See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/271080745

Cognitive fallacies and criminal investigations

ARTICLE in SCIENCE & JUSTICE · JANUARY 2015

Impact Factor: 1.42 · DOI: 10.1016/j.scijus.2014.12.007

READS

54

1 AUTHOR:

Hans Ditrich
University of Vienna
28 PUBLICATIONS 132 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Hans Ditrich
letting you access and read them immediately. Retrieved on: 15 February 2016
SCIJUS-00498; No of Pages 5
Science and Justice xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Science and Justice

journal homepage: www.elsevier.com/locate/scijus

Cognitive fallacies and criminal investigations

Hans Ditrich
Department I/9 SIAK, Institute for Science and Research, Federal Ministry of the Interior, Herrengasse 7, Vienna 1014, Austria

a r t i c l e i n f o a b s t r a c t

Article history: The human mind is susceptible to inherent fallacies that often hamper fully rational action. Many such miscon-
Received 15 May 2014 ceptions have an evolutionary background and are thus difficult to avert. Deficits in the reliability of eye-
Received in revised form 15 December 2014 witnesses are well known to legal professionals; however, less attention has been paid to such effects in crime
Accepted 17 December 2014
investigators.
Available online xxxx
In order to obtain an “inside view” on the role of cognitive misconceptions in criminalistic work, a list of fallacies
Keywords:
from the literature was adapted to criminalistic settings. The statements on this list were rated by highly experi-
Criminal investigations enced crime scene investigators according to the assumed likelihood of these errors to appear and their severity
Cognitive errors of effect. Among others, selective perception, expectation and confirmation bias, anchoring/“pars per toto” errors
Error management and “onus probandi”—shifting the burden of proof from the investigator to the suspect—were frequently consid-
ered to negatively affect criminal investigations.
As a consequence, the following measures are proposed: alerting investigating officers in their training to cogni-
tive fallacies and promoting the exchange of experiences in peer circles of investigators on a regular basis. Fur-
thermore, the improvement of the organizational error culture and the establishment of a failure analysis
system in order to identify and alleviate error prone processes are suggested.
© 2014 Forensic Science Society. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction concerning the implications of cognitive fallacies in crime investigators

There can be no doubt that every suspect is entitled to the highest
possible standards of investigation in a modern legal system. Despite
potential constraints from budgetary restrictions or political trends,
such a high-level-approach to criminal examinations forms an integral
part of a fair trial and thus is a basal human right. However, even taking
best intentions for granted, errors do occur at all levels and are inherent-
ly unavoidable (e.g., [1]). This fact is often attributed to “Murphy's law”
but was already accepted in ancient Rome. However, while Seneca's
(4 BC–65 AC) aphorism, “errare humanum est—to err is human”—is
common knowledge, the rest of this sentence “… sed persevare
diabolicum—to persist (in error) is diabolical”—is less often cited.
Any attempt at improvement requires intense investigation of the
causative factors (e.g., [2,3]) that lead to an error. Instead, the first im-
pulse of individuals and organizations is often to cover up eventual mis-
takes. Such a “denial behavior” can itself already be regarded as a
cognitive distortion strategy (e.g., [4,5]). Inappropriate error-handling
strategies not only violate individual rights, with often tragic conse-
quences, but also negate the basis of scientific progress [6]—gaining in-
formation from eliminating mistakes.
The unreliability of human perception had already been demonstrated
for eye-witnesses by Franz von Liszt over a century ago. However, studies
E-mail address: hans.ditrich@bmi.gv.at.
are gaining attention in the last decades (see, e.g., [7] for references).
There are several, partly conflicting, error-handling strategies. In the
production of goods or services, a reduction of errors (defined as varia-
tions), preferably to zero, is the goal of quality management and quality
assurance programs. Instead, an open approach to mistakes and a positive
attitude that reduces personal fears of admitting them is often promoted
in organizations that are devoted to innovation, research and develop-
ment. The latter often favor a position that regards errors as an opportu-
nity for improvement. Accordingly, flexibility in problem management
and collective solving strategies, as well as successful handling of unex-
pected complications, are regarded as assets to be refined [8,9]. None of
these approaches seems ideally suited for criminalistic investigations.
Formal quality management systems by their very nature can only
measure aberrations from a “perfect” prototype. Consequently, such
control systems are aiming at highest homogeny. The establishment of
quality standards (e.g., [10]) is certainly essential in criminalistic work.
However, merely following some methodical guidelines might be insuf-
ficient for reaching an accurate conclusion even in rather small, techni-
cal fields of forensic investigations (e.g., gunshot residue analysis [11]).
In the classical scientific process, the peer-review system, as well as
follow-up studies, serves as error correction measures. However, this
approach is usually unfeasible in forensic investigations [12]. For medi-
cal procedures and doubtful cases of death, autopsies may be regarded
as an (ultimate) quality control.
Generally, in the legal system, the well-established practice of error
correction involves appeal and a new hearing. Nevertheless, this

