ArticlePDF Available

Choosing remedies after accidents: Counterfactual thoughts and the focus on fixing “human error”

Authors:

Abstract and Figures

The present research is motivated by an interest in why organizational decision makers so often respond to accidents with remedy plans that focus narrowly on correcting human error rather than more environment-focused plans or more encompassing plans. We investigated the role of counterfactual thinking in the decision-making tendency toward human-focused plans. Our experiments indicated that even in a domain where human-focused remedies were not otherwise appealing, many participants decided on human-focused remedies after they had generated an "if only" conjecture about the accident. This reflects that human actions are often selected as the focus of "if only" conjectures and, importantly, that this focus "locks in" and carries through to subsequent remedy decisions. Our hypothesis that remedy plans are produced from "if only" thoughts was supported over several alternative interpretations. We discuss implications for research on the relation between counterfactual thinking and adaptive learning.
Content may be subject to copyright.
Psychonomic Bulletin & Review
1999, 6 (4),579-585
Choosing remedies after accidents:
Counterfactual thoughts and
the
focus
on
fixing "human error"
MICHAEL
W.
MORRIS,
PAUL
C,
MOORE,
and
DAMIEN
L. H. SIM
Stanford University, Stanford, California
The
present
research is motivated by an interest in why organizational decision makers so often re-
spond to accidents with remedy plans thatfocus narrowly on correcting human error
rather
than more
environment-focused plans or more encompassing plans. We investigated the role of counterfactual
thinking in the decision-making tendency toward human-focused plans. Our experiments indicated
that even in a domain where human-focused remedies were not otherwise appealing, many partici-
pants decided on human-focused remedies after they had generated an
"if
only" conjecture about
the
accident. This reflects that human actions are often selected as the focus of
"if
only" conjectures and,
importantly, that this focus "locks in" and carries through to subsequent remedy decisions. Our hy-
pothesis
that
remedy plans are produced from
"if
only" thoughts was supported over several alterna-
tive interpretations. Wediscuss implications for research on
the
relation between counterfactual think-
ing and adaptive learning.
Accidents incite
judgment
and
decision making. Re-
sponses to accidents have captured the interest
of
orga-
nizational researchers studying actual decisions in field
settings (March
& Simon, 1958) and
of
cognitive psy-
chologists studying decisions in laboratory tasks (Kah-
neman
& Tversky, 1982). Among the more important in-
sights from organizational research is that when managers
choose
remedies
after
accidents,
they
often
focus on
human
factors to
the
exclusion
of
relevant
factors in
workers' environments (Perrow, 1984; Vaughan, 1996).
Organizational researchers have also drawn attention to
institutionalized procedures
of
analyzing accidents, such
as the practice
of
identifying a "root cause" through a re-
constructive simulation
of
the chain
of
events (Carroll,
1995). Interestingly, these procedures require a strategy
of
diagnosis like that which cognitive psychologists have
called
"counterfactual
thinking" or
"if
only"
thinking
(Kahneman
& Miller, 1986; Kahneman & Tversky, 1982).
The present research takes this parallel as a point
of
de-
parture. We build on evidence from research on acade-
mic tasks suggesting that people's
"if
only" conjectures
about a negative outcome shape their subsequent plans
for preventing recurrences (Roese, 1994). We hypothe-
size
"if
only" conjectures after accidents shape the focus
of
subsequent remedy decisions. Hence,
"if
only" con-
jectures focusing on actions
of
employees would lead to
Wewould like to thank Pam Haunschild, ltamar Simonson, Jim March,
and Vittorio Girotto for their helpful comments, and Aimee Orolet and
Steve Su for assisting in the coding. A more detailed report of empirical
results is available on request from the authors. Correspondence should
be addressed to M.
W. Morris, Graduate School of Business, Stanford
University, Stanford, CA 94305-5015 (e-mail:
morris_michael@
gsb.stanford.edu).
plans to change the employees rather than changing their
environment.
Before introducing the studies testing our hypothesis
about the role
of
"if
only" conjectures in shaping remedy
decisions, we shall first review some related literature
that highlights the importance
of
this hypothesized influ-
ence on remedy decision making. First, we will review
conclusions from field research on industrial accidents
about the context and patterns
of
remedies chosen to pre-
vent accident recurrences. Next we will review evidence
from research on the heuristics guiding
"if
only" thinking.
Responses to Accidents in Organizations
An enduring idea in organizational research is that man-
agerial decisions are often made through standard oper-
ating procedures rather than through thorough analysis
(March
& Simon, 1958). Although the solutions provided
by such procedures are not always optimal, managers tend
to settle on the first plausible solution rather than search
more exhaustively for the optimal solution-c-t'satisficing"
rather than optimizing (March, 1994). In many industries,
the standard procedure for responding to an accident is to
identify a "root cause" (Carroll, 1995). To do this, a deci-
sion
maker
reconstructs or simulates the chain
of
events
leading to the accident. After large disastrous accidents,
such as those involving the Three Mile Island nuclear plant
or the Space Shuttle Challenger, such reconstructions have
become public, drawn-out affairs. After smaller accidents,
the same process occurs in miniature as a manager prepares
to report on the root cause
of
the accident.
Responding to an accident by searching for the root
cause
of
a given outcome can be contrasted with other ways
of
analyzing the problem. A different approach is to con-
sider the current accident as one case in a larger data set
579
Copyright 1999 Psychonomic Society, Inc.
580
MORRIS,
MOORE,
AND
SIM
of
accidents
and
nonaccidents in the organization's expe-
rience
and
to
analyze
the covariation
of
outcomes
and
causal factors (see Kelley, 1967). Evidence
both
from the
laboratory
and
from
the field suggests
that
a failure to
consider a given
case
in light
of
past
cases prevents dis-
covery
of
some
causal
factors
that
may
be important.
The laboratory research
of
MeGill (1989)
has
shown that
discovery
of
some causal factors requires
that
the current
case be compared
with
past
cases-that
participants im-
plicitly ask,
"Why
did
the outcome occur on this occasion
and not on
past
occasions?" Organizational research on
accident diagnoses suggests a similar pattern. In the case
of
the Three Mile Island nuclear plant, the initial diagno-
sis, based on a close reconstruction
of
the
accident
event,
was revised when the accident was analyzed in the context
of
the organization's historical experience
with
related
incidents (Perrow, 1981). Similarly, the initial attribution
of
the Challenger disaster to an inappropriate launch deci-
sion was qualified after broader analysis (Vaughan, 1996).1
Another theme in organizational research on responses
to accidents is the tendency
of
decision
makers
to focus
on human factors
rather
than factors in the machine or
organizational
environment.
Studies
across
industries
confirm
that the
most
prevalent diagnosis
of
industrial
accidents is
"human
error," and remedies
tend
to focus
on
changing
human
factors
(through
replacing
or re-
training employees) rather than changing
machine
or or-
ganizational factors (Norman, 1990, 1992). In Vaughan's
(1996) words, "invariably, the accepted explanation [for
industrial
accidents]
is some
form
of
'operator
error,'
isolating in the media spotlight someoneresponsible for the
hands-on work: the captain
of
the ship, a political func-
tionary, a technician, or middle-level
managers"
(p.
393).
In the aviation industry, for example, 75%
of
fatal aviation
accidents in a recent 5-year-period were
blamed
on pilot
error
(National
Traffic
Safety Board, 1990).
