[Driver Attitude] [Older Drivers] [Licensing Older Drivers]
[Ergonomic Intro] [HF & Litigation] [HF Consulting]
Best viewed with
Netscape 3.0!
Last updated 17 July 1997
Erik C. B. Olsen, M.S.
Human factors expert testimony plays an important role in litigation and forensics. This panel presents key issues of human factors litigation of interest to the public, academic, and professional sectors, as well as to present and future lawyers, business people, and consultants.
INTRODUCTION
A wide variety of human factors expert testimony plays an important role in
litigation and forensics. Furthermore, human factors and ergonomics forensic
consultants and professionals prepare their testimony based on a scientific basis and
may be called upon by attorneys for limited, specific opinion or, as is more likely the
case, to assist the attorney (and the court) in fully understanding the issue at hand in
a broad sense. Although often expensive and seemingly inefficient, the courts
provide a "powerful tool to correct the flaws that irresponsible engineering,
management practices, amateur entrepreneurs, or other agencies foist on the using
population" (Hornick, 1994).
Forensics experts provide valuable contributions within the justice system and to the society at large. This panel presents perspectives from human factors professionals based on their own experience for better understanding the process involved in presenting clear and useful testimony.
Reference
Hornick, R. J. (1994). Forensics experts fire back (letter; comment). Ergonomics in design, July, 4-5.
WHEN TO WARN: LEGAL VERSUS HUMAN FACTORS PERSPECTIVES
Donald P. Horst
One of the frequently litigated areas in which human factors experts can contribute
is "failure to warn." However, the scientist and the lawyer view the relevant issues
from very different perspectives. These differences can create problems for the
human factors professional. Some find the mixture of law and science so awkward
that they are unwilling to do legal consulting at all. Those who do work in
litigation may find that they have limited credibility in the eyes of their audience –
the lawyers, judges, and jurors. In addition, the focus of the legal system on failure
to warn may have impacts on the design and use of warnings that the human
factors professional may consider undesirable from a safety standpoint.
The contrast between the legal and scientific perspectives is familiar to anyone who
has worked in this area. Some differences are listed below, with comments on
problems they create for the expert who addresses questions of when to warn.
Goals: the client's best interest versus scientific accuracy
The difference in goals and philosophy between law and science has probably been felt by every human factors professional who has worked in litigation. There may be no more awkward mixture than that of advocacy and impartiality. The human factors expert can fall back on the methodological guidelines of our profession, at least in part. However, our understanding of unintentional experimenter bias may leave us less comfortable than experts in other fields who are less sensitive to such issues. In the laboratory, we go as far as to construct double blind experimental designs to avoid bias, while the forensic consultant, by contrast, has full knowledge of the outcomes desired by the client and a vested interest in pleasing the client. The effects of bias, if any, on expert opinions and on warnings that are currently in use would be difficult to measure, but the solution seems clear – continuing research and initiation of open discussion outside the context of litigation.
Ultimate authority: law versus data
One of the most obvious differences between law and human factors lies in the ultimate authorities to which they appeal. For example, in some states, a "presumption of effectiveness" for warnings is written into the law. The human factors expert may wish to run experiments to examine effectiveness, but the law cannot be changed by facts. In any case, the lawyer may be reluctant to conduct experiments, because he or she does not know the results in advance and fears that the data might provide ammunition for the opposing side. The legal appeal to laws, regulations, and standards affects questions of both when and how to warn. Human factors experts are often asked to refer to warnings guidelines or rules of thumb in assessing whether warnings were adequate, even though both research and common experience have demonstrated that the guidelines are neither necessary nor sufficient to insure effectiveness of the warnings.
Subject matter: duty versus effects
Even the basic subject matter of the lawyer and the scientist may differ dramatically. The lawyer is concerned with responsibility and with the duties and rights the various parties may have under the law. The human factors expert may be familiar with the law and may have strong personal convictions about these matters, but they are not usually questions that can be addressed experimentally, even in principle. At the same time, the human factors expert may be able to provide information that is relevant to anyone who is addressing the legal questions. The information may take the form of experimental data on the effects of warnings, or survey data bearing on what the public wants or believes. A potential conflict arises when the client asks for an opinion on a subject that human factors does not address directly. How should the human factors expert respond and where should the lines be drawn?
