Expert witnesses
Human factors expert witness: how do courts test opinions on use errors and obvious risks?
A human factors expert explains how people perceive, understand and act with a product, and whether its design and warnings fit what users actually do. Courts test that opinion like any other expert opinion. In 2003 the Third Circuit upheld the exclusion of a human factors theory that had no literature or tests behind it, while the same expert was allowed to tell the jury how a child under stress would use a jet ski throttle.

What does a human factors expert witness do in a product case?
The expert explains how users perceive, understand and act with the product, and whether its design, controls, labels and instructions fit those users. The opinion goes to defect, to the adequacy of warnings and to causation.
In Hickerson v. Yamaha Motor Corp., U.S.A., decided on February 20, 2018, the Fourth Circuit recounted the district court's finding that a claim of inadequate warnings on a personal watercraft "implicate[s] the study of human factors and other industry standards" and lies generally beyond the common knowledge of a jury.
A human factors opinion answers a short list of questions. What could the user see, hear and feel at the moment of use? What did the user know about the risk, and from where: a label, a manual, training or general knowledge? What were users of this product likely to do, including under time pressure or stress? Would a different design, control or warning have changed that?
The people who give these opinions come from different fields. In the cases discussed below, the human factors expert in a jet ski case held a doctorate in experimental psychology and taught psychology, the expert who proposed new warnings in another watercraft case was a mechanical engineer, and the warnings expert in a third case was described by the court as having experience with warnings but not as a design engineer. For medical devices the vocabulary comes from the FDA's human factors guidance.
Which parts of a defect statute turn on what users know and do?
In Ohio, the design defect test weighs foreseeable uses and alterations, the likely awareness of product users of the risks, and what a reasonably prudent consumer would expect. The warnings statute excludes open and obvious risks.
Under Ohio Revised Code 2307.75(A), a product is defective in design or formulation if its foreseeable risks exceed its benefits when it left the manufacturer's control. Section 2307.75(B) lists the factors for the risk side. Factor (B)(1) is the nature and magnitude of the risks "in light of the intended and reasonably foreseeable uses, modifications, or alterations of the product." Factor (B)(2) is "The likely awareness of product users, whether based on warnings, general knowledge, or otherwise, of those risks of harm." Factor (B)(5) asks how far the design is more dangerous than a reasonably prudent consumer would expect when the product is used in an intended or reasonably foreseeable manner.
Three of these factors are about people. A human factors expert addresses which uses and alterations were foreseeable, what users knew of the risk and from which source, and what an ordinary user expected. The design engineer addresses the mechanism and the alternatives.
Two provisions favor the manufacturer. Section 2307.75(E) says a product is not defective in design if the harm was caused by an inherent characteristic that is a generic aspect of the product, cannot be eliminated without substantially compromising its usefulness or desirability, and "is recognized by the ordinary person with the ordinary knowledge common to the community." Under section 2307.76(B), a product is not defective for lack of a warning about a risk that is open and obvious or a matter of common knowledge.
In both provisions the test is what an ordinary person perceives and knows. A defense human factors expert describes what was visible and generally known about the risk. A plaintiff's expert describes what this user could perceive in the conditions of the accident: lighting, noise, the position of the body, the time available.
How does the FDA define a use error for medical devices?
As a user action or lack of action that differed from what the manufacturer expected and caused a result the user did not expect, was not caused solely by device failure, and did or could result in harm.
The FDA guidance Applying Human Factors and Usability Engineering to Medical Devices was first issued on February 3, 2016. The version published on the HHS guidance portal carries the issue date August 3, 2026. Section 3 defines a use error as "User action or lack of action that was different from that expected by the manufacturer" and that caused a result different from what the user expected, not caused solely by device failure, which did or could result in harm.
The same section defines abnormal use as a "Conscious, deliberate act or deliberate omission of an act that is counter to or violates normal use" and that lies beyond any further reasonable user interface-related risk control by the manufacturer. A critical task is "A user task which, if performed incorrectly or not performed at all, would or could cause serious harm," and the guidance counts compromised medical care as harm.
The two definitions divide the user's conduct into two groups. Use errors are what the manufacturer's human factors process is meant to find and control. A deliberate violation of normal use falls outside that process. In a device case, which of the two the user's action was is often the first question put to a human factors expert.