http://dx.doi.org/10.1016/j.scijus.2014.12.007
1355-0306/© 2014 Forensic Science Society. Published by Elsevier Ireland Ltd. All rights reserved.

Please cite this article as: H. Ditrich, Cognitive fallacies and criminal investigations, Sci. Justice (2014), http://dx.doi.org/10.1016/
j.scijus.2014.12.007
2 H. Ditrich / Science and Justice xxx (2014) xxx–xxx
approach also has its eminent drawbacks. First, it requires knowledge
(and proof) of where, when and in which context a mistake happened
as the burden of proof rests on the appellant. Second, a certain financial
capacity and sufficient time are required, the latter being a resource that
is normally not refundable (e.g., child custody, business options, etc.).
Consequently, while recognizing that 100% perfection is not sustain-
able—even if it might be achieved temporarily—the establishment of a
supplementary system of error handling at the organizational level
might further improve the overall quality of the legal system.
2. Incentives
Mistakes may be divided into several categories. Many technical or
structural mistakes can be avoided by organizational measures, such
as better equipment, modified operating rules or improved training.
A similar approach pertains to qualification/competence/profession-
alism issues. High levels of training and education are essential and have
to be provided, ensured and continuously refined. However, personal
experience and mental agility are essential qualities (compare [13]),
that unfortunately can be transferred by formal teaching only to a limit-
ed extent. Personal integrity and commitment of the investigating offi-
cers are additional qualities of central significance. To select for, keep,
maintain and monitor such characteristics are demanding challenges
for both the investigating officer and his/her organization.
In addition to such demands, an often underestimated problem that
afflicts successful problem solving at the individual (and consequently
also the organizational) level are cognitive failures that subconsciously
influence the perception and evaluation of evidence. However, as forensic
investigations—like any other—involve human assessment and interpre-
tation, the cognitive premises of human cognition must not be ignored.
Numerous examinations relating to erroneous action and their
prevention have been carried out in the fields of aviation (see
e.g., [14–17]), clinical medicine (e.g., [18–22]) and also in general man-
agement (e.g., [9]). Systematic error analysis is also often carried out for
sites where an accumulation of traffic accidents is encountered. In con-
sequence, architectural measures (road design, traffic signs/lights, mir-
rors, roundabouts, etc.) are implemented independently of the eventual
personal faults (e.g., [23,24]).
Still, there are comparatively few studies on the influence of the
individual's cognitive disposition to forensic problem solving. However,
certain aspects such as decision making [25,26] or the coordination of
experts at crime scenes [27], and different aspects of criminological
work like the interpretation of fingerprints [28–31] or the reliability of
person identification by witnesses (e.g., in CCTV recordings, line ups
or photo-collections) have been investigated [32–35].
3. Assessment
The professional opinion of highly experienced crime scene officers
was obtained for assessing the relevance of cognitive bias for criminal
investigations. These officers had a minimum of 10 years, mostly over
20 years of professional practice and are in charge of investigation
teams as leading officers.
A list of cognitive errors in medical diagnosis [18] was abridged and
adapted to criminal scenarios. The resulting list of statements was
rated by the officers according to their personal opinion on a five grade
verbal scale from appearing “very often” to “never.” The officers were
asked to select those problematic concepts that they regard as most com-
monly appearing and having the strongest adverse effect on practical
criminalistic work based on their expert knowledge. Accordingly, the rat-
ings are based solely on the individual experience of the investigators.
No ranking or quantitative interpretation can be derived from these
statements as the number of available individuals was too small to allow
statistically significant evaluations. Still, using a structured approach,
the experience of qualified investigators may provide useful “inside in-
formation” to assess caveats in their professional duties. The problem
Please cite this article as: H. Ditrich, Cognitive fallacies and criminal investigations, Sci. Justice (2014), http://dx.doi.org/10.1016/
j.scijus.2014.12.007
awareness of practitioners may or may not represent the most critical
