United States courts serve an important "gatekeeping" function with respect to the admission into evidence of expert testimony and any accompanying demonstrative evidence. Given the complexity of many aviation accident and product liability matters, expert testimony is often vital to the presentation of a defense to the jury. With advancements in technology, experts are increasingly relying upon computer-generated demonstratives, including computer animations and virtual reality depictions, to present more accurate accident reconstructions and/or flight simulations, but the advanced technology can present a challenge in ensuring that the demonstrative and expert testimony will be admissible at trial.
Admissibility of Expert Testimony Generally – In carrying out their "gatekeeping" functions, the United States Supreme Court has held that courts must assess proposed expert testimony in order to ensure that it "rests on a reliable foundation." Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 597, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). In this regard, courts must find that: (1) the witness is qualified to be an expert; (2) the opinion is based upon "reliable data and methodology;" and (3) the expert's testimony on a particular issue will "assist the trier of fact." Nimely v. City of New York, 414 F.3d 381, 396–97 (2d Cir. 2005).
Additionally, courts must consider the following factors, which bear on the determination of reliability: (1) whether a theory or technique has been or can be tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the technique's "known or potential rate of error… and the existence and maintenance of standards controlling the technique's operation;" and (4) whether a particular technique or theory has gained general acceptance in the relevant scientific community. Daubert, 509 U.S. at 593–94. The Supreme Court has noted that its Daubert inquiry is a "flexible one" and does not "constitute a definitive checklist or test." Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 150, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Thus, the factors are not necessarily exclusive and not all factors may be applicable to all experts or in every scenario.
Admissibility of Computer-Generated Demonstratives – In keeping with the flexibility of the Daubert standard, courts have generally refrained from creating rigid standards for the admissibility of computer-generated demonstratives. Rather, courts have preferred to leave the determination of admissibility of evidence to the discretion of the trial judge, who is in the best position to consider and analyze the offered evidence. With respect to any evidence, the primary considerations are whether the evidence is relevant, and whether its probative value to jurors outweighs any potential prejudicial effect.
However, each individual item of evidence must be authenticated. Thus, in order for an expert to present or rely upon a demonstrative that includes virtual reality or computer animation, the evidence must be supported by testimony which establishes that the matter in question is what the proponent claims it to be. Federal Rule of Evidence 901(a). In determining whether virtual reality and/or computer animations are reliable, courts have generally found that the foundation must be laid for the technology itself, as well as for the expert methodology that underlies the technology. In practice, this generally results in the presentation of testimony from both the expert who prepared the data underlying an animation, as well as the computer technician or engineer who used that data to create the animation. See Clark v. Cantrell, 339 S.C. 369, 386, 529 S.E.2d 528, 536 (2000). Thus, courts may consider the Daubert factors of whether the evidence has been tested, has been subjected to peer review, has a known or potential rate of error and has gained acceptance in the relevant field, with respect not only to the methodology of the expert but also with respect to the software or program that created the demonstrative. See Valente v. Textron, Inc., 931 F. Supp. 2d 409, 420 (E.D.N.Y. 2013), aff'd, 559 Fed. Appx. 11 (2d Cir. 2014).
As with other any other expert opinion, computer-generated recreations which depict disputed evidence "must be based on scientific, identifiable, and objective facts." Cox v. State, 849 So.2d 1257, 1273 (Miss. 2003). If courts perceive that an expert's opinion is based on any "speculation" or "conjecture," the expert's opinion, and any demonstrative associated with that opinion, will be inadmissible at trial or on a motion for summary judgment. See Cal. ex rel. Harris v. Safeway, Inc., 651 F.3d 1118, 1149 n. 4 (9th Cir. 2011). Any animation that is not based on accurate data or "actual, physical measurements" from the scene of an incident may be deemed merely speculative, see Parvin v. State, 113 So. 3d 1243, 1251 (Miss. 2013), and recreations which do not exactly mirror the conditions which existed at the time of an accident may not be admissible. See Chase v. General Motors Corp., 856 F.2d 17, 19-20 (4th Cir.1988).
Thus, to ensure admissibility, an expert preparing any form of computer-generated accident reconstruction or simulation must ensure not only that the technology and methodology used has been tested and is generally accepted, but that every detail of the reconstruction reflects the exact conditions of the accident. Where an expert recreated an automobile accident via a live-action videotape depicting the route driven on the night of the accident, and then created a virtual reality depiction which made use of the videotape in addition to computer animation and simulation, the evidence was deemed inadmissible not by reason of the technology, but for deficient data input. Harris v. State, 152 S.W.3d 786, 793–94 (Tex. App. 2004). The party presenting the evidence conceded that the speed of the vehicle in the videotape and the perspective of the camera were not accurately depicted, as the camera had been located in the center of the backseat of the vehicle, rather than in the driver's seat. Accordingly, the trial court held that the inaccuracies represented in the videotape would be misleading and confusing to the jury and, as the virtual reality depiction relied upon the videotape, both pieces of evidence were inadmissible. The trial court's holding was upheld on appeal.
In a more recent case, a computer simulation model was held inadmissible on the basis of the technology as well as the data input. In Valente v. Textron, Inc., 931 F. Supp. 2d 409, 416 (E.D.N.Y. 2013), aff'd, 559 Fed. Appx. 11 (2d Cir. 2014), an expert created a computer simulation in order to determine what conditions would produce the type of instability necessary to have caused a golf cart to roll over. The software program that the expert used is one in which the user creates formulas and equations, which the program then solves. The expert, an accident reconstructionist, wrote his own computer code for the model, including all relevant algorithms and formulas. The code which constituted his simulation model had never been used or validated by anyone outside of the expert's company and the model was proprietary to the company, meaning that it was not commercially available. With respect to the technology, the court found that the simulation model did not meet the reliability requirement because it had not been validated, did not have a known error rate, had not been subjected to peer review and did not have general acceptance in the scientific community.
Further, the court found that certain components of data that the expert entered into the model were unreliable. The expert's formulas relied upon a tire model that had previously been used for automobiles, not golf carts, in addition to other models which had been created and verified for automobiles. The expert testified that the particular models used were irrelevant to the ultimate opinion reached regarding instability of the vehicle. However, the expert conceded that at least one of his automobile-based assumptions had to be corrected after he discovered that a portion of the calculation was different for golf carts. The court stated that it could not know whether the expert's remaining automobile-based equations were reliable in the context of a golf cart rollover simulation and the computer simulation was, thus, found inadmissible. As the expert's ultimate opinion regarding the instability of the vehicle relied upon the simulation, the expert's testimony was excluded altogether.
Thus, experts intending to rely upon computer-generated demonstratives must ensure that the computer software and/or programs creating the demonstrative meet the general Daubert standards of having been tested, having been subjected to peer review, having known error rates and having been generally accepted in the scientific community. In addition, the experts must ensure that their own underlying methodology separately meets these Daubert standards and that each individual piece of data entered into the software and/or program is reliable and accurately reflects the exact conditions of the accident. Given the sheer volume of data variables which would necessarily comprise a flight simulation and/or aviation accident recreation, especially in a virtual reality depiction, there may be many opportunities for an opposing party to attack the reliability of such a demonstrative, which makes thoughtful adherence to the Daubert standards at every layer of input and conclusion an important aspect of the creation of the demonstrative.