The
per-
centage
of
accidents attributed to
human
error
may be
greater than
deserved
because
ofthe
cognitive processes
involved after accidents.
The mere prevalence
of
a focus on
"human
error" does
not necessarily indicate nonoptimal decision making. In
some domains
of
organizational accidents, human-focused
remedies may be the only sort feasible. Or, human-focused
remedies may be
more
efficacious
than
other
possible
remedies that focus on the environment or on interactions
of
factors. Nevertheless, there are some
domains
that do
not fit these descriptions, and yet human-focused remedies
are favored (Norman, 1990).
The
present research inves-
tigates a possible contributor to human-focused remedy
decisions
in
accident
domains
where
human-focused
remedies are
not
the
only possible
remedy
or the
most
efficacious
remedy. We
propose
that
a
prompt
for
"if
only" thinking leads to an early narrowing
of
focus on a
single factor, often a
human
action, and this focus "locks
in" and carries
through
to
remedy
decisions.
"If
Only
...
" Conjectures and Remedy Decisions
Accidents have
been
studied by
psychologists
inter-
ested in people's thinking about counterfactual or "might
have
been"
scenarios (Kahneman & Miller, 1986; Kahne-
man
& Tversky, 1982). One form
of
counterfactual think-
ing that
occurs
after negative,
unexpected
outcomes such
as accidents is expressed by conjectures that the outcome
could have
been
averted
"if
only"
one
antecedent event
had
been
changed.
An
interesting
feature
of
"if
only"
conjectures is that they emphasize
how
a change to only
one factor
could
have
made
a difference. The outcome is
traced to a single, critical link in the causal chain. By con-
trast,
other
ways
of
framing explanations for an outcome
foster more emphasis on combinations
of
multiple factors.
The
narrow
focus
of
"if
only"
conjectures
has
led
re-
searchers
to
study
how
people
select
one focal factor
from the
many
possible factors antecedent to an outcome
(Roese, 1997; Wells, Taylor, & Turtle, 1987). Research
on this
question
has
uncovered
heuristics
guiding
the
focus
of
"if
only" conjectures,
such
as that
human
ac-
tions are favored over nonhuman factors having the same
causal relationship to the outcome (Girotto, Legrenzi, &
Rizzo, 1991). An account for this tendency is that human
actions
are
more "mentally
mutable"
than other factors;
that is, it is easier to
imagine a
change
in human action
(a
person
having done x instead
of
y) than to imagine a
change
in an environmental
factor
(the
replacement
of
one
machine
with a different machine). Importantly, the
relative
ease
of
bringing to
mind
the alternative
human
behavior x does not
mean
that this behavior would be rel-
atively
easy
to bring about in reality. In reality, changing
environmental factors
might
be
easier
than controlling
human
behavior.s
In
the
present
research, we
test
the
hypothesis
that
human-focused
remedy decisions
are
caused by human-
focused
"if
only" conjectures. A
precedent
for the notion
that
remedy
decisions are directly
shaped
by initial
"if
only" conjectures comes from research on individual aca-
demic performance (Roese, 1994). There is both labora-
tory
and
field evidence that the content
of
students'
"if
only" conjectures after negative feedback determines the
content
of
their plans for improved performance in the fu-
ture (Roese, 1994; Roese
& Olson, 1995). Certain kinds
of
"if
only" conjectures beget certain kinds
of
corrective
plans. In this vein, we hypothesize
that
human-focused
"if
only"
conjectures
may lead to
human-focused
remedy
decisions.
PILOT
STUDY
The
tendency
for
"if
only"
conjectures
to
engender
human-focused
remedy decisions is
of
greatest interest
in domains where human-focused remedies would not be
otherwise chosen. We sought to
study
a type
of
accident
in which
human-focused
remedies
are
often chosen even
though
they
are rejected by
people
who
fully evaluate a
variety
of
different
remedy
plans. In designing
our
vi-
gnette, we drew on studies
of
industrial accidents that point
to situations where
human
operators have to respond to
changes in the machine environment (Norman, 1992; Rea-
son, 1990). Girotto et al. (1991)
provided
the everyday
example
of
how a driver
of
an old car compensates for the
"IF ONLY" THOUGHTS AND REMEDY DECISIONS 581
machine's degradation by developing skills that take over
functions once handled by the machine, such as signal-
ing. In so doing, the driver's role becomes more salient.
After an accident, this driver is a more likely focus
of
blame and remedy attempts than would be the driver
of
a new, fully functioning car. Yet it is precisely in the case
of
the old car, not the new car, where diagnosis and rem-
edy should focus on the machine rather than on its human
operator. A larger example
of
this pattern
of
remedy de-
cisions for this type
of
accident was the recent termination
of
the Russian cosmonauts involved in accidents on the
space station, Mir, even though the accidents occurred dur-
ing their efforts to repair components that had stopped
functioning (Specter, 1997).3 Norman (1992) presented
case studies
of
industrial settings where the focus on the
human side
of
the human-machine interface led to a fail-
ure to prevent recurrences
of
easily avoidable accidents.
In our pilot study, we presented a description
of
an ac-
cident
of
this type and checked how participants would
analyze the
strengths
and
weaknesses
of
human
and
environment-focused plans for remedying the problem.
Method
We presented 54 undergraduates at a major California university
a story entitled "Accident at the Lawn Mower Factory," which de-
scribed a factory that had recently started to produce a new lawn
mower product:
The new lawn mower is built on the same assembly line as the old one
was but has required a few changes in the manufacturing process.
The story continued by describing how workers have reacted to the
change:
After the change, some assembly line workers have grumbled that
mowers no longer arrive to their work station in a steady flow.One com-
plained about "rushing to finish the first mower in time for a second,
but then having to wait forever for the third to arrive." Long-time work-
ers have started the practice of "reaching up the line" to work on a
mower before it is secured at their work stations. This has nearly caused
accidents, and the foreman has cautioned against it.
Next was a description of the accident:
Today an accident occurred in the factory that resulted in a line shut-
down for over three hours. Paul, a long-time worker was "reaching up
the line," lost his balance, and caused a mower to fall on some delicate
quality control instruments.
The story ended by conveying that "in addition to the high replace-
ment costs
of
the broken instruments, the lost production
of
this
highly demanded product is extremely costly for the company."
We assessed participants' responses by asking them to consider
two general approaches to preventing accidents, "changing the work-
ers" and "changing the work environment." "Changing the workers"
was defined as either hiring different people or training the current
people into different modes
of
behavior. "Changing the work envi-
ronment" was defined as redesigning the organizational policies or
the physical structure
of
the setting or the machinery. Participants
were told that both options were fiscally feasible and were asked to
consider these two options with respect to several criteria. Three
items captured different aspects
ofthe
perceived efficacy of a solu-
tion: "Is this approach likely to solve the problem completely?" "Is
this approach likely to provide a lasting solution?" and "Could this
approach, by itself, eliminate the problem
of
reaching-related acci-
dents?" A final item probed the perception that a particular remedy
would not only be inefficacious but also counterproductive (i.e., "Is
it likely that this solution will be counterproductive?"). All ratings
were taken on a Likert-type scale
(I
= not at all; 7 = very much).