sources of errors but certainly can be regarded as a substantial source
of information. Nevertheless, it should be kept in mind that categoriza-
tions are intrinsically theoretical models and that real world mistakes
may be the combined product of several fallacies.
3.1. Selective perception, expectation and confirmation bias—the tendency
to search for, agree with or interpret information in a way that confirms
one's own presumptions
This mental bias is common to nearly everyone. However, every pre-
caution should be taken to avoid this mistake in a criminal investigation
(see also e.g., [36,37]). While it might not always be favorable for per-
sonal self-esteem to question, critically review and eventually declare
one's own hypotheses to be false, this is certainly indispensable—not
only in forensic work but in every knowledge gaining process (e.g.,
[6]). In crime scene work, an overlooked or disregarded piece of evi-
dence is often lost forever. Consequently, traces that are seemingly irrel-
evant or even disagree with the “working hypothesis” also deserve
careful documentation and preservation.
Concealing exculpatory evidence from the defendant and the court
was reported as the most common serious form of official misconduct
[2, p. 66]. However, whether such deficiencies are the result of willful
misconduct or a consequence of negligent action is a matter of liability,
though initially of lesser importance to the individual under suspicion.
Still, if such negligence is a consequence of cognitive fallacies there
might be procedural measures to avoid it.
3.2. Anchoring/“pars per toto”—overestimating the validity of (even inap-
propriate) partial information, i.e., to “anchor” a decision on a previously
registered detail while disregarding new, supplementary facts
Professional experience is undoubtedly an important asset. Previ-
ously collected information concerning other issues can profoundly
support the process of finding and evaluating evidence. However, it is
rather tempting to conclude the following: “When I see a bird that
walks like a duck and swims like a duck and quacks like a duck, I call
that bird a duck.”1 This theorem is often valid, but it should be kept in
mind that there are about 150 different species in the family of ducks
(Anatidae). Accordingly, decisions based on a few, familiar features
may probably be correct very often (and time saving, thus “economic”)
but regrettably not always, as they might also ignore other crucial de-
tails (see also [38]).
In contrast, long tried and tested methods should not be dismissed.
Overconfidence in a—usually modern—technical method might also be
included in this group of conceptual failures. For example, at the begin-
ning of the century the collection of DNA, data was frequently consid-
ered as “ultima ratio” in criminalistics, thought to deliver doubt-free
evidence and thus rendering many other investigation methods outdat-
ed. However, this resulted in adverse effects when more traditional ap-
proaches were neglected. While DNA typing increasingly becomes a
standard technique, it is also demystified and more conventional
crime scene procedures are employed again.
3.3. Burden of proof/“onus probandi”—shifting the burden of proof from the
investigator to the suspect
Most legal systems respect the ancient roman legal rule “in dubio pro
rheo.” Even though this principle of “being innocent until proven guilty”
is occasionally ignored in prosecution, it is fundamental to all modern,
democratic, juridical frameworks. However, this standard is sometimes
even disregarded in the formulation of legal acts, for example, in the
fields of product liability or environmental safety issues. Shifting or
1
Quote attributed to the poet James W. Riley around 1883-1885.
 H. Ditrich / Science and Justice xxx (2014) xxx–xxx
reversing of the burden of proof regrettably often forms the rationale
behind large scale surveillance or public monitoring measures.
While some success may be attributed to such methods, it is logically
impossible and often also practically difficult to prove that something
“was not there” or “didn't happen.” The argument from ignorance2—
the absence of evidence is no evidence of absence …—may not always
lead to valid conclusions and certainly implies caution when evaluating
evidence.
Nevertheless, in any investigation, it is clearly the obligation of the
investigator to provide evidence for the guilt and/or innocence of a sus-
pect as part of the “fair trial” principle (e.g., §3 StPO in Austria).
3.4. The cognitive biases outlined above were picked as having the highest
de facto potential to influence a criminal investigation adversely. In addi-
tion, the following errors were also marked as likely to appear and obstruc-
tive to perceive facts
