Admitting computer animations: more caution and new approach are needed.

• Author: Chatterjee, I. Neel

COMPUTER-generated evidence has been admitted as both substantive and demonstrative evidence. Computer simulations are admitted under high standards because they draw new conclusions for a jury. Computer animations, on the other hand, are admitted under lower demonstrative evidence standards because courts analogize them to charts or diagrams. This legal distinction is premised on numerous misunderstandings concerning the nature of human perception of computer animations.

Similar standards should be employed for both simulations and animations. Based on various theories of admissibility, this article develops a three-step process for admitting computer animations as substantive evidence.

COMPUTER ANIMATION

Computer animation graphically depicts the testimony of a witness or witnesses.(1) In its simplest form, it is a picture of a static object. For instance, the animation could be a simple picture of a baseball. Theoretically, the animation makes no conclusions, but as it becomes increasingly complicated, assumptions and theories begin to take on a more simulated character. Thus, a standard is needed to admit "simulated animations."

In a simulation, a computer draws conclusions based on limited facts. An example of a simulation is a calculation of the flow of toxic waste through the soil given only the consistency of the soil and the amount of toxic waste put in the ground. A simulation may or may not have a graphic depiction.

This article uses an example of an animated event to demonstrate the problems that arise when animations or simulations are used in trials. The event animated is a player hitting a home run during a baseball game but also possibly hitting a victim with the baseball. The animation is used to demonstrate the testimony of a witness who claims to have observed the event. An issue in the case is how hard the victim was hit, if at all.

The issues raised are: * possible theories of admissibility for the computer-generated evidence; * problems of reliability encountered in animating events; * most appropriate method of admitting animation, including laying a proper foundation for the computer programs used to animate the event, the underlying scientific theories and authentication of the animation if a witness is available; and * concluding that a higher standard for admitting computer animation is necessary to ensure fairness.

By subjecting animation to a more rigorous test, the proposed method reduces the likelihood a jury will be misled by unreliable evidence.

THEORIES OF ADMISSIBILITY

1. Historical Background

   Not until the mid-1970s did litigants begin to exploit the graphic capabilities of computers. Computers in the early 1970s were limited in capacity and could not accurately reflect or compute many events used in litigation.(2) At that time, they were used infrequently as litigation aids. The first major case concerning the admissibility of a computer simulation, Perma Research and Development v. Singer Co.,(3) was decided in 1976.(4)

   In Penna Research, the Second Circuit upheld the trial court's admission of expert testimony based on computer simulations. The plaintiff's case relied almost exclusively on two expert witnesses. One expert's testimony was based on the results of a computer simulation to determine whether an anti-skid braking device was improvable. When the plaintiff refused to turn over the simulation's underlying data and theorems to the defendant, the defendant argued that without that information it did not have a fair opportunity to cross-examine the plaintiff's witness, and it moved to exclude the expert's testimony.

   The majority of the Second Circuit panel disagreed and held that the defendant had not shown that it did not have an adequate basis on which to cross-examine. Judge Van Graafeiland wrote a strong dissent, arguing that the defendant should have access to the information because the credibility of the simulation could not be attacked without it. He also noted that more exacting standards were needed for the introduction of computerized evidence. Thus, the majority's standard of admissibility, although challenged, was whether the opposing counsel could show it did not have an adequate basis for cross-examination.

   Throughout the 1980s, computer simulations proliferated in litigation. The admissibility standard was somewhat unclear, but a number of cases began to develop rules.(5) Since simulations normally used scientific theories, courts applied the standards for admitting scientific testimony and theories.

   These cases involved expert testimony based on simulations. The simulations did not show any pictures or animations. They generally applied complex mathematical calculations and simply produced numbers and answers. As computer technology developed and animation began to enter the courtroom, courts had to develop a new method to deal with graphic depictions of testimony.