Results
Because
of
the intercorrelation among the items tap-
ping the perceived efficacy
of
changing the workers (mean
r = .50) and changing the environment (mean r = .32), in-
dices were formed by averaging the three ratings for each
remedy. We submitted these to a repeated measures analy-
sis
of
variance (ANOVA)
and
found that participants
judged human-focused changes to be far less efficacious
(M
= 3.31) than changes to the work environment
[M
=
5.51;
F(l,50)
=
74.64,p
< .001]. Another way to sum-
marize the results in evaluating the strengths
of
the two
plans was that human-focused plan was rated better by
only 13%
of
participants.
Ratings
of
weaknesses
of
the plans showed a consis-
tent pattern. The likelihood
of
counterproductive results
was rated higher for a human-focused intervention (M =
4.26) than an intervention focused on the work environ-
ment [M = 3.31; F(1,50) = 9.37, p < .005].
In sum, results reveal that participants who thoroughly
consider decision options recognize the relative limitations
of
human-focused remedies in this accident context. The
new assembly line creates a strong temptation to reach up
the line, so retraining or replacing the workers would be
unlikely to prevent a recurrence.t
EXPERIMENT 1
Experiment 1 presented the same accident case as in
the pilot study, but introduced a prompt to respond to the
accident with an
"if
only" thought. We put several groups
of
participants in the role
of
a manager responding to the
accident at the lawn mover factory. With several different
formats, we checked the association between the content
of
"if
only" conjectures and
of
subsequent remedy decisions.
Method
Two groups
of
participants were put in the role
of
manager and
asked to respond to the lawn mower accident story described above.
These different replications, in which the format
of
measures also
varied, will be reported as Experiments IA and lB.
Experiment lA. A group
of
management students (n = 58) were
prompted to make
"if
only" conjectures in response to the outcome
and then to make decisions about how to remedy the problem. Their
counterfactual task asked them to complete a sentence beginning
"If
only
...
" and ending with "this accident would not have hap-
pened." This allowed us to assess the factors on which people spon-
taneously focus on when undoing an accident. Afterward, partici-
pants were told to suppose that the factory allowed a limited amount
of
funds to use in trying to reduce the chances
of
future accidents.
They were told to consider the options
of
doing nothing at all,
of
trying to change employees' behavior, and
of
trying to change the
machine environment. Participants then wrote a summary
of
their
decision.
Experiment
lB.
To check whether the findings generalize be-
yond the population
of
management students, museum goers (n =
93) were asked to choose which
of
two options was more similar to
their own intuitive reaction to the accident. The counterfactual op-
582 MORRIS, MOORE, AND SIM
Table 1
Frequency of Person- and System-Focused Remedies
by Managers as a Function of Having
Generated
Person-
and
System-Focused
"If
Only" Statements in
Experiment
lA
"If
Only" Conjecture Type
Worker Had Assembly Line
Not Reached Was Designed Better Other
Remedy Focus
n
%
n % n % Total
Retrain workers 15 56 3
13
3
38 21
Redesign line
11 41 18
78 2 25 31
Other
I 4
2
9 3 38 6
Total 27 23 8 58
Note-x
2(1,58) = 17.79,p < .001.
tions were,
"If
only Paul had not reached up the line, the accident
would not have happened," and
"If
only the assembly line had been
designed to deliver engines in a steady flow, the accident would not
have happened." Later, they were presented with two remedy op-
tions and asked to select the one that they would recommend. The
remedy options were, "Educate the workers about unsafe behaviors
on the assembly line," and "Investigate possible design changes to
the assembly line."
Results
In Experiment
lA,
both
the
"if
only" conjectures and
remedy plans were
coded
into three categories: (1) those
focused on the human operator
of
the machine, (2) those
focused on the machine environment, and (3) those fo-
cused elsewhere. Two coders agreed on over
90%
of
the
cases. In counterfactual conjectures, the modal response
(47%
of
participants) focused explicitly on the worker's
action
("If
only Paul
hadn't
reached up the line"). A con-
siderable fraction
of
responses (40%) focused on the as-
sembly line, and the remaining fraction (13%) focused
on other aspects
of
the worker's environment such as the
management. In response to the remedy decision task, a
large fraction
of
participants (36%) generated plans that
focused narrowly on changing the employees' behavior
(e.g., "Hold workshops to change the workers' behavior"),
(Notice that although this is not a majority
of
participants,
the rate
of
preference
for human-focused
remedies
is
strikingly higher than the 13% rate at which participants in
the pilot study favored human-focused over environment-
focused plans.) The remaining participants addressed the
problem through either a redesign
of
the assembly line
(53%; e.g., "plan to regularize the assembly line flow so
that it is steady and continuous") or some other approach
(10%; e.g., "bring in a consultant" or
"do
nothing").
Although the increased frequency
of
preference for
human-focused
remedies
(relative to the
pilot
study)
suggests support for the hypothesis, more direct evidence
comes from looking at remedy decisions conditional on
participants'
"if
only"
thoughts, as shown in Table 1.
Among participants who went down the cognitive path
of
a human-focused
"if
only" conjecture, 56% decided
on a human-focused remedy. By contrast, among those
whose
"if
only" conjecture focused on the machine en-
vironment, only 13% decided on a human-focused rem-
edy. The association between human-focused
"if
only"
thoughts
and
human-focused remedies was highly sig-
nificant
[X
2(1,58)
=
l7.79,p
< .001].
Experiment 1B closely replicated the basic results
of
Experiment 1A with a different sample
of
participants
and a different response format. Again, we found that a
substantial fraction
of
participants favored the human-
focused
"if
only" conjecture (48%)
and
a substantial frac-
tion favored the human-focused
remedy
(34%). More
importantly, the tendency toward human-focused reme-
dies was far higher among participants whose
"if
only"
thought focused on the human operator (49%) than among
participants whose
"if
only" thought focused on the ma-
chine environment (21%). The association between
"if
only" thought and remedy decision content reached a high
level
of
statistical significance [X
2(1
,93) =
8.l0,p
< .005].
EXPERIMENT
2
The results
of
Experiment 1 established an association
between counterfactual thinking
and
later remedy deci-
sions. Yetthese results do not prove that remedy decisions
are generated from
"if
only" conjectures. Three alternative
interpretations for the effect
in Experiment 1could be raised.
First, it could be argued that decisions about remedies come
first in people's minds and
"if
only" thoughts are derived
afterward to support preexisting remedy plans. Second, it
could be argued that the association merely reflects an im-
plicit pressure to be consistent across tasks. Third, asking
for a counterfactual statement and asking for a remedy may
be parallel ways
of
eliciting a single representation
of
the
situation. Experiment 2 tests these interpretations by in-
vestigating whether priming counterfactual thoughts in-
fluences remedy decisions, whether influence runs in the
opposite direction ("remedy primacy interpretation"),
whether influence runs in both directions ("consistency
pressure interpretation"), or whether both are influenced
by a single representation
of
the situation ("common rep-
resentation interpretation"). We predicted that there would
be influence in the direction from counterfactual thoughts
to remedy decisions, but not in the opposite direction.