Methodology and procedures
The legal framework defines an adversarial approach to resolving conflicts
and differs in principle from the scientific approach that is, ideally, a cooperative
approach to reaching consensus. Some examples of methodological conflicts are:
Selective versus comprehensive definition of the problem (e. g., One expert
may be asked, "Is it a good warning?" the other, "Did the warnings play any role in
the accident?").
Legally defined terms versus operationally defined terms.
A reasonable degree of scientific certainty (i. e., [p > .50] versus [p < .05]).
The "legal document" model of warnings versus the behavioral impact model
While the plaintiff's lawyer often argues that additional warnings were needed even when the plaintiff was well aware of the hazard, the defense often argues that warnings were provided even when they were limited to obscure statements in documents that the plaintiff never read. Neither argument may make sense to the human factors expert, but there may be no way to reconcile the human factors approach with the lawyers' theories, even when human factors issues are central to the case.
Summary
The differences between legal and human factors perspectives create two interrelated challenges for the human factors expert who addresses failure to warn. One is to convince a skeptical audience that we are scientists with something of value to say. The other is to develop guidelines for when (and how) to warn, based on our own, scientific methods.
RESPONDING TO ADMISSIBILITY CHALLENGES OF
ERGONOMICS/HUMAN FACTORS TESTIMONY
Dieter W. Jahns
Motions to exclude expert-witness testimony (aka "motions in limine" ) are a
common tactical strategy in civil cases before both federal and state courts. There is
nothing "personal" about them, and they are used against experts in all
professional/occupa-tional fields. My presentation will summarize on what basis
admissibility challenges are made and how to assist client attorneys to counter such
challenges for qualified ergonomists/human factors practitioners.
Ergonomic Systems Development (ESD) involves analysis, design, and testing
and evaluation to enhance human performance, workload, safety/health, and
economic value by means of technological innovations. When accidents happen,
(now often renamed "unintended injuries" by public health and safety advocates)
the ergonomist is often called upon to investigate and render an expert opinion as
to human-factors aspects of accident/injury causation (Jahns, 1990). What an expert
can do at trial is governed by a) rules of civil procedure, b) rules of evidence, and c)
discretion of the bench. It should be obvious that the expert is involved to educate
the trier-of-fact (judge and/or jury) with regard to scientific information which is
beyond "common sense" or personal experience in a generic sense. The expert must
act in an impartial manner and do nothing that would imply partisanship or
interest in the outcome of any case. That is, the expert is there to provide facts, not
to usurp the job of the jury which is to evaluate facts relative to legal theories of
fault or blame.
Challenges to expert testimony are usually based on a) lack of qualifications by
the expert, b) insufficient foundation to render an opinion, and/or c) relevancy of
the proposed testimony to the issue at bar. In general, the motion in limine will
claim that the proposed testimony will "unduly arouse the prejudice or sympathy of
the jury", or "create side issues", or will "consume undue amount of time to
introduce" or "be cumulative" or (especially for ergonomics/human factors)
"invade the province of the jury." While the forensic ergonomist must avoid any
behavior that may be construed as the practice of law, the ergonomist and lawyer
must integrate their expertise in drafting a memorandum in opposition to the
motion to exclude the ergonomist's testimony. The lawyer will focus on legal
precedents; the ergonomist will focus on the relevance of his/her education,
training, experience, license/certification as they relate to the investigative work
that was performed to render expert opinions. Judges are now "the gate keepers" for
determining "the line between science and superstition" (Huber, 1990, pg. 113).
Therefore, an accomplished forensic ergonomist must know legal theory and legal
constraints, know the rules of civil procedure, know the rules of evidence, be
prepared to practice in the roles of consultant, analyst, investigator/researcher,
report generator, and ethical representative of the profession before the court. An
honest, well-researched memorandum (e.g., Quesenberry, 1994) can accomplish a lot
in allowing expert testimony and in keeping ergonomics from being falsely accused
of being one of the "junk sciences" (Huber, 1990).
References
Huber, P. W. (1990), Pathological science in court. DAEDALUS (Journal of the American Academy of
Arts and Sciences), Risk, 119(4), 97-118.
Jahns, E. W. (1990), Forensic human factors: Caveats for the profession. Human Factors Society
Bulletin, 33 (4), 3-4.
Quesenberry, S., Esquire, (1994) Plaintiff's memorandum in opposition to defendant's motion to exclude
Dieter Jahns, Filed October 19, 1994, Superior Court Washington, County of San Juan, Cause #93-2
05157-4, Friday Harbor, WA, pg. 1-8.