2. Admissibility as Demonstrative Evidence

   At first, animations were classified as demonstrative evidence. The standards for admitting demonstrative evidence require that the demonstration fairly and accurately reflect the underlying oral testimony and aid the jury's understanding.(6) A witness must authenticate the demonstration.(7) Theoretically, demonstrative evidence does not add anything new to the case. It has no probative value and is used only to illustrate verbal testimony. Its purpose is to assist the trier of fact in putting the facts together.

   New York v. McHugh(8) is the first reported case to address a litigant's use of computer graphics during trial. The defendant attempted to introduce a computer-generated animation to establish that adverse weather conditions, not the defendant's alleged drunk driving, caused an automobile accident. The prosecution argued, as in Perma Research, that it should have access to the underlying formulas and techniques. The trial court overruled the objection, equating the animation to demonstrative evidence and terming it "more akin to charts and diagrams."

3. Admissibility as Substantive Evidence

   After McHugh, courts characterized computer evidence as either substantive or demonstrative. Generally, substantive evidence was not graphic in nature and was used to prove a fact in issue or to support a conclusion. Computer-generated substantive evidence usually was called a "computer simulation." Courts held that simulations had independent probative value, added new facts to the case, and thus had to meet substantive rather than demonstrative evidence standards.

   Graphic depictions admitted under a demonstrative standard were classified as "computer animations." Because demonstrative evidence is not considered to have probative value, the standard of admissibility for it is generally lower than for substantive evidence. Thus, animations are admitted under a lower standard than simulations.

   If evidence is characterized as substantive, further inquiry must be made. At least three standards now are used for admitting animations under a substantive approach: (1) scientific testimony, (2) accident reconstruction and (3) motion picture re-enactment.

   1. Scientific Testimony

      Scientific testimony usually is admitted as expert testimony governed by Rule 702 of the Federal Rules of Evidence. In Daubert v. Merrill Dow Pharmaceuticals Inc.(9) the U.S. Supreme Court held that "scientific" implies a grounding in the methods and procedures of science. If such scientific knowledge will assist the trier of fact to understand the evidence, it is admissible.

      The states have interpreted their evidentiary rules according to a number of different standards. The most popular approach for admitting scientific evidence was established in Frye v. United States,(10) in which the defendant attempted to introduce evidence from a polygraph test. The court held that the polygraph results were inadmissible because polygraph testimony had not gained adequate acceptance in the scientific community. The Frye approach effectively excludes "junk science" and untested hypotheses from the courtroom,(11) but it has been criticized as too restrictive in admitting valid new principles of science.(12) Although the Supreme Court held in Daubert that the standards of the Federal Rules of Evidence have superseded the Frye standard for admissibility of scientific evidence in federal court, there is a question as to the extent of the change wrought by Daubert,(13) and some state courts may continue to follow Frye.

      Some states follow alternative standards articulated by the federal circuits prior to Daubert. For example, courts have rejected Frye and adopted a four-part test for reliability. In United States v. Williams, the Second Circuit held that the more sensible approach was to measure: (1) the potential rate of error, (2) the existence and maintenance of standards, (3) the care and concern with which a scientific technique was performed and whether it appears to lend itself to abuse, and (4) whether the technique stands in an analogous relationship to routinely admitted scientific evidence.(14)

      Another approach was outlined in United States v. Downing,(15) in which the Third Circuit tried to create a rule that would not unduly restrict relevant evidence while safeguarding against evidence derived from unfounded scientific principles.(16) The court held that reliability stems from a number of factors, including general acceptance, existence of specialized literature dealing with the technique, qualifications and professional stature of the expert, non-judicial uses of the technique, rate of error of the technique, and expert testimony supporting the validity of the technique. This standard has been followed by many courts but has not been accepted by a majority.(17)

      Professor McCormick offers a similar multifactor balancing test to assess the reliability of the animation techniques.(18) He argues that the court should consider the potential rate of error in using the technique...