"IF ONLY" THOUGHTS AND REMEDY DECISIONS 583
Manipulated Subject of
"If
Only" Conjecture
Person None Environment
Table 2
Endorsement
of Person-
and
System-Focused
Remedies as a Function
of
Manipulated
"If
Only"
Conjecture
in
Experiment
2A
Remedy
Focus
M SD M SD M SD
Retrain workers 5.06. 1.60 4.73. 1.64 4.37. 2.03
Redesign line 5.35. 1.32 5.77.
b
1.27
s.zr,
.98
Difference - .29. 2.02
-1.04.
b
1.84
-1.84
b
2.39
Note-Different subscripts by row indicate significant differences in
means
(p
< .05).
Method
Participants.
In an experiment (Experiment 2A) testing whether
remedy decisions are affected by priming
of
"if
only" content, 62
undergraduate students were randomly assigned to conditions in
which they were led to generate a human-focused counterfactual
statement, no counterfactual statement, or an environment-focused
counterfactual statement before making remedy decisions. In the
comparison experiment (Experiment 2B), testing whether
"if
only"
thoughts are affected by priming
of
remedy decision content, 73 un-
dergraduates students were randomly assigned to conditions in
which they generated a human-focused remedy, no remedy, or an
environment-focused remedy before making a counterfactual con-
jecture.
Design,
Procedure,
and
Materials.
Participants began by read-
ing the accident vignette. In the primary experiment (2A), partici-
pants in one condition went directly to the remedy decision task,
without being prompted to think counterfactually beforehand (no
"if
only" prompt). Participants in the other two conditions were
prompted to complete a counterfactual conjecture with a particular
conte.it focus: Participants generated a conjecture focusing either
Oil
the worker
("If
only the worker
__
") or on the machine envi-
ronment
("If
only the assembly line
__
"), All participants rated
two remedy decisions on 7-point scales, one tapping willingness to
attempt a human-focused remedy and the other tapping willingness
to attempt an environment-focused remedy.
The procedure
of
the second experiment (2B) simply inverted the
independent and dependent variables. In two conditions, partici-
pants were
first
primed
with
human-focused
or environment-
focused remedies by requiring them to complete a sentence describ-
ing a possible remedy (the sentences began, "I would try to train
the workers to
__
" or "I would try to redesign the assembly line
to
__
"), In the third condition, participants were not exposed to
any possible remedies. In all three conditions, participants then
rated the similarity
of
their own thoughts to two counterfactual con-
jectures (the counterfactuals were,
"If
only Paul had not reached up
the line, this accident would not have happened," and,
"If
only the
assembly line delivered the engines in a steady flow, this accident
would not have happened").
Results
Table 2 shows the results from Experiment 2A testing
whether the counterfactual thoughts about an accident
influence decisions about how to remedy the problem.
As expected, the tendency to differentially endorse the
human- focused remedy was greatest among participants
who had been primed with human-focused
"if
only" con-
jectures and least among participants who had been primed
with environment-focused conjectures. The linear con-
trast was significant
[F(1 ,59) =
4.19,p
< .05], reflecting
that the two content-primed groups differed from each
other [t
=
-2.24,
P < .05]. Another way to summarize
these data that facilitates comparison with previous stud-
iesis the percentageofparticipantswho favored the human-
focused plan over the environment-focused plan, which
was 24% in the human-focused counterfactual condition,
12% in the no-counterfactual condition, and 9% in the
environment-focused counterfactual condition.
In contrast to Experiment 2A, there was no effect
of
the manipulation in Experiment 2B, which tested for in-
fluence
of
remedy decisions on endorsement
of
"if
only"
conjectures. The degree
offocus
on human action (mea-
sured by the difference between endorsement
of
human
and environment counterfactuals) was nearly identical in
the human-remedy prime condition
(M
= - .30), the no-
prime condition (M
=
-.31),
and the environment-remedy
prime condition
[M= - .54; contrast
F(l
,70) = .18].5
GENERAL
DISCUSSION
The present studies support the hypothesis that
"if
only" conjectures after accidents shape decisions about
how to prevent recurrences. Experiment 1 documented the
statistical association between the content of counterfac-
tual conjectures and
of
remedy decisions. The causal link
between the focus
of
an
"if
only" conjecture and that
of
remedy decisions was supported in Experiment 2A. In Ex-
periment 2B, no support was found for three alternative
explanations based on the ideas, respectively, that remedy
decisions are cognitively prior to counterfactual conjec-
tures, that participants alwaysstriveto maintain consistency
of responses, and that asking for counterfactual statements
and asking for remedy decisions are parallel means
of
eliciting a common representation
of
the situation.
We have argued that one reason to study the processes
influencing accident remedy decisions is that psycho-
logical factors may contribute toward a tendency to focus
narrowly on correcting "human error" after industrial
accidents. In particular, we have suggested that
"if
only"
conjectures are drawn toward human actions and hence
may engender human-focused decisions at a greater rate
than would occur otherwise. Our results are consistent
with this argument. Experiment I documented that a
substantial fraction
of
participants focused on human ac-
tions when generating an
"if
only" conjecture about an
accident. This was the case even though the accident was
designed to be one for which a narrowly human-focused
plan would not be favored in a direct analysis
of
differing
plans (and, in fact, was not favored by participants in the
pilot study). Taken together with the evidence that con-
jectures influence subsequent decisions, the results lend
credence to the argument that human-focused remedy
decisions may occur as a by-product
of
human-focused
"if
only" conjectures. Given that these decisions have
been identified as potentially problematic by accident re-
searchers, it is important to consider their possible psy-
chological roots.
Implications
for
Theory
The present research adds to the emerging literature
on counterfactual thinking and decision making. Most
584 MORRIS, MOO RE, AND SIM
past research on the relation between
"if
only" thinking
and adaptive decision making has emphasized its salutary
role. This has been the dominant message in theoretical
statements (e.g., March, Sproull, & Tamuz, 1991) and in
empirical findings from performance domains (e.g., Mor-
ris & Moore, 1998; Roese, 1994). These studies have em-
phasized that
"if
only" thoughts about a negative perfor-
mance engender adaptive plans for avoiding recurrences
of
the outcome.
By contrast, the present research indicates that
"if
only"
thoughts influence remedy plans, but it calls into question
the adaptiveness
ofthese
plans. Our pilot study indicated
that participants
did
not prefer human-focused remedy
plans for the accident after a thorough analysis. Although
there is no definitive basis for calling a
plan
better or
worse in the case
of
a hypothetical accident, the pilot
study suggests
that
participants
recognize
a human-
focused plan as worse. Yet, the
"if
only" conjectures gen-
erated by many participants in Experiment 1 led them to
human-focused plans. Hence, our experiment illustrates
a domain where
"if
only" thinking may impede partici-
pants from choosing the best decision option. Together
with other
recent
empirical
findings
(Sim
& Morris,
1998) and theoretical analyses (Sherman & McConnell,
1995), the present findings contribute to an emerging
countervailing theme in the literature that decision mak-
ing is not always well served by
"if
only" thinking. In-
stead, it is a mixed blessing for decision
making-helpful
in some domains, harmful in others.
Issues for Future Research
An important issue for future research is identifying
the boundary conditions under which
"if
only" conjectures
impede adaptive decision making about accidents. This re-
search will clarify the benefits or costs
of
procedures-
formal and
informal-that
encourage decision makers to
generate
"if
only" conjectures about accidents. Given that
"if
only" conjectures may increase the likelihood
of
a
focus on human actions, one candidate for a moderating
variable is the efficacy
of
human-focused remedies in a
given accident domain.