USE OF COMPUTER PROGRAMS & GRAPHICS IN ACCIDENT RECONSTRUCTION
David A. Thompson
Introduction
Communicating a physical facility or action to a jury is necessary to describe
how or why a specific physical accident happened at that facility or was caused by
that action. Often, an expert witness may attempt this with a verbal explanation,
possibly using gestures or other body language. However, this approach requires the
juror, who generally knows nothing about the actual physical situation, to make a
mental transformation from the intellectual description of the witness to a three
dimensional one in a clear, reliable manner "on the fly" in order to comprehend
what actually existed and occurred. Unless this mental transformation is done in a
high fidelity manner, the juror will not have an accurate factual basis on which to
base his or her opinion, and justice may not be well served. Obviously, the juror
lives in a 3-D world and requires ideas and facts converted to his or her world to
fully comprehend.
This communication problem is avoided when direct representations of the
three dimensional physical world are used directly in the presentation of the facts of
a case, allowing a direct 3-D to 3-D mental transformation by the juror. Models of
stairways, power tools, and actual stepladders are very helpful in this regard. Even
2-D diagrams of ladders or stairways, floor plans, or photographs of machines from
several angles permit a relatively simple 2-D to 3-D transformation on the part of
the juror.
However, how does the human factors professional represent people in such
2-D or 3-D models or diagrams in a meaningful manner that permits reasonable
high fidelity accident reconstruction? Is it possible to have both static and dynamic
representations of physical situations and behaviors? Is it possible to represent 5th
percentile, slender females as well as 95th percentile, heavy males, and at selected
points in between? How can the humans being modeled be put in representative
postures related to the most probable accident scenarios?
I will present a variety of examples where human mannequins have been
used to represent various critical events in accident scenarios. The software used is
Manequin* supplied by the Biomechanics Corporation of America, and supposedly
developed for solving selected NASA personnel space problems. With some effort,
it is possible to posture a wire frame mannequin of a human in virtually any
posture that the human can reasonably be expected to assume. (Interestingly, since
the software mannequin will not assume impossible human postures, it can be used
to evaluate the feasibility of questionable human reaches, stretches, twisting, and
visual ranges. Movements being evaluated can include all degrees of freedom of
which a particular limb is capable.) Movement of the separate body parts, including
individual fingers, is possible. In addition, simple drawing tools are available to
show the relative relationships of desks, handrails, steps, and other physical
features.
Once the mannequin is postured in a particular configuration, it has a 3-D
capability and may be viewed from any direction including an isometric view. It is
even possible to apply forces to the hands or feet and measure the resulting torques
on shoulders and legs. One may also choreograph a crude animation of a brief
motion sequence, or display the "normal" cone of vision.
Simple extensions of the mannequin's actions are possible and are helpful to
the human factors professional. The body's center of mass can be drawn in to show
moments of tipping or bending force when climbing a vertical ladder or reaching
inside an automobile trunk. Other forces can also be shown, such as the amount
and direction of hand force on handrails, or traction forces during heel strike when
walking. Several mannequins of differing size and gender may be shown and
manipulated in the same scenario.
The potential user should be warned, however, that significant bugs remain
in standard copies of the software, and no upgrades to correct them are planned.
Although the software is very useful in selected applications, it can be very
frustrating to use in certain object drawing applications. While one can draw objects
in 3-D to accompany the mannequins (which are always in 3-D) this can be very
tedious and difficult to do without the inclusion of small dimensional errors.
THE HUMAN FACTORS EXPERT IN ROAD USER LITIGATION, OR CONFESSIONS OF A HIGHWAYMAN
Richard A. Olsen
About 400 of my 500 cases over the last 21 years as an expert witness or
consultant have been concerned with behaviors of drivers and pedestrians in
highway litigation. In this presentation I will discuss a few of the more important
issues that have concerned me in this practice. First, what we call ourselves has
implications for what we (I) do and for what juries and attorneys think we do. If the
word "engineering" is used in a title, does it help or hinder others in
understanding? How about the word "psychologist?" And do we help or do we
confuse when we testify about "research?" In raising the issue of complex analysis,
computer simulation, and statistical testing, do we run the danger of raising costs
without clarifying the issues?
Road users, the road environment, and the vehicle's controls and reactions
obviously have implication for engineering analysis. Without background in road
design, delineation, lighting, signs and markings, vehicle dynamics, and accident
reconstruction, the human factors specialist may be reduced to opinions on "human
behavior." There are cases that require analysis in areas of the clinical psychologist,
but despite the assumptions of many attorneys and judges, that is not what I,
personally, am there for.