Section 4.1 explains where use-related hazards come from: aspects of the user interface that cause the user to fail to perceive, read, interpret, understand or act on information from the device. The list that follows includes device use that "necessitates physical, perceptual, or cognitive abilities that exceed the abilities of the user" and device use that "is inconsistent with the user's expectations or intuition about device operation."
What does the FDA guidance say about labeling, training and validation testing?
It ranks design first, protective measures second and information for safety, including training, last. Validation testing should use representative users in realistic conditions, generally with at least 15 participants per distinct user group.
Section 7 cites ANSI/AAMI/ISO 14971, which lists risk control options "in order of preference and effectiveness": inherent safety by design, then protective measures in the device itself or in the manufacturing process, then information for safety, which includes training users to avoid the use error. The guidance places design changes to the device and its user interface ahead of labeling and training, because labeling and training "rely on the user to remember" information. Labeling may not be at hand during use, and knowledge from training can decay. If a design change cannot remove a use error completely, the guidance still calls for updating the labeling and training.
Section 8 sets two conditions for human factors validation testing: the participants represent the intended (actual) users of the device, and the test conditions are sufficiently realistic to represent actual conditions of use. Section 8.1.1 says that "in general, the minimum number of participants should be 15." If the device has more than one distinct population of users, each population needs at least 15 participants. The FDA treats populations as distinct when their characteristics would likely affect their interactions with the device.
Section 6.2 names sources of known use-related problems: the FDA's MAUDE database of device reports, the CDRH list of medical device recalls and early alerts, and the manufacturer's customer complaint files. The guidance says that "All known use errors and use-related problems should be considered in the risk analysis" for a new device.
For a human factors expert in a device case, these sections define the records to request: the use-related risk analysis, which the guidance defines as the systematic use of available information to identify use-related hazards and estimate use-related risk, the formative evaluations, the validation test report with the participant groups, and the complaint files and MAUDE reports for the device. The expert compares the incident with them: whether the user's action was a known use error, whether the task was treated as critical, and how validation participants performed it.
Which human factors opinions did the Third Circuit admit and exclude in Calhoun v. Yamaha?
The expert could testify that a child under stress would squeeze a throttle shaped like a bicycle brake. His theory that stress clenching activated the throttle was excluded because no literature or tests supported it.
In June 1989, 12-year-old Natalie Calhoun rode a rented Yamaha Wavejammer WJ500G at the Palmas del Mar resort in Puerto Rico, made a sudden turn and crashed into an anchored boat. She died of head and neck injuries. A warning label listed 14 as the minimum recommended age. The plaintiffs challenged the "squeeze finger throttle" design and the warnings.
Their human factors expert, Dr. Edward W. Karnes, held a doctorate in experimental psychology and was an emeritus professor of psychology at Metropolitan State College in Denver. Judge Louis H. Pollak of the Eastern District of Pennsylvania let him testify that because of the throttle's similarity to a bicycle brake, "a child in a stress situation would naturally squeeze the mechanism in order to stop the jet ski," and that the warning made the vehicle unreasonably dangerous and defective, "especially for youthful operators."
Three parts of his testimony were excluded. The first was the theory that under stress a person tends to clench the hands and inadvertently activate the squeeze finger throttle: "There was no literature confirming this theory, nor demonstrable tests." The second was his own conclusion that the throttle design was defective. The third was the opinion that the minimum age should be 16, because he had nothing to say to support "that number rather than a number higher or lower."
In Calhoun v. Yamaha Motor Corp., U.S.A., Nos. 02-4098 and 02-4162, decided on November 18, 2003, Chief Judge Scirica wrote for the Third Circuit that, lacking support, the clenching testimony "was speculative and unreliable," and affirmed. The court also upheld limits on a marine engineer who, at the time of his report, "had never operated a jet ski" and had tested no alternative throttle. The jury found for Yamaha on the strict liability claims.
The admitted opinion tied a user's response to a visible feature of the product, the bicycle-brake shape. The excluded opinions claimed a physiological mechanism and a specific age with no study behind either.
Must a human factors expert have used the product?
Not always. In Smith v. Ingersoll-Rand the Tenth Circuit held that firsthand knowledge is not a condition of admissibility, and most of the expert's testimony there concerned how the manufacturer developed and marketed the machine.
On April 5, 1993, in Las Cruces, New Mexico, groundsman Ron Smith was using a sledgehammer to free a jammed sensor ski while an Ingersoll-Rand MW-6520 milling machine backed up. The operator, apparently unaware of him, turned the wheels, and Smith's right foot was trapped under the front tire. His leg was amputated above the knee.
