our ability to develop an accurate
assessment of risks. Chief among these are heuristics and biases, which can
overtake reasoned analyses and decisions. Heuristics are practical
problem-solving methods that serve as shortcuts in our cognitive thinking,
influenced by our life experiences. Heuristics are important to consider
because they may cause us to arrive at erroneous conclusions. Instead of
stepping from A to B to C to D, heuristics allow us to jump directly from A to
D. If the current situation is aligned to the foundation of our heuristic, this
is good because the heuristic saves us time. However, if information in steps B
or C indicate a different path, jumping from A to D can result in a completely
different and wrong conclusion.

Biases impact heuristic thought processes and have a major
impact on how we identify, analyze and evaluate risks. Important biases in risk
perception include:

Anchoring bias. Our thinking is influenced by the
first relevant data point we encounter when considering any situation. For
example, if we are purchasing a used vehicle, the first person to offer a price
establishes a range of reasonable prices in everyone’s minds, anchored around
the first stated value.

Availability bias. People tend to make judgements and
decisions based on new, recent or dramatic information. Our memories fade
quickly and the significance of what happened two years ago pales in contrast
with what we read in daily—or hourly—news feeds. In our always-on social media
lives, the dual effect of availability and anchoring poses a dangerous
combination. Early reports of events can be fraught with inaccuracies, which
are generally corrected later. The anchoring effect of the early news, however,
can overpower the later, accurate—and perhaps less dramatic—information.

Confirmation bias. People also tend to believe
information that supports our position or preconceptions and discount other
data, regardless of how accurate or relevant it is. In his book The Black
Swan
, Nassim Nicholas Taleb noted, “We will tend to more easily remember
those facts from our past that fit a narrative, while we tend to neglect others
that do not appear to play a causal role.”

Conservatism bias. Unlike the availability bias, we
may also tend to discount new data or evidence in favor of knowledge we have
obtained over time. Consumer companies, for example, are often slow to
recognize and adapt to changing consumer preferences.

Information bias. Many individuals and organizations
seek more and more information about a situation even though the additional
information will not affect decisions on how to act or react.

Other Factors Influencing Risk Perception

In addition to biases, a number of other factors also affect
the way we perceive and deal with risks. One of the most important is
experience and familiarity. If we do not have first-hand knowledge of a risk,
we tend to discount both the likelihood that it will materialize and its
possible consequences. This factor is two-sided: On one hand, having direct
experience with a risk makes it seem more likely; on the other hand, constant
exposure to a risk makes it so familiar that we often discount the consequences,
perhaps because we have adapted to living with it.

Another factor is time relevancy. We tend to magnify the
importance of risks that have occurred recently, compared to risks that have
not occurred for some time. For example, after the September 11 terrorist
attacks, aviation safety and security was paramount, and significant measures
were implemented to secure air travel that remain in place to this day.
Terrorists are just as likely, if not more so, to attack other modes of
transportation or public gathering locations such as shopping malls and sports
venues, however.

Control also impacts risk perception. People tend to
discount risks if they feel in control of the situations exposing them to those
risks. As David Ropeik points out in his book How Risky Is It, Really?,
we believe that it is safer to drive than fly to a distant location, yet
statistics show it is much more likely for a traveler to die in a traffic
accident. A similar situation exists for the risk of being distracted using a
mobile phone while driving: Hands-free technology is considered a viable risk
reduction technique because the driver does not hold a physical handset while
driving, yet research indicates that the real risk is the mental distraction of
talking with someone and processing information not relevant to operating a
vehicle.

Putting a face to a risk and its consequence also has an
outsized effect on perception of a risk. Notice how the American Society for
the Prevention of Cruelty to Animals (ASPCA) and humanitarian aid organizations
use this to influence our contributions. Heart-wrenching images of neglected
animals and malnourished children stir our emotions and drive us to donate.

Weighing downside risks versus benefits is another important
consideration. When looking at the pros and cons of a decision and considering
the risks, people often discount downside risk consequences in inverse
proportion to the perceived upsides or benefits. The greater the upside, the
more they discount the likelihood that the downside risk will occur. This
phenomenon is evidenced in many failed acquisitions where the expected benefits
never materialize.