⁎ “Knowledge bias” should not be confused with the selective percep-
tion, expectation and confirmation bias complex mentioned before.
Instead, this error results from the tendency to choose the most fa-
miliar option, i.e., to “walk the well-known path” and thus seems
more related to anchoring bias. However, while it is highly probable
that a solution that already worked for several times before will also
be suited to solving future problems, better alternatives may be
overlooked due to a lack of flexibility.
Knowledge bias may also be a consequence of (inherently inflexible)
standard operating procedures (SOPs) that are often endorsed by
quality management systems. Nevertheless, standardization (and
certification) of operating procedures can only assure formal com-
pliance with an approved method for dealing with repetitive and
substantially homogenous processes. Thus, quality management
rules can only be applied to a certain limit to “new,” unforeseen
problems. While a certain standardization of the criminal investiga-
tion process is clearly indispensable, the question of how to provide
the highest investigative quality when dealing with “extraordinary”
criminal cases remains.
⁎ Thinking in stereotypes—“overgeneralization” can also be viewed as
a special form of the anchoring/“pars per toto” fallacy. Prejudices are
often derived from socio-cultural differences, thus discriminating
“us” from “the other” based on common, distinctive features.
When certain properties of individuals are generalized as character-
istic of the whole group, such an attitude may lead to inappropriate
reactions, breach of the principle of equality and acting out eventual
tendencies of xenophobia and racism. Conversely, “attractive”
suspects may face bias as well, e.g., in assessing their likelihood of
committing certain crime types [39] or in judgments of their ac-
countability [40]. This bias is not only an issue for juries or judges
but may also play a role in the initial investigations, e.g., in an inter-
rogation or a line-up.
⁎ Similar to the former problem are the “fallacy of division” and the “fal-
lacy of composition” biases. Properties that are found to be true in one
part of a complex issue do not necessarily imply that the whole com-
plex is correct. Vice versa, if a statement is accepted as being true, this
does not necessarily mean that all of its details must be true as well.
3.5. The following issues were not particularly marked in the survey but
were still regarded as problematic upon questioning, albeit considered as
less substantial in the initial steps of an investigation
Many widespread misconceptions—especially those regarding (sta-
tistical) probabilities, such as assuming a homogenous distribution of
data values (hasty generalizations, black swan blindness, gambler's
2
attributed to John Locke, 1689.
Please cite this article as: H. Ditrich, Cognitive fallacies and criminal investigations, Sci. Justice (2014), http://dx.doi.org/10.1016/
j.scijus.2014.12.007
3
fallacy, clustering error)—can be expected to adversely influence solving
a criminal case. This group of errors is based on positing that a hypoth-
esis based on a sequence of observations is true, even though the under-
lying data are insufficient (i.e., not enough observations were made).
Closely related to the latter misconceptions is the “search satisficing”
error. This is often promoted or even provoked by economic consider-
ations. Hence, if a search is called off as soon as a presumably conclusive
feature is discovered, resources are certainly spared but important evi-
dence might be lost. As such, a compromise—completeness versus effi-
ciency (and swiftness)—is required in every search.
All humans—not only criminal investigators—are prone to “cum/post
hoc ergo propter hoc” errors, i.e., accepting spurious correlations. How-
ever, neglecting the axiom that correlation is not equivalent to causation
has severe consequences in the interpretation of evidence.
For example, a personal trace, such as DNA or a fingerprint, can only
be considered as evidence that a suspect was in contact with a given ob-
ject (disregarding the possibility of secondary transfer for the moment).
To deduce a progression of events from such a trace remains a supposi-
tion based on probability (see also [41]). While the detection of causal-
ity is certainly the most important issue in all research, this subject
remains often rather speculative in some criminal investigations. Still,
attempts of fitting the evidence to the conclusions—or “Procrustes” at-
tempts at explanations—should be avoided.
4. External factors
Significant contributions to the identification of mistakes that oc-
curred during the investigation of a crime often come solely from the
side of the defendant/lawyer, by appealing, occasionally from the