Another, more subtle, moderating variable may be the
configuration
of
causes in a given accident domain. Here,
the important aspect
of
"if
only" conjectures is that com-
pared with other ways
of
framing causal attributions for
accidents, they lead to a narrower focus on one antecedent
factor rather than a broader focus on multiple, interact-
ing antecedent factors. Hence,
"if
only" conjectures may
impede decision making when diagnostic narrowness is
a liability. In other words,
"if
only" conjectures may en-
gender inefficacious plans to prevent accident recurrences
when the
causal
structure
of
the
domain
is
complex
(Jervis, 1996). The decisions that result would exemplify
what Dorner (1997) has called "unconditionalized plan-
ning"-remedies
that adjust one factor in a system with-
out taking into account chain reactions that may cause
new problems. In a recent investigation
of
such pitfalls,
participants reacted to outcomes in a complex domain by
following heuristic rules to generate counterfactual con-
jectures. Consequentially, these conjectures focused on
an antecedent factor that lacked any causal relation to the
outcome
and
that participants themselves evaluated as
erroneous upon later analysis (Morris, Sim, & Moore,
1999). Adaptive learning from counterfactual thinking
in complex domains may require that human intuition be
bolstered by computer simulations
of
the causal config-
uration
that
enable exploration
of
interaction effects
(Moore & Thomsen, 1996).
Models
of
adaptive
learning
from
counterfactual
thoughts must consider the types
of
outcomes that pro-
voke such thoughts. In addition to thoughts that "undo"
accidents, there is potential relevance in thoughts that
"undo" nonaccidents-that is, thoughts that posit an ac-
cident occurring after one has not. Under most circum-
stances, these conjectures are infrequent because the ex-
perience
of
a nonaccident lacks motivating or salient
factors that would provoke counterfactual thinking. An
exception,
of
course, is the class
of
nonaccidents that
qualify as
"near
accidents." Indeed, research on aviation
pilots' responses to near accidents observes the frequent
occurrence
of
counterfactual
thoughts
about
how it
"could have been worse" as well as thoughts about how
it
"could
have been better," yet only the latter, upward
comparisons enabled adaptive learning (Morris & Moore,
1998).
Although
our
focus heretofore has been the role
of
counterfactual thoughts in learning, learning is not the
only function
of
counterfactual thinking, and it is an im-
portant project for future research to explore other func-
tions. The kinds
of
counterfactuals that serve the func-
tion
of
learning differ from those which serve functions
related to people's emotional needs. There is consider-
able evidence that the function
of
coping with negative
affect is served by thoughts about how an outcome might
have been worse (Roese & Olsen, 1995). And, counter-
factual conjectures can serve a person's motive
of
world-
view preservation (Tetlock, 1998). For example, after an
accident has occurred, one can preserve a beliefthat the
setting is essentially safe by envisioning how the acci-
dent was almost averted.
In
summary,
future
research
on the psychology
of
counterfactual thinking and its interaction with the struc-
ture and content
of
domains may provide further insights
into its role in adaptive learning and other functions. These
insights, in turn, may elucidate the psychological pro-
cesses underlying consequential patterns
of
decision mak-
ing after accidents that have been described by organi-
zational researchers.
REFERENCES
CARROLL,
J. (1995). Incident reviews in high-hazard industries: Sense
making and learning. Industrial & Environmental Crisis Quarterly,
9,175-198.
DORNER,
D. (1997). The logic offailure: Recognizing and avoiding
error
in complex situations. Reading, MA: Addison-Wesley.
GILBERT,
D. T., &
MALONE,
P.
S. (1995). The correspondence bias. Psy-
chological Bulletin,
117,21-38.
GIROTTO,
v.,
LEGRENZI,
P., & RIZZO, A. (1991). Event controllability
in counterfactual thinking. Acta Psychologica, 78, 111-133.
"IF ONLY" THOUGHTS AND REMEDY DECISIONS 585
GLADWELL,
M. (1996, January 22). Blowup. The New
Yorker,
pp. 32-36.
JERVIS,
R. (1996). Counterfactuals, causation, and complexity. In P. E.
Tetlock
& A.
Belkin
(Eds.), Counterfactual thought experiments in
world politics
(pp. 309-316). Princeton:
Princeton
University Press.
KAHNEMAN,
D., &
MILLER,
D. T. (1986).
Norm
theory:
Comparing
re-
ality to its alternatives.
Psychological Review, 93, 136-153.
KAHNEMAN,
D., &
TVERSKY,
A. (1982). The
simulation
heuristic. In
D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under un-
certainty: Heuristics and biases
(pp.
201-208).
New
York:
Cam-
bridge University Press.
KELLEY,
H. H.
(1967).
Attribution
theory
in
social
psychology.
In
D.
Levine (Ed.), Nebraska Symposium on Motivation (pp. 192-238).
Lincoln:
University
of
Nebraska Press.
MARCH,
J. G. (1994). A primer on decision making: How decisions hap-
pen.
New York: Macmillan.
MARCH,
J. G., &
SIMON,
H. A. (1958). Organizations. New York: Wiley.
MARCH,
J. G.,
SPROULL,
L. S., &
TAMUZ,
M. (1991).
Learning
from
samples
of
one or fewer. Organizational Science, 2, 1-13.
MeGILL,
A. L. (1989). Context effects in
judgments
of
causation. jour-
nal
of
Personality & Social Psychology, 5, 189-200.
MOORE,
P.
C,
&
THOMSEN,
J. (1996). Reliving the past through com-
putational simulation: Testing the accuracy oforganizational coun-
terfactual thinking.
Unpublished manuscript.
MoRRIS, M.
w.,
&
MOORE,
P. C (1998). Learning from a brush with
danger.' Evidence that pilot Iearningfrom dangerous incidents is en-
abled by counterfactual thinking and hindered by organizational ac-
countability
(Research
Paper No. 1492).
Stanford
University, Grad-
uate School
of
Business.
MORRIS,
M.
w.,
SIM, D. L. H., &
MOORE,
P. C. (1999). Erroneous "if
only" thoughts: Heuristics
of
counterfactual thinking lead to errors in
diagnosing accidents.
Unpublished manuscript.
NATIONAL
TRAFFIC
SAFETY
BOARD
(1990). Annual review
of
aircraft ac-
cident data.
us. air carrier operations calendar year 1987 (Report
No.
NTSB/ARC-90/01,
Government Accession No. PB 91/119693).
Washington,
DC: Author.
NORMAN,
D. A.
(1990,
April 11). Cognitive science in the cockpit. Paper
presented at the Aerospace
Human
Factors
Symposium,
NASA
Ames
Research Center, Moffet Field, CA.
NORMAN,
D. A. (1992). Turn signals are thefacial expressions
of
auto-
mobiles.
Reading, MA: Addison-Wesley,
PERROW,
C
(1981).
Normal
accident
at
Three
Mile
Island. Society,
18(5), 17-26.
PERROW,
C. (1984). Normal accidents: Living with high risk systems.
New York:
Basic
Books.
REASON,
J. (1990). Human error.
Cambridge:
Cambridge
University
Press.
ROESE,
N. J. (1994).
The
functional basis
of
counterfactual
thinking.
journal
of
Personality & Social Psychology, 66, 805-818.