Risks like automobile accidents and workplace safety
incidents have rich historical data sets that lend themselves to mathematical
modeling and projections of future occurrences. Advanced software can perform
sophisticated calculations to estimate the “level of risk” based on likelihood
and consequence curves. However, for many business-level risks, scant data
exists to determine curves that realistically model likelihood and consequence.
Yet many executives rely on calculated “value-at-risk” figures to make
strategic and tactical business decisions. Running thousands of calculations
using Monte Carlo techniques implies accuracy and validity, yet the inputs used
for likelihood and consequence can be unrealistic.

Techniques to Compensate

Despite the seemingly insurmountable challenges to determine
an accurate accounting of risks and risk levels, there are some relatively
simple techniques that can be used to provide a counterbalance. Remember that
risk management is a journey, and programs should improve over time.
Incorporating one or more of the following techniques will help move toward
more accurate risk identification, analysis and evaluation.

Use calibration exercises. In his book The Failure
of Risk Management
, Douglas Hubbard discusses using calibration exercises
prior to a risk identification session to guide individuals’ estimates of risk
likelihood and consequence. People are generally overconfident in their
abilities, and calibration exercises provide a dose of reality. Subsequent risk
identification and analyses are usually more realistic and accurate following
calibration sessions.

Adjust likelihood and consequence scales. Many risk management
programs determine risk levels by combining estimates of likelihood of
occurrence and magnitude of consequence. Simple techniques like adding or
multiplying the ratings together are often used, and then a scheme is derived
to decide thresholds for risk significance. An observed shortcoming of this
approach is that low-likelihood but high-consequence risks are overlooked. Yet
it is precisely these risks that significantly harm organizations. A useful
counterbalance is to place more weight on the consequence scale compared to the
likelihood scale to maintain the visibility of lower-likelihood but
higher-consequence risks.

Ask the same question multiple ways. As discussed
earlier, heuristics and biases impact the way we perceive risks and risk
levels. Our thought processes are affected by the wording and presentation of a
question or scenario. Employing a technique that solicits information about
potential risks using differing language or contexts can help detect variances
in risk perception and guide follow-up work to determine more accurate
information. Well-constructed risk surveys often employ this technique to great
advantage.

Ask the same question of multiple people. Similar to
asking the same question multiple ways, presenting questions to multiple people
highlights disparities in risk perception. Risk surveys can be sent to people
from different parts of an organization and the results analyzed to detect wide
variances in risk perception. For example, higher up the leadership chain,
views of risks tend to focus on longer-term strategic risks, while those in the
lower- to mid-levels focus on more operational and tactical risks. Similarly,
individuals in different functional areas will frequently view risks and risk
levels differently based on their individual heuristics and biases. This does
not mean one is right and the other wrong. Rather, it highlights the necessity
and usefulness of considering multiple factors in identifying significant
risks.

Ask the same question at different times. Current events
and work issues affect thought processes. If a risk survey or other risk
identification process is conducted at roughly the same time every year,
results will skew risks to those that are most front-of-mind at that time. For
example, if risk assessments are conducted around strategic planning time,
longer-term strategic risks will appear more important. If assessments are
conducted during heavy manufacturing periods, such as an inventory build-up for
holiday sales, manufacturing capacity and supply chain risks will seem more
important. To counter these tendencies, conduct risk assessments at varying
times during the year.

Ensure the risk management process is continuous.
Risks are dynamic and a risk management program should be as well. Building on
the technique above, multiple risk identification processes and frequencies
should be used and should be time-independent. Emerging risks do not wait for
regularly scheduled risk assessment workshops. Use multiple risk sensing
platforms to identify new risks and detect the onset of a known risk.
Well-defined risk indicators accompanied by robust data acquisition and
analytics are useful for this purpose. Every organization has a unique
operating structure and rhythm, and risk management processes benefit by aligning
to these. Maintaining a real-time or near-time focus on risks and risk
treatment helps limit the potential that a critical risk will be missed or not
addressed.

Use technology to help. Technological solutions are
valuable in managing risk information, acquiring and analyzing risk data, and
automating information flows and decisions. A key benefit is the ability to
remove psychological and emotional influences in processing risk data. Advances
in artificial intelligence and data analytics can provide cost effective and
valuable insight into risk environments and emerging risks. And risk management
software platforms can aggregate risks, automate risk assessments, and track
risk treatment actions.

Developing, implementing and evolving a risk management
program is challenging at best. Psychological influences on risk perception can
negatively impact the validity and focus of risk mitigation measures included
in a program. Understanding these influences and integrating methods to
compensate for them can substantially improve the effectiveness and value of
any risk management initiative.

Leave a Reply

Your email address will not be published. Required fields are marked *