media or also just from the wisdom of hindsight.
It is a common practice to attribute and explain failures on the basis
of personal incompetence. However, individual culpability is not always
sufficient to explain flaws in complex investigations that involve nu-
merous, apparently capable and often also highly experienced officers.
Nevertheless, some errors can be attributed to human failure of a single
investigator or a few individuals. Still, inadequate actions may originate
from widely different premises.
There can be no doubt that police work and especially crime scene
investigations impose high stress levels on the officers in charge. Right
from the beginning, in addition to the challenges of crime scene work it-
self, there is strong pressure from superiors, the public and the media, at
least in seemingly spectacular cases. High pressure to succeed lasts on
otherwise often under-supported units. This high stress level at a
crime scene, and also insufficient training or personal motives such as
burnout and resignation or, on the other hand, hubris and craving for
recognition, may lead to inadequate action. Pressure from media can
have strongly adverse effects on an investigation. If not handled proper-
ly, the request from journalists to achieve (and communicate) results
quickly almost inevitably provokes premature conclusions.
Some additional factors that can lead to failure have a methodical
background. Applying sometimes poorly understood new scientific
methods to an investigation may lead to “tunnel vision”—thus neglecting
other, maybe less trendy alternatives. In general, relying on authorities
without challenging the validity of their competence can be problematic.
Moreover, written sources should be evaluated critically, as not every
study is based on consensus and even correct information may lead to
wrong conclusions (e.g., anchoring). This is certainly difficult for anyone
who is unfamiliar with the specific subject; however, it should be re-
membered that an expert can only be an expert in a specific field.
Usage of statistical methods—while indispensable in many scientific
and practical questions—may have only limited benefits in criminal in-
vestigations. Even the highest levels of statistical significance can only
prove correlations and do not provide information on causalities. Forensic
reasoning on a statistical basis is restrained by the induction problem—
the “black swan fallacy.” Furthermore, objects to be statistically evaluated
4 H. Ditrich / Science and Justice xxx (2014) xxx–xxx
should ideally differ in only a very few properties, a condition that is nor-
mally not found when looking at criminal cases.
Obstacles to a successful investigation may also be generated from
within the organization. Frequently, intra-organizational pressures
such as the compulsion to “solve” a certain amount of cases—implying
sentencing as guilty—can result in a biased collection and evaluation
of evidence, thus disregarding “fair trial” and the principle of “innocent
until proven guilty” mentioned before (e.g., [34]).
It should also be kept in mind that in strongly hierarchical organiza-
tions such as the police, an unconventional, potentially appropriate or
creative approach at a lower level might be obstructed by superiors
for a variety of reasons. Furthermore, the Dunning–Kruger effect—
overestimating one's own competence [42]—may result in a “chain re-
action” that hinders innovative initiatives—especially under “bottom-
up” circumstances. The adverse consequences of the Dunning–Kruger
effect cannot be over-estimated—in organizations in general, as well
as in criminal prosecution services. The combination of incompetence
and hubris underlying this fallacy inevitably leads to selective percep-
tions or cognitive dissonance experiences, resulting in inappropriate ap-
preciation of circumstances and consequently in flawed actions.
5. Conclusions
It seems evident that, in many criminal cases that were not optimally
investigated, the crucial faults were made at the very beginning of the in-
vestigation, often at the crime scene itself. Most other, subsequent stages
of a criminal procedure can be reviewed, repeated and also corrected if
necessary. However, in crime scene work, each piece of evidence
that is initially neglected may be lost forever. Considering this pivotal po-
sition in a legal procedure, the question of how optimum performance in
the initial stage of an investigation can be assured seems fundamental.
In addition, a general improvement in error management for
criminalistic work seems desirable. Focusing on the possible effects of
cognitive biases, a few provisions may help in diminishing—if not
preventing—some of the negative consequences:
