Animation vs Simulation


In the realm of litigation, the term “animation” is usually associated with computer generated exhibits that help explain the opinion or testimony of a witness. To be admissible in court, it must be judged as relevant, accurate, and fair. Compare this with a “simulation” which typically forms the basis for expert testimony, and must be prepared using generally accepted scientific principles.

These two terms appear in court decisions on admissibility going back to the 1990s. Given this precedent, and these details, you’d expect judges to first determine whether a proposed exhibit is an animation or simulation, in order to know what standards for admissibility should apply. However, it is common for computer generated exhibits to contain both animated and simulated elements, making them difficult to categorize.

I’ve been thinking about this issue over the past week, after watching two forensic experts, Eugene Liscio and Matthew Noedel, talk about the difference between shooting incident re-creation vs re-construction. (aka animation vs simulation)

If you watch more of Eugene’s videos (which I highly recommend for anyone involved in gun violence cases) you’ll hear him repeatedly emphasize that it’s the process that matters most, regardless of whether an exhibit is labeled an animation or simulation.

Now, let’s get back to that Las Vegas bus crash case. Fat Pencil built an accurate 3d model and used reverse projection photogrammetry to determine the path of the bus through the intersection. We collaborated with accident reconstruction and biomechanics experts to determine the most likely scenario for how a bicycle rider was fatally injured in the crash, and how the result might have been different if the bus had been equipped with rear tire guards.

We even mocked up possible positions of the bus and bicycle before the crash to test visibility against different witness statements. This, in itself, is not unusual for Fat Pencil. We’ve done it for dozens of cases, usually during a live meeting or video call with attorneys, investigators, and expert witnesses. What was novel in this case is the plaintiff’s attorney offered the 3d model itself as an exhibit during trial, so he could recreate scenarios leading up to the crash in real time with the jury watching. The judge agreed to this demonstration, so long as defense counsel was given equal opportunity to ask for scenarios that he wanted the jury to see.

I was concerned that jurors might get confused about how much of what they were seeing was based on science vs speculation. Here’s how I introduced the demonstration, paraphrased from the trial transcript:

I should point out that while we do use photogrammetry and 3d visualization together, they are not the same thing. Using 3d visualization, you can actually test a whole variety of circumstances… some based on science like photogrammetry, others based on witness statements, and still others that may be completely hypothetical. You’ll see all three of those today, and it’s important to think about the foundation or the evidence behind what you see in the 3d visualization. Just because you see a picture on the screen doesn’t mean it’s based on a strong foundation.

March 5, 2018 Trial Transcript

In other words, what the jury got to see was part simulation, part animation, and fully transparent about the process that informed my testimony. It was effective, and I say this not because of the outcome (which was good for the plaintiff), but because jurors stayed highly engaged throughout the demonstration. This was not an obscure discussion of technical terms, but a visual story playing out in front of their eyes… one that allowed them to evaluate different points of view and reach their own conclusion about which scenario was most credible.

My experience on this case, and many others over the past ten years, reinforces my belief that elements of both animation and simulation should be considered when developing a visual story. Today, if someone asks me whether to create animations or simulations for a case, here’s what I'll suggest:

  1. Computer generated animations are most effective when based on an accurate model that incorporates all relevant evidence.
  2. Simulations are most effective when presented as part of a compelling visual story that explains the science behind the conclusion.

Joshua Cohen is a Principal at Fat Pencil Studio