Among the problems in the field now becoming known as "ergonomics" is
the question of defining and distinguishing the human factors specialist (HF) and
the human factors engineer (HFE). In itself, "ergonomist" is not any more
understandable than these HF and HFE labels, though it would encompass both. I
use "HFE" and "engineering psychologist" in my work, in spite of the latter term's
more academic connotations, because I bring engineering training and design
experience into the analysis of questions in litigation as well as the HF areas of
specialization. This has advantages, but it presents problems of its own. It is
essential that each expert witness describe clearly what his or her own qualifications
are when opinions are to be given.
In my case, the advantages of the HFE label, including the ability to offer
opinions as an accident reconstructionist, can be valuable for at least three reasons.
First, the driver is certainly influenced by the design of the entire transportation
system, and no analysis is complete without considering technical aspects. Second,
the assumption of the psychologist as counselor tends to undercut the quantification
and hard science aspects that HFE (or HF) brings to the analysis. Equations,
drawings, and tables are not presented for only that reason, but a quantitative
approach does bring the full message of the potential of ergonomics in design. The
judge, jury, and attorneys usually have respect for algebra, geometry, and statistics in
an opinion, especially if the material is presented in a way that allows them to
follow the arguments and conclusions, if not all the math. Third, the broadening of
investigative coverage often allows one expert to cover the necessary breadth in an
economical manner, though I regularly work with a traffic engineer or an accident
reconstructionist as well. And yes, cost control and excessive fees are another
concern of mine for the future of the expert.
There are also disadvantages to implying one is an engineer and a
psychologist. The two fields are somewhat complementary, but each is more than
enough in itself for a lifetime practice. There is a need to specialize within each
field, and when expertise is claimed in both, greater narrowing of knowledge in each
field is implied. Many expert witnesses have had to face the questions; "Oh, so you
are an expert in everything, is that right?" One answer is, "yes, everything having
to do with the driver's motivations and reactions in this situation." That may be
the objective of an HF analysis, but seldom can one say "everything" has been
explained. Such an answer would soon be put in that category of "things I wish I
hadn't said." There are limits to what one person can know, and the litigation
process recognizes that. Where the stakes are high, armies of experts will be paraded
through the court. Each adds to the picture the attorney is constructing, with
overlapping areas to lend emphasis to the extent the court will permit.
But many cases do not have "large" stakes. More often, the action is trying to
obtain the limits of one or more insurance policies, totally in the neighborhood of
$100,000. One or two experts cover what can be said for each side. An estimate of a
speed of impact from damage is appropriate from a HFE/reconstructionist in such
cases, rather than from a separate reconstructionist. The accuracy of estimates is
usually limited by the lack of details in the information available. It would not
make sense to pursue an elaborate computer analysis or reenactment when the
outcome depends on which (arbitrary) assumptions are made for the inputs. Even
in bigger cases, the back-of-an-envelope computation will give the same results as a
frame-by frame movie analysis if the underlying data are approximate. When you
show convincingly that you are working with input values of ±20%, a computer
printout to four decimal places fails to impress even the most computer-naive
juror.
My experience has led me to start with simplifying assumptions and to offer
basic opinions to the extent that the information allows. Because I have written and
reviewed papers for, and attended, a number of research organizations and their
committees for many years, I am able to stay up on current work fairly well. I
seldom do a literature search, as such, but rather pull out a specific paper or
reference when it applies to issues in a current case. The issues in much litigation
are not complex compared to those in a research setting. Making the issues sound
complex, or offering to do "scientific studies" that really do not have sufficient
controls to be reliable, adds to the cost without adding to the enlightenment of the
jury. The difference between an attempt to explain a concept in court and to snow a
jury can be one of degree and conscience. Computer simulation, in my opinion, is
especially prone to obfuscation and misuse. I used some of the early reconstruction
programs, ca. 1980, but found this rapidly became a specialty in itself. Also, it is
important to an ethical practice that the difference be made clear between an
observation of driver behavior at a site, for example, and a real analysis of such
behavior.
There is room for disagreement in the preceding discussion (for more, see Olsen, 1992). What holds for me and my experience does not necessarily hold for you and yours. An expert witness must consider such distinctions, however, both in preparing a presentation and in the critique of the analysis presented by someone else. That is the intent of this presentation.