⁎) At the individual level, the crime scene investigators in charge or offi-
cers leading a criminal investigation should be made aware of the in-
fluence of cognitive biases. In accordance with their specific role in the
legal process, prosecutors, coroners, forensic medical experts and
judges might also benefit from pertinent training (compare [27]).
While it is unlikely that basic human cognitive predispositions can
easily be changed, it may be assumed that a certain attention towards
possibly misleading conceptions might be acquired by adequate train-
ing. Such an educational program module should point out the behav-
ioral roots of cognitive fallacies and include coaching as well as
practical training (e.g., video-controlled exercises and group feed-
back). Officers who participate in criminal investigations should be
made aware of their own cognitive premises. Once alerted, the inves-
tigators might be more mindful to the effects of otherwise fully un-
conscious influences. At least they should be made aware of the
most common bias, i.e., the conviction that oneself is free of biases.
⁎) Furthermore, officers from different investigative teams should
meet on a regular basis to present and openly discuss “peculiar”
cases—how they were treated and possibly solved, what were the
obstacles and how they were overcome. The usual (vertical) infor-
mation flow—reporting just to superiors, prosecutors and judges—
hinders the sharing of experiences and thus restricts the most valu-
able source of information for practitioners. This holds even true for
cases covered by the media, as accuracy and completeness of public
information is often insufficient for professional commitments. Such
a structured exchange of experience would likely contribute to the
“seven distinct critical skill sets for top performance in CSE (crime
scene examiners)” as postulated by Kelty et al. [13].
⁎) Attempting to improve the error culture at the level of an organiza-
tion is a rather complex task. Examinations of personal misconduct
Please cite this article as: H. Ditrich, Cognitive fallacies and criminal investigations, Sci. Justice (2014), http://dx.doi.org/10.1016/
j.scijus.2014.12.007
are usually carried out by superiors, special commissions or legal
bodies. Individual culpability normally results in disciplinary mea-
sures, liability or even penal law. Uncovered structural problems
can be dealt with comparatively easily, e.g., by refining training pro-
grams, modified operation rules, better equipment or improved su-
pervision.
However, as failures are not always just a consequence of faulty
equipment or individual mistakes, organizations should also critical-
ly evaluate some of their own premises. Hierarchical structure, uni-
formity as well as peer pressure are inherent characteristics of law
enforcement organizations (among others). Unfortunately, the
same factors are fundamental triggers for inappropriate actions. Ac-
cordingly, a corporate strategy should be developed that is open to
acknowledging mistakes—independently of possible liability, disci-
plinary or legal questions. Understandably, many experts—not only
in law enforcement—have a certain disposition to cover mistakes
in order not to harm the public image of their profession. This
attitude, however, may also hinder the rational analysis of errors
and consequently their mitigation. A policy of error denial—even
when based on considerations to maintain society's confidence in
the legal system—inhibits the improvement of investigative process-
es and discourages the critical evaluation of actions. Consequently, a
setting where errors are regarded as an incentive for improvement—
notwithstanding all personal implications—is required. This is
certainly a demanding task; however, numerous pertinent manage-
ment consulting companies or counselors for organizational culture
development can be recruited for implementing such a strategy.
⁎) Errors that already have been uncovered should be investigated by a
commission that is as far as possible apart from the regular hierarchy
where the error occurred. Members should conform to high profes-
sional standards but be from outside to avoid routine-blindness. Ob-
viously, influences from peer pressure as well as from superiors
must be prevented.
Such a commission should not be targeting personal culpability nor
should it attempt to repair a mistake, i.e., solve a case. The former
should remain in the competence of the established disciplinary au-
thorities; the latter would be “cold case management” and thus a
different task. Instead, the commission should focus on the premises
that resulted in an error, investigating mainly the operational and
structural causes.
Economic constraints may suggest that only spectacular, “impor-