ROEsE, N. J. (1997). Counterfactual thinking.
Psychological Bulletin,
121, 133-148.
ROESE,
N. J., &
OLSON,
J. M. (1995). Functions
of
counterfactual think-
ing. In N.
1.
Roese
& J. M. Olson (Eds.), What might have been: The
socialpsychology
of
counterfactual thinking (pp. 169-197). Mahwah,
NJ: Erlbaum.
SHERMAN,
S. J., &
MCCONNELL,
A. R. (1995). Dysfunctional implica-
tions
of
counterfactual
thinking:
When
alternatives to reality fail us.
In N
J. Roese & J. M. Olson (Eds.), What might have been: The so-
cial psychology
of
counterfactual thinking (pp. 199-231). Mahwah,
NJ: Erlbaum.
SIM,D L. H.,
&
MORRIS,
M. W. (1998). Representativeness in counter-
factual thinking: The principle that antecedent
and
outcome
corre-
spond
in magnitude. Personality & Social Psychology Bulletin, 24,
595-609.
SPECTER,
M. (1997, August 20). Jeers sting Mir mission control which be-
moans a money pinch.
The New YorkTimes, p. A I.
ThTLOCK,
P. E. (1998). Close-call counterfactuals and belief-system de-
fenses: I was not almost
wrong
but
I was almost right. Journal
of
Per-
sonality & Social Psychology, 75, 3, 639-652.
VAUGHAN,
D. (1996). The Challenger launch decision: Risky technol-
ogy, culture. and deviance at NASA.
Chicago, IL: University
of
Chicago
Press.
WELLS,
G. L.,
TAYLOR,
B. R., & TuRTLE, J. W. (1987).
The
undoing
of
scenarios. Journal
of
Personality & Social Psychology, 53, 421-430.
NOTES
1.
Given
these limitations, a natural question,
of
course, is why orga-
nizations
promulgate the procedure
of
reconstructing accidents.
Some
scholars
have
argued
that these
procedures
act as
"rituals
of
reassur-
ance"
that
bracket
off
the
accident
as a
singular
event
and
isolate re-
sponsibility
to narrow factors within
that
event, thereby restoring
con-
fidence
in the
safety
of
normal
operations
(Gladwell, 1996; Perrow,
1984).
2.
It is
worth
clarifying the relation between human-factor focus in
"if
only"
thinking
about
accidents
(Girotto, Legrenzi, & Rizzo, 1991)
and
the
correspondence
bias in
thinking
about
causes
of
action (Gilbert &
Malone, 1995). A similarity is that both
biases
involve an exaggeration
of
the
extent
to which
outcomes
arise from
human
actors as
opposed
to
their
environments. The two biases differ, however, in that
human
ac-
tions
are
the
"effect"
to be
explained
in the
correspondence
bias,
whereas
they
are the
"cause"
ascribed
to
explain
accidents in
"if
only"
thinking.
3.
Of
course, in this example, the
option
of
replacing the
human
ac-
tors
may
have been the only
option
that
seemed
economically feasible.
In
our
experimental materials, we
ensured
that the
economic
consider-
ations
would
not pull in favor
of
either
human
or
nonhuman
remedies.
4.
Although
the
primary
purpose
of
the pilot study was to assess par-
ticipants'
endorsement
of
human-focused
and
environment-focused
remedies
for the accident problem, we also explored several
other
fac-
tors to
check
their impact on
participants'
reactions to the vignette. To
check
an interpretation
based
on an
implied
demand, we varied
whether
the
change
to a new
product
was
said
to have
"required"
or merely to
have
"involved"
changes
to the work process. To check an interpretation
based
on
implied
comparison
cases (McGill, 1989), we varied the pres-
ence
of
information
about
past accidents.
Half
of
participants saw the
story
above, which
makes
no mention
of
previous accidents, whereas
the
other
half
saw a story with the
added
detail that similar accidents
had
occurred
in the past,
before
the redesign.
If
the selection
of
a
causal
background
is critical to
participants'
judgments
about this case,
then
information
about
past
accidents
should
change
their
question
from
"why
the accident
occurred
now
and
not
in the past" to
"why
the acci-
dent
happened
to Paul
and
not to
other
employees," shifting causal focus
from the
work
environment to the
worker's
behavior, Analyses
of
the
interactions
ofthese
variations
with
the within-subject variable revealed
that
participants'
perception
of
the
superior
efficacy
of
changes
to the
environment
was not
moderated
by
these
variations.
That
is, it
made
no
difference
whether
adaptation to the new
product
was
said
to have
"re-
quired"
or merely to have
"involved"
the change to the
manufacturing
process
[F(l,50)
= .63, n.s.]. Likewise, it
made
no difference
whether
the
story
mentioned
that
accidents
had
also
occurred
in the
past
[F(l
,50) = .80, n.s.].
5.
Although
caution is always due in
arguing
from results
supporting
the
null
hypothesis,
most
of
the reasons for this caution do not apply to
the
argument
we derive from the
contrasting
results
of
Experiments
2A
and
28.
Because
these studies differed
only
in that we inverted the ma-
nipulated
and
dependent variables, the null result
cannot
simply reflect
a
lack
of
clarity in the experimental materials.
Nor
is a
lack
of
statisti-
cal
power
a compelling
account
because
Experiment
2B has a larger N
than
Experiment
2A, yet has a
much
smaller
effect size.
(Manuscript received
June
25, 1998;
revision accepted for
publication
August
4, 1999.)
... The idea that accidents are a gap within a regulatory system which needs to be established is common. Morris et al (1999) [4] demonstrated that decision-makers have a tendency to blame human errors and not technological system failures in the event of accidents when they have to give an answer like "if this …. then the accident would not have happened". ...
... The idea that accidents are a gap within a regulatory system which needs to be established is common. Morris et al (1999) [4] demonstrated that decision-makers have a tendency to blame human errors and not technological system failures in the event of accidents when they have to give an answer like "if this …. then the accident would not have happened". ...
Article
Full-text available
In modern aeronautical systems, one of the biggest challenges for the management structures is to maintain the control at all levels. Operational safety and efficiency impose the need to control all the associated risks and hazards; thus, in order to achieve organizational performance, a very important aspect is to establish and develop a strong organization with respect to operations and objectives. Nevertheless, performance cannot be achieved without control; the continuous technological development and the environmental variabilities have a great impact on the organizational management processes. Organizations are very complex and they will continue to expand due to the increasing demands of flight operations. The capacity to adapt, considering the permanent transformations in the society, represents a continuous process that needs to be carefully carried in order to diminish or eliminate the errors that may occur due to organizational factors. Controlling each operational step from the beginning represents the premises for obtaining stability and performance.
... Human error is frequently determined to be causal in workplace incidents [88,89]. Yet, professional workplace investigators have been shown to have a human error bias in which they tend to allocate a disproportionately greater amount of responsibility for the event to worker action than the evidence would support (e.g., [34,90,91], cf. ...