Reference
Olsen, R.A. (1992) Human factors engineering experts in product design. Products Liability Law
Journal, 4 (1), 23-41. Combining human factors analysis and accident reconstruction: Illustrations from
transportation accident cases.
HUMAN FACTORS ISSUES IN ANALYSIS OF TRAFFIC ACCIDENTS
Kenneth Ziedman
Introduction
Expert testimony relating to human performance and behavior is now
commonplace in litigation with traffic accidents. Analysis of individual accidents is
challenging in that it forces the expert to apply general human factors knowledge to
a specific instance of human/vehicle/roadway interaction and to specify those
factors or conditions which are most related to an individual accident. This process
is generally conducted in the face of incomplete knowledge of the accident
circumstances, including limitations concerning knowledge of the perceptions and
responses of involved parties and witnesses. Further, the expert's analysis is
conducted and presented in the context of adversarial criminal or civil legal
proceedings, the goal of which is to place responsibility for an event. Ideally, the
expert's role is one of technical analysis to help explain human factors aspects of an
accident to assist the jurors in an ultimate decision of responsibility.
The focus of this report is on human factors data used or desired by experts in
formulating conclusions in litigation. The value of examining such data in the
context of forensic analysis is (a) analysis of actual accidents can enrich human
factors understanding of the interactions between operator, vehicle and roadway, (b)
a potentially useful database of human factors case histories and applications may be
buried in legal proceedings, (c) the identification of research needs may arise from
limitations experienced by practitioners in forensic analysis, and (d) examination of
how human factors data are used in litigation may help to improve the quality of
forensic analysis. In addition, the results of the study may be useful in human
factors education in providing a group of accident case histories and data
applications for a common area of human performance.
This project involved summarization of human factors data used in accident
analysis from persons involved as testifying experts. Contributors were asked to
provide case reports containing a brief summary of the accident, human factors
issues, data used in analysis and any additional data that would have been helpful
but was not available, as well as references to human factors data they found useful
in accident analysis.
A further purpose of this project was to compile a bibliography of data sources useful for analyzing vehicle and road user performance, with emphasis on sources of quantitative data that can be applied to forensic analysis.
Methodology
A data base of accident descriptions, human factors issues and data limitations
was developed based on the 59 case reports submitted by contributors. Accidents
were coded according to simple collision-type categories (e.g., vehicle/vehicle,
vehicle/pedestrian, etc.). Human factors issue and data limitation categories were
defined based primarily on functional performance
areas (e.g., perception/decision/reaction time, visibility and conspicuity, etc.).
Summaries of each case report were prepared and entered into the data base in a
consistent format. The accident scenarios, human factors issues and
data limitations associated with each case report were reviewed, coded and
tabulated. Finally, the results were integrated and summarized.
Results and Discussion
The case reports resulted in 201 issue items and 115 data limitation items.
The most frequently noted human factors issues were visibility and conspiciuty,
decision and cognition, user characteristics and perception/reaction. The most
frequently noted data limitation areas were visibility and conspicuity,
perception/reaction, workload, and decision and cognition. Specific topics in
which additional data would be especially useful in understanding accident
causation were: (a) user performance in complex perception, decision and response
situations, (b) driver visual search and attention, and (c) driver interaction with
roadways and signing, including effects of expectation and familiarity. Examples of
the types of data that contributors felt would be useful for accident analysis will be
presented.
The study results are limitated in that traffic accidents that are at issue in
litigation are probably not a random sample of all traffic accidents and in that the
total number of case reports were fairly small. However, many of the data limitation
topics do suggest research areas that could provide useful information for traffic
safety decisions. Further, it is likely that failures in user/system interaction that lead
to litigation are similar to the failures in accidents that do not result in litigation.
A final issue to be discussed is the manner in which human factors data are
usedin forensic analysis and the application of controlled research results to specific
accidents.
Reference
Ziedman, K. (1995, 9 January). Human factors issues in forensic analysis of traffic accidents (draft) for Committee on Vehicle User Characteristics, A3BO2, Transportation Research Board.
This paper is referenced as:
Olsen, E. C. B., Ellsworth, O. T., Horst, D. P., Jahns, D. W., Thompson, D. A., Olsen, R. A, and Ziedman
K. (1995). Human Factors in Litigation (panel). In Proceedings of the Silicon Valley Ergonomic
Institute's 1st Annual Conference (ErgoCon '95), San Jose, California.
[Driver Attitude] [Older Drivers] [Licensing Older Drivers]
[Ergonomic Intro] [HF & Litigation] [HF Consulting]