The plaintiffs' expert in human factors engineering, Dr. Edward Karnes, testified that Ingersoll-Rand "failed to conduct an adequate human factors analysis of the milling machine before marketing it," that poor visibility and noise hindered communication among workers, and that the machine should have had mirrors so the operator could see the groundsmen. He called the company's argument that mirrors create a false sense of security a rationalization that human factors literature had tested and disproven.
Ingersoll-Rand argued that neither Karnes nor the safety expert Vincent Gallagher had firsthand experience with milling machines. The district court ruled that this went "to the weight and not the admissibility of the testimony." In Smith v. Ingersoll-Rand Co., No. 98-2340, decided in 2000, the Tenth Circuit found no abuse of discretion and wrote that "firsthand knowledge is not requisite to the admissibility of an expert opinion." Most of the testimony concerned the company's development and marketing procedures, where firsthand observation of the machine would shed little light. The court quoted Kumho Tire: the expert must employ "the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." It affirmed the judgment in its entirety.
In Calhoun the lack of experience counted against a design opinion. The marine engineer there had never operated a jet ski when he wrote his report and had run no tests of alternative throttle designs, though he admitted he could have. The district court let him describe the squeeze finger throttle but barred his opinion that it was unsafe because it resembled a bicycle brake, and barred his testimony on warnings. The Third Circuit upheld those limits.
Can a human factors expert testify that users ignore all warnings?
Not without data. In Wells v. Kawasaki the Tenth Circuit upheld the exclusion of an opinion that the way people use a personal watercraft makes warnings of any kind ineffective, which rested on a few depositions and one article.
In summer 2015, 16-year-old Nicole Rider, now Wells, was a passenger on a rented 2013 Kawasaki personal watercraft at Lake Powell. She fell backward during acceleration, and the water jet caused internal injuries, including a ten-centimeter tear in her large intestine. She needed a colostomy. The label on the back of the craft said "WEAR PROTECTIVE CLOTHING," and the owner's manual said that "[n]ormal swim wear will not adequately protect you." She wore an ordinary bathing suit. A luxury seat with more hip support was offered on other 2013 models.
Her warnings expert, Joellen Gill, took the view that users would not read the label and that "the manner in which people typically use and experience the [PWC] render warnings of any kind ineffective," so Kawasaki should have looked for a design solution. The District of Utah excluded her opinions and those of the engineer Dr. Anand Kasbekar, whose self-designed seat drag test had no peer review and no known error rate. Without the experts, the design and warning claims failed on summary judgment.
In Wells v. Kawasaki Motors Corp., U.S.A., No. 20-4004, an unpublished order and judgment of March 29, 2021, Judge Phillips wrote for the Tenth Circuit that Gill's primary opinion assumed a design alternative existed when "no data supports" one, and that her other opinions rested on "a few deponents' testimony and one article" and on general studies. The court cited Joiner for the rule that a court need not admit an opinion connected to the data only by the ipse dixit of the expert, and affirmed.
The claim that users ignore warnings is itself a claim about behavior. In Wells it needed research or data about how people use that kind of product, and the expert had none.
What happens if the user never read the warning?
The case then turns on the warning that was there. In Hickerson v. Yamaha the plaintiff admitted she did not read the warnings, and the Fourth Circuit held them adequate as a matter of law.
In June 2012, on Lake Hartwell, a passenger fell off the back of a 2011 Yamaha VXS WaveRunner into the jet thrust and suffered internal injuries. She was wearing a bikini. The label on the glove box began "WEAR PROTECTIVE CLOTHING" and said that "Severe internal injuries can occur if water is forced into body cavities." A label on the rear platform told passengers to "[w]ear a wetsuit bottom or clothing that provides equivalent protection."
Her expert, Dr. Anand Kasbekar, a mechanical engineer, proposed an alternative warning system. The district court excluded it because the proposals were neither tested nor based on specific relevant research or studies. The Fourth Circuit added that he "supported his inadequate warning opinion with no research, data, or scientific theories."
The court applied comment j to section 402A of the Restatement (Second) of Torts: "Where warning is given, the seller may reasonably assume that it will be read and heeded." Hickerson admitted she did not read the warnings. The court found that they "plainly warn of the danger of the orifice injuries Hickerson sustained," pointed to their placement in multiple locations on the front and rear of the craft and to their direct and specific wording, and held: "Thus, the PWC's warnings were adequate as a matter of law." The design claim failed under comment j as well.