Preprint
Full-text available
Psychological research demonstrates how our perceptions and cognitions are affected by context, motivation, expectation, and experience. A mounting body of research has revealed the many sources of bias that affect the judgments of experts as they execute their work. Professionals in such fields as forensic science, intelligence analysis, criminal investigation, medical and judicial decision-making find themselves at an inflection point where past professional practices are being questioned and new approaches developed. Workplace investigation is a professional domain that is in many ways analogous to the aforementioned decision-making environments. Yet, workplace investigation is also unique, as the sources, magnitude, and direction of bias are specific to workplace environments. The workplace investigation literature does not comprehensively address the many ways that the workings of honest investigators' minds may be biased when collecting evidence and/or rendering judgments; nor does the literature offer a set of strategies to address such happenings. The current paper is the first to offer a comprehensive overview of the important issue of cognitive bias in workplace investigation. In it I discuss the abilities and limitations of human cognition, provide a framework of sources of bias, as well as, offer suggestions for bias mitigation in the investigation process.
... Morris and Moore (2000) found that generating counterfactuals about flying near misses could increase pilots' intentions to perform incident-avoiding behaviors. Relatedly, Morris et al. (1999) found that counterfactual thinking can enhance adaptive learning and motivation to think about self-improvement solutions. A more recent empirical study demonstrated that counterfactual thinking could enhance individuals' knowledge by facilitating learning from past events, which in turn could lead to safer behavior (He et al., 2020). ...
Article
Full-text available
Safety climate evolves in reflection of the effectiveness of organizational safety management efforts. The present study identified patterns of safety climate change over time and examined the role of error disclosure climate and counterfactual sharing in relation to safety climate change patterns. Online surveys were administered in a Chinese hospital three times at approximately 1-month intervals; the final sample included 451 healthcare workers nested within 62 teams. A latent growth mixture modeling approach was adopted to identify representative patterns of safety climate change at both individual and team levels and the predictors of those patterns. Three patterns of safety climate trajectories, declining (16%), improving (39%), and maintaining (45%), were identified at the individual level. Positive error disclosure climate and counterfactual sharing were significantly associated with increased probability of membership in the improving and maintaining trajectories compared to the declining trajectories. Counterfactual sharing mediated the relation between error disclosure climate and membership of safety climate trajectories. At the team level, two patterns of safety climate change, declining (15%) and improving (85%), were identified. Team counterfactual sharing was significantly associated with increased probability of membership in the improving trajectories compared to the declining trajectories. The current study demonstrated that an open and non-judgmental culture and the practice of sharing errors can contribute to improving safety climate over time.
... This seems especially likely when safety deliberations end following reductive, human error attributions for accident causes (e.g., Holden, 2009). Research has shown that people tend to target human errors for corrective measures following accidents (Morris et al., 1999), perhaps because these types of human-focused solutions come to mind more readily than systems-focused solutions (Coombs & Holladay, 2010;Holden, 2009). Further, misguided understandings of human error may affect the way incidents are reported and investigated (e.g., Woodcock, 2008;Woodcock et al., 2005) and further obscure effective recourse options that might arise from system-centered approaches (Dekker, 2011), resilience approaches (Woods & Hollnagel, 2017), or other emerging paradigms (e.g., Stanton & Harvey, 2017). ...
Article
Full-text available
Attributions of the causes of accidents to human error are problematically reductive, yet such attributions persist in media coverage. Few experiments have examined how human error attributions affect people’s perceptions. An experiment compared attributions of accidents to “human error” versus other causes (“mechanical failure,” “technical error,” or “computer error”). Participants (N = 971) from an online sample read one of 50 real news excerpts describing accidents from a broad array of domains (e.g., aviation, automobiles, manufacturing, and infrastructure, among others). Stories kept the same or similar details, with only the causal attribution altered to compare human error to other causes. With human error attributions, participants were in greater agreement with the statement that an individual deserved to be punished for the accident and in less agreement that an organization or company was responsible for the accident. People also perceived past human error accidents to have been more preventable, although ratings of prospective preventability were not significantly different for human error versus other attributions. The idiosyncratic details of particular accidents contributed more variance to perceptions than the causal attribution. The same pattern of results was replicated in a second experiment (N = 1195), and new analyses found no evidence that the relationship between causal attributions and perceptions was moderated by the personal relevance of the news story. Our findings suggested that, when an accident is attributed to human error in media, the public may be less likely to expect examination or mitigation of systemic shortcomings (e.g., in design, organizational practices, etc.) that precipitate accidents.
... This seems especially likely when safety deliberations end following reductive, human error attributions for accident causes (e.g., Holden, 2009). Research has shown that people tend to target human errors for corrective measures following accidents (Morris et al., 1999), perhaps because these types of human-focused solutions come to mind more readily than systems-focused solutions (Coombs & Holladay, 2010;Holden, 2009). Further, misguided understandings of human error may affect the way incidents are reported and investigated (e.g., Woodcock, 2008;Woodcock et al., 2005) and further obscure effective recourse options that might arise from system-centered approaches (Dekker, 2011), resilience approaches (Woods & Hollnagel, 2017), or other emerging paradigms (e.g., Stanton & Harvey, 2017). ...
Preprint
Full-text available
Attributions of the causes of accidents to human error are problematically reductive, yet such attributions persist in media coverage. Few experiments have examined how human error attributions affect people’s perceptions. An experiment compared attributions of accidents to “human error” versus other causes (“mechanical failure,” “technical error,” or “computer error”). Participants (N = 971) from an online sample read one of 50 real news excerpts describing accidents from a broad array of domains (e.g., aviation, automobiles, manufacturing, and infrastructure, among others). Stories kept the same or similar details, with only the causal attribution altered to compare human error to other causes. With human error attributions, participants were in greater agreement with the statement that an individual deserved to be punished for the accident and in less agreement that an organization or company was responsible for the accident. People also perceived past human error accidents to have been more preventable, although ratings of prospective preventability were not significantly different for human error versus other attributions. The idiosyncratic details of particular accidents contributed more variance to perceptions than the causal attribution. The same pattern of results was replicated in a second experiment (N = 1195), and new analyses found no evidence that the relationship between causal attributions and perceptions was moderated by the personal relevance of the news story. Our findings suggested that, when an accident is attributed to human error in media, the public may be less likely to expect examination or mitigation of systemic shortcomings (e.g., in design, organizational practices, etc.) that precipitate accidents.
... Folger and Cropanzano's (1998) general theory of fairness posits that stakeholders consider the locus of control in attributing responsibility for the crisis. Limited research shows a general bias toward internal causes for accidents (Morris et al. 1999). ...
Article
Full-text available
This study argues that actions taken during a crisis within the responsibility compass impacts reputation. The lens of responsible leadership was chosen from among the different foci of leadership scholarship because of its emphasis on relational and ethical dimensions. The focus of this study was the actions undertaken by the CEOs of United and Delta Airlines after the forceful removal of the passengers and its influence on the reputation capital as measured by the changes in the respective share prices of the airlines. The overarching question was whether there was clear statistical evidence that linked the actions undertaken as part of the crisis management efforts and the reputation of the airlines. Findings from this study suggest that actions taken as part of responsible leadership compass explained the variations in the share prices. In the case of United Airlines almost 70% of the variations in the share prices could be explained by responsible leadership, whereas in the case Delta airlines only 50.6% of the variations in the share value could be predicted based on the role of leadership in crisis management.