tant” cases should be reviewed. However, everyday routine errors
might in particular be identified by an established error analysis process
(see e.g., [43]), resulting in an overall reduced error rate and conse-
quently better performance and higher efficiency in the long run. Anal-
ogous to, e.g., traffic safety, such studies may identify “hot spots” where
errors accumulate and thus provide the opportunity to neutralize such
“danger zones” by suitable measures. By recognizing and marking pro-
cesses that are particularly prone to errors, it might also be possible to
avoid small adverse effects that would otherwise be too “insignificant”
for official consideration. Although prominent cases make headlines in
the media, it is rather the small faults in a criminal process that influence
the performance of a legal system and also increase overall costs.
Establishing a board of error analysis (and improvement) might
therefore be well worth the effort with regard to investigative perfor-
mance as well as economically. Analogous to boards that are regularly
established for controlling disciplinary and financial matters in most
private and public administrations, it is suggested here that criminal in-
vestigations should be evaluated by a group of experts that is operating
independent from the regular organizational hierarchy on an institu-
tionalized basis.
Summarizing, it may be stated that criminal investigators are—like
all humans—prone to cognitive fallacies. Nevertheless, the professional
experience and problem awareness of senior officers is an important
 H. Ditrich / Science and Justice xxx (2014) xxx–xxx
resource of information. Several types of cognitive errors were marked
as having a high potential of adversely influencing criminal investiga-
tions. This first-hand knowledge was thus integrated in this study, in ad-
dition to academic studies from the literature.
The implications of misconceptions for the results of an inquiry may
be severe. It is therefore suggested that the awareness of the officers to
adverse cognitive effects should be increased by dedicated training and
in peer group discussions.
Additionally, interdependencies and the background of activities, as
well as the influence of the organizational framework, should be ana-
lyzed by an independent board in order to examine eventual flaws in
criminal investigations. Such assessments should target the develop-
ment and further refinement of a system that—while recognizing that
a failure rate of zero is unattainable—absorbs mistakes and minimizes
the adverse effects of errors.
Acknowledgments
The author is greatly indebted to the officers that helped in this
study and feels obliged to respect their request to remain anonymous.
References
[1] P. Watzlawick, The situation is hopeless, but not serious: (the pursuit of unhappi-
ness), Norton, New York, 1983, ISBN 978-0-393-31021-4.
[2] S.R. Gross, M. Schaffner, Exonerations in the United States, 1989–2012, Report by
the National Registry of Exonerations, University of Michigan Public Law Working
Paper 277, 2012, pp. 1–103.
[3] I.E. Dror, Combating bias: the next step in fighting cognitive and psychological con-
tamination, J. Forensic Sci. 57 (2012) 267–277.
[4] I. Stringer, Cognitive Distortions—A Practitioner's Portfolio, Forensic Psychology
Practice Ltd, Boldmere, 1999. 1–27.
[5] C. Tavris, E. Aronson, in: Mistakes were made (but not by me) (Ed.), Law and Disor-
der, Harvest Harcourt, Orlando, ISBN: 978-0-15-101098-1, 2008, pp. 127–157.
[6] K. Popper, Objective Knowledge: An Evolutionary Approach, Clarendon Press, Ox-
ford, 1972, ISBN 0-19-875024-2.
[7] I. Dror (Ed.)Cognitive Factors in Forensic Science, Science and Justice—Topical Col-
lection Issue, 2014 (http://scienceandjusticejournal.com/content/scijus-forensic –
acc. 27. 3. 2014).
[8] R.L. Helmreich, Error management as organisational strategy, Proceedings of the
IATA Human Factors Seminar, Bangkok, 1998, pp. 1–7.
[9] C. van Dyck, M. Frese, M. Baer, S. Sonntag, Organizational error management culture
and its impact on performance: a two-study replication, J. Appl. Psychol. 90 (2005)
1228–1240.
[10] EA-5/03 Guidance for the implementation of ISO/IEC 17020 in the field of crime
scene investigation, http://www.european-accreditation.org/n1/doc/EA-1-01.
pdf2008 (acc. 12. 11. 2013).
[11] H. Ditrich, Distribution of gunshot residues—the influence of weapon type, Forensic
Sci. Int. 220 (2012) 85–90.
[12] H. Ditrich, Scientific background of criminal investigations—does “forensic science”
exist? Forensic Perspectives, SIAK Scientific Series, vol. 6, 2011, pp. 13–40.
[13] S.F. Kelty, R. Julian, J. Robertson, Professionalism in crime scene examination: the
seven key attributes of top crime scene examiners, Forensic Sci. Policy Manage. 2