The Fourth Circuit wrote that the district court had not excluded the warnings claims merely because they lacked expert testimony: it required Hickerson to support them with any admissible evidence. A human factors expert who says a warning failed starts with the actual labels, their place on the product and their wording, and then explains why this user did not or could not act on them.
What should a human factors expert's file contain?
The product or an exemplar, every warning and instruction as the user met them, the user's task and conditions, the literature or tests behind each claim about behavior, and for medical devices the manufacturer's human factors records.
The cases set out what a court looks for. In Calhoun the admitted opinion linked user behavior to a feature of the product, and the excluded opinions had no literature or tests. In Wells the opinion that warnings of any kind would fail rested on a few depositions and one article. In Hickerson the alternative warnings were untested and the warnings opinion cited no research or data. In Smith the expert's criticism of the manufacturer's process rested on depositions, discovery documents and the human factors literature on mirrors.
From these rulings the file starts with the product or an exemplar of the same model, and with every warning and instruction as the user met them: on the product, on the packaging, in the manual, in training materials. It describes the user's task step by step and the conditions of the accident, such as speed, lighting, noise and the time available to react. Each statement about how people behave cites a study, a test or a published source, and if the expert proposes a different warning or design, the file shows how it was tested or what research supports it.
For a medical device the file adds the records named in the FDA guidance: the use-related risk analysis, formative evaluations, the human factors validation report with its participant groups, complaint files and MAUDE reports. The expert states which version of the guidance applied when the device was developed, because the guidance was first issued in 2016 and has a 2026 version.
How does IAE find a human factors expert for a product case?
In the IAE request, choose the case type Product liability, name the product and the use or warning at issue. Within 5 business days IAE sends 2-3 candidates with CVs, hourly rates and prior testimony lists.
The product liability expert witness page lists human factors and warnings specialists who address what a user would notice on the label and in the manual, including medical device use errors under the 2026 FDA human factors guidance. The request also asks for the state and court, your side, the expert disclosure deadline, and the parties and their counsel for the conflict check. Please leave out privileged or sealed information.
Within 1 business day IAE replies whether it can search in that field by that deadline. Each candidate confirms in writing that he or she has no ties to the parties or their counsel, and the candidate file includes testimony in the last 4 years and publications in the last 10 years.
The search fee is paid in advance, and candidate names and CVs come after payment. If no candidate fits, the second search is free; if IAE finds no candidates, it refunds the fee. The expert's hourly rate never depends on the outcome of the case or on the content of testimony. Under the agency option the expert bills through IAE on one invoice.
IAE is not a law firm. Case strategy and the decision to retain an expert stay with the attorney.
Sources
- FDA, Applying Human Factors and Usability Engineering to Medical Devices, guidance issued August 3, 2026 (originally issued February 3, 2016), HHS guidance portal
- Ohio Revised Code 2307.75, Product defective in design or formulation
- Ohio Revised Code 2307.76, Product defective due to inadequate warning or instruction
- Calhoun v. Yamaha Motor Corp., U.S.A., Nos. 02-4098 and 02-4162 (3d Cir. Nov. 18, 2003)
- Smith v. Ingersoll-Rand Co., No. 98-2340 (10th Cir. 2000), slip opinion archive
- Wells v. Kawasaki Motors Corp., U.S.A., No. 20-4004 (10th Cir. Mar. 29, 2021)
- Hickerson v. Yamaha Motor Corp., U.S.A., No. 17-1075 (4th Cir. Feb. 20, 2018), in the appendix to the petition for certiorari, No. 17-1589
Read also
All articles
State of the art defense in product liability: what must the manufacturer's expert prove about the technology available when the product was made?
The state of the art defense asks what a manufacturer could have known and built on the day the product left its hands, not on the day of the injury
Toy injury expert witness: small parts, age grading and ASTM F963 in a children's product case
A toy injury case usually turns on a short list of questions: was the toy meant for a child under three, did any part fit inside the federal small
Machine guarding expert witness: what does the engineer examine when a worker is hurt by an unguarded machine?
When a worker is hurt at the point of operation of a machine, the case usually turns on the guard: how the machine was designed to keep hands outExpert witnesses
Need an expert witness for your case?
Describe the question, the case type and the court. Within 5 business days you receive 2-3 candidates with CVs, hourly rates and lists of prior testimony.