... Examining workplace safety behaviors, generating upward selffocused counterfactuals about a near miss flying incident increased pilots' intentions to perform incident-avoiding behaviors (Morris & Moore, 2000). Testing organizational incident remedies showed that generating counterfactuals increased the likelihood of proposing personal-behavior focused remedies, rather than environmental/organizational-focused ones (Morris, Moore, & Sim, 1999). These reflect that counterfactual thinking can sometimes promote adaptive learning and internal or self-focused rather than external or other-focused solutions. ...
Article
Introduction: Information processing theories of workplace safety suggest that cognition is an antecedent of safety behavior. However, little research has directly tested cognitive factors as predictors of workplace safety within organizational psychology and behavior research. Counterfactuals (cognitions about "what might have been") can be functional when they consist of characteristics (e.g., "upward' - focusing on better outcomes) that alter behavior in a manner consistent with those outcomes. This field study aimed to examine the influence of counterfactual thinking on safety behavior and explanatory mechanisms and boundary conditions of that relationship. Method: A sample of 240 medical providers from a hospital in China responded to three surveys over a four-month time frame. Results: Results showed that upward counterfactuals were positively related to supervisor ratings of safety compliance and participation. These relationships were mediated by safety knowledge but not by safety motivation. Upward counterfactuals were more strongly related to safety behavior and knowledge than downward counterfactuals. As expected, safety locus of control strengthened the mediating effects of safety knowledge on the relationship between upward counterfactuals and safety behavior. Conclusions and Practical Applications: The findings demonstrated that counterfactual thinking is positively associated with safety behavior and knowledge, thus expanding the variables related to workplace safety and laying some initial groundwork for new safety interventions incorporating counterfactual thinking.
Article
Psychological research demonstrates how our perceptions and cognitions are affected by context, motivation, expectation, and experience. A mounting body of research has revealed the many sources of bias that affect the judgments of experts as they execute their work. Professionals in such fields as forensic science, intelligence analysis, criminal investigation, medical and judicial decision-making find themselves at an inflection point where past professional practices are being questioned and new approaches developed. Workplace investigation is a professional domain that is in many ways analogous to the aforementioned decision-making environments. Yet, workplace investigation is also unique, as the sources, magnitude, and direction of bias are specific to workplace environments. The workplace investigation literature does not comprehensively address the many ways that the workings of honest investigators' minds may be biased when collecting evidence and/or rendering judgments; nor does the literature offer a set of strategies to address such happenings. The current paper is the first to offer a comprehensive overview of the important issue of cognitive bias in workplace investigation. In it I discuss the abilities and limitations of human cognition, provide a framework of sources of bias, as well as, offer suggestions for bias mitigation in the investigation process.
Article
Full-text available
Online data breaches are recurrent and damaging cyber incidents fors organizations worldwide. This study examines how organizations can effectively mitigate reputational damages in the aftermath of data breaches by hacking, through situational crisis communication strategies. Comparable data breach crises do not have an equally negative impact on organizational reputation. Base responses such as comprehensive and exhaustive guidelines, and detailed explanations about the incident to consumers helped to reduce the damage. Corporations responding to data breaches by hacking benefit from admission of responsibility in spite of the initial characterization of such crises as victim crisis types. Organizations that primarily relied on one single strategy, performed better than those that inconsistently blended strategies. Particularly denial was ultimately detrimental to organizational reputation. Self-disclosure allowed companies to positively influence media reporting. Social media communication did not play an important role in the response of the organizations involved. The consistent and timely adoption of compensation, apology, and rectification strategies, combined with reinforcing strategies such as ingratiation and bolstering, positively influenced reputational recovery from the crisis.
Chapter
Patient safety has become a priority of all healthcare systems. Global policies have been implemented to reduce the number of incidents related to patient safety. There are still many patients who suffer damages derived from healthcare interventions. The impact of such events extends not only to family members but also to the professionals themselves and to health institutions creating a significant economic burden. Most studies have been directed to the epidemiology of adverse events (AEs), an intent to analyze their causes and their consequences. Many safe practices have been promoted to reduce these risk events. Recently, aspects related to post-AE performance have become more relevant. Risk management involves, among other actions, the identification, notification, and analysis of the AE that point out potential failures in the system and the root cause with the ultimate goal of establishing improvement actions and avoiding their recurrence. Gradually these strategies have been incorporated into institutional policies, thereby improving the culture of safety. However, there are still parts of the process that require research and development such as disclosing AE to patients and families and supporting professionals after being involved in an AE.
Article
Counterfactuals are mental representations of alternatives to the past and produce consequences that are both beneficial and aversive to the individual. These apparently contradictory effects are integrated into a functionalist model of counterfactual thinking. The author reviews research in support of the assertions that (a) counterfactual thinking is activated automatically in response to negative affect, (b) the content of counterfactuals targets particularly likely causes of misfortune, (c) counterfactuals produce negative affective consequences through a contrast-effect mechanism and positive inferential consequences through a causal-inference mechanism, and (d) the net effect of counterfactual thinking is beneficial.
Article
It is hypothesized that causal explanations for an occurrence vary as a function of the causal background against which the occurrence is considered. Three experiments are presented that test propositions regarding the operation of the causal background in the selection of causal explanations. Findings indicate that factors previously shown to affect subjects' attributions-specifically, role (actor vs. observer), covariation information (consensus and distinctiveness), and quality of performance (positive vs. negative)-may do so by guiding subjects' selection of a causal background. Evidence indicates that these factors may not have the predicted effect on subjects' attributions when competing cues, such as context or wording of the causal question, suggest the relevance of conflicting causal backgrounds.
Article
Presents a theory of norms and normality and applies the theory to phenomena of emotional responses, social judgment, and conversations about causes. Norms are assumed to be constructed ad hoc by recruiting specific representations. Category norms are derived by recruiting exemplars. Specific objects or events generate their own norms by retrieval of similar experiences stored in memory or by construction of counterfactual alternatives. The normality of a stimulus is evaluated by comparing it with the norms that it evokes after the fact, rather than to precomputed expectations. Norm theory is applied in analyses of the enhanced emotional response to events that have abnormal causes, of the generation of predictions and inferences from observations of behavior, and of the role of norms in causal questions and answers. (3 p ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Organizations learn from experience. Sometimes, however, history is not generous with experience. We explore how organizations convert infrequent events into interpretations of history, and how they balance the need to achieve agreement on interpretations with the need to interpret history correctly. We ask what methods are used, what problems are involved, and what improvements might be made. Although the methods we observe are not guaranteed to lead to consistent agreement on interpretations, valid knowledge, improved organizational performance, or organizational survival, they provide possible insights into the possibilities for and problems of learning from fragments of history.
Article
Normal Accidents analyzes the social side of technological risk. Charles Perrow argues that the conventional engineering approach to ensuring safety--building in more warnings and safeguards--fails because systems complexity makes failures inevitable. He asserts that typical precautions, by adding to complexity, may help create new categories of accidents. (At Chernobyl, tests of a new safety system helped produce the meltdown and subsequent fire.) By recognizing two dimensions of risk--complex versus linear interactions, and tight versus loose coupling--this book provides a powerful framework for analyzing risks and the organizations that insist we run them. The first edition fulfilled one reviewer's prediction that it "may mark the beginning of accident research." In the new afterword to this edition Perrow reviews the extensive work on the major accidents of the last fifteen years, including Bhopal, Chernobyl, and the Challenger disaster. The new postscript probes what the author considers to be the "quintessential 'Normal Accident'" of our time: the Y2K computer problem.