(2011) 175–186.
[14] R.L. Helmreich, On error management: lessons from aviation, Br. Med. J. 320 (2000)
781–785.
[15] D. Maurino, Threat and Error Management (TEM), Proceedings of the Canadian Avi-
ation Safety Seminar (CASS) Vancouver, Threat and Error Management (TEM) in Air
Traffic Control, ICAO 2005, 2005 (http://www.skybrary.aero/bookshelf/books/515.
pdf - acc. 6. 5. 2013).
Please cite this article as: H. Ditrich, Cognitive fallacies and criminal investigations, Sci. Justice (2014), http://dx.doi.org/10.1016/
j.scijus.2014.12.007
5
[16] US Department of Defense, The Human Factors Analysis and Classification System
HFACS 2005, http://www.uscg.mil/safety/docs/ergo_hfacs/hfacs.pdf (acc. 17. 10.
2013).
[17] I.E. Dror, Cognitive science serving security: assuring useable and efficient biometric
and technological solutions, Aviat. Secur. Int. 12 (2006) (2006) 21–28.
[18] P. Croskerry, The importance of cognitive errors in diagnosis and strategies to min-
imize them, Acad. Med. 78 (2003) 775–780.
[19] J. Zhang, V.L. Patel, T.R. Johnson, E.H. Shortliffe, A cognitive taxonomy of medical er-
rors, J. Biomed. Inform. 37 (2004) 193–204.
[20] J.W. Hurst, Cognitive errors (Can they be prevented?), Am. J. Cardiol. 101 (2008)
1513–1517.
[21] I. Patel, R. Balkrishnan, Medication error management around the globe: an over-
view, Ind. J. Pharm. Sci. 72 (2010) 539–545.
[22] I.E. Dror, A novel approach to minimize error in the medical domain: cognitive neu-
roscientific insights into training, Med. Teach. 33 (2011) 34–38.
[23] http://www.usroads.com/journals/rmej/9801/rm980102.htm (acc. 18. 10. 2013).
[24] G. Lovegrove, T. Sayed, Macro-level collision models for evaluating neighborhood
traffic safety, Can. J. Civ. Eng. 33 (2006) 609–621.
[25] I. Helsloot, J. Groenendaal, Naturalistic decision making in forensic science: toward a
better understanding of decision making by forensic team leaders, J. Forensic Sci. 56
(2011) 890–897.
[26] I. Dror, The ambition to be scientific: human expert performance and objectivity, Sci.
Justice 53 (2013) 81–82.
[27] R. Julian, S. Kelty, J. Robertson, “Get it right the first time”: critical issues at the crime
scene, Curr. Issues Crim. Justice 24 (2012) 25–37.
[28] I.E. Dror, D. Charlton, Why experts make errors, J. Forensic Identif. 56 (2006)
600–616.
[29] T.A. Busey, I.E. Dror, Special abilities and vulnerabilities in forensic expertise, The
Fingerprint Sourcebook, US Dept. of Justice, Washington, 2012, pp. 15/1–15/23
(Chapter 15).
[30] J.R. Vanderkolk, Examination Process, The Fingerprint Sourcebook, US Dept. of Jus-
tice, Washington, 2012, pp. 9/1–9/26 (Chapter 9).
[31] P.A.F. Frazer-Mackenzie, I.E. Dror, K. Wertheim, Cognitive and contextual influences
in determination of latent fingerprint suitability for identification judgments, Sci.
Justice 53 (2013) 144–153.
[32] G.L. Wells, M. Small, S. Penrod, R.S. Malpass, S.M. Fulero, C.A.E. Brimacombe, Eyewit-
ness identification procedures: recommendations for lineups and photospreads,
Law Hum. Behav. 22 (1998) 603–647.
[33] T.A. Busey, G.R. Loftus, Cognitive science and the law, Trends Cogn. Sci. 11 (2007)
111–117.
[34] A.C. Krix, M. Sauerland, Im Zweifel für den Angeklagten: Über Fehlerquellen von
Zeugenaussagen, Inquisitive Mind 1 (2011) (2011) 12–21.
[35] D. Canter, L. Hammond, D. Youngs, Cognitive bias in line-up identifications: the im-
pact of administrator knowledge, Sci. Justice 53 (2013) 83–88.
[36] S.D. Charman, The forensic confirmation bias: a problem of evidence integration, not
just evidence evaluation, J. Appl. Res. Mem. Cognit. 2 (2013) 56–58.
[37] S.M. Kassin, I.E. Dror, J. Kukucka, The forensic confirmation bias: problems, perspec-
tives and proposed solutions, J. Appl. Res. Mem. Cognit. 2 (2013) 42–52.
[38] M.J. Saks, D.M. Risinger, R. Rosenthal, W.C. Thompson, Context effects in forensic sci-
ence: a review and application of the science of science to crime laboratory practice
in the United States, Sci. Justice 43 (2003) 77–90.
[39] M. Saladin, Z. Saper, L. Breen, Perceived attractiveness and attributions of criminal-
ity: what is beautiful is not criminal, Can. J. Criminol. 30 (1988) 251–259.
[40] H. Sigall, N. Ostrove, Beautiful but dangerous: effects of offender attractiveness and
nature of the crime on juridic judgement, J. Pers. Soc. Psychol. 31 (1975) 410–414.
[41] A. Biedermann, P. Garbolino, F. Taroni, The subjectivist interpretation of probability
and the problem of individualization in forensic science, Sci. Justice 53 (2013)
192–200.
[42] J. Kruger, D. Dunning, Unskilled and unaware of it: how difficulties in recognizing
one's own incompetence lead to inflated self-assessments, J. Pers. Soc. Psychol. 77
(1999) 1121–1134.
[43] Department of Defense Human Factors Analysis and Classification System (DOD
HFACS).