While writing his dissertation, Dan Simon began to wonder how judges make decisions not from a legal, sociological, or economic perspective but rather from a psychological one. Today, the USC law professor has built a career investigating how factors of the mind—such as memory, false confessions, and the framing of interviews—influence rulings in the criminal justice system.
Fred Lawrence: This podcast episode was generously funded by two anonymous donors. If you would like to support the podcast in similar ways, please contact Hadley Kelly at hkelly@pbk.org. Thanks for listening.
Musical Interlude
Lawrence: Hello, and welcome to Key Conversations with Phi Beta Kappa. I'm Fred Lawrence, Secretary and CEO of the Phi Beta Kappa Society. This podcast features conversations with Phi Beta Kappa visiting scholars who spend one academic year with us. They travel to up to eight Phi Beta Kappa affiliated colleges and universities, partake in the academic life on those campuses, and present a lecture on a topic in their field. Lectures are always free and open to the public. For a full schedule and to learn more about the program, visit pbk.org.
For today's episode, I'm so pleased to introduce you to Dan Simon, a professor of law and psychology at the University of Southern California. Dan is the author of In Doubt: The Psychology of the Criminal Justice Process, which focuses on the psychological causes of errors in the criminal process. He frequently speaks to judges and law enforcement personnel about psychological insights into the criminal process, such as the limited accuracy of evidence used in the criminal process, especially human testimony. Welcome, professor.
Dan Simon: I am grateful for being here. Thank you very much, Fred.
Lawrence: Dan, it's good to have you here. Criminal justice reform is certainly a big topic again—maybe it always is—with the First Step Act of 2019 heralded by some as a significant criminal justice reform. Others have criticized it, but it's also caused for a reevaluation of the Violent Crime Control and Law Enforcement Act of 1994, the so-called 1994 Crime Act. Now you've worked in a fascinating, and I'll call it troubling, corner of this area of the law: the very reliability of the way in which we make decisions of guilt in our system. And, I want to talk to you about your work in that, and how much we can rely on the whole system, but before we do that, I want to ask you a little bit about your own background. Tell us a little bit about where you grew up and what that was like.
Simon: I grew up in Israel, and I was educated at Tel Aviv University before coming to the United States and doing some graduate work in the United States. That is, advanced degrees in law. I taught for three years at the University of Haifa in Israel before moving to take the job at USC in Los Angeles.
Lawrence: What led you into the fields of law and psychology? When did you first realize that those subjects were related and that the relationship between the two was going to be of great interest to you?
Simon: Yeah, I really got, literally, sucked into this subject during my SJD degree, which as some of your listeners might know, is a sort of a PhD-like degree in law that's offered in just a handful of universities in the States. It's a dissertation degree, and I started off examining a question of “How do judges decide cases in times of national security?” That's a question that interested me since I had previously been a human rights lawyer in Israel, where the state of emergency is sort of both legally declared but also palpable in much of the Israeli courts’ jurisprudence. During the four year period of writing a dissertation, I started asking myself a deeper question, and that is, “How do judges make decisions at all?” That is, even if you take away the exigencies of an emergency set of circumstances.
And that brought me to do a lot of very serious reading: a good two years worth of being cooped up in a small apartment reading a lot of psychology—experimental psychology, mostly—in the fields of decision making and reasoning. And, my thesis ultimately was more general than I had originally began with, and that had to do with how do judges reason? How do they reach decisions? How do they explain their decisions from a psychological perspective rather than from a legal, or sociological, or economic perspective?
Lawrence: So, I did say that I wanted to spend some time talking about your work on the accuracy of evidence that's used in the criminal justice process, and it is very provocative work, so let's start with the investigatory process itself long before we ever get to trial. You've written a lot about the role that the investigatory process and the way it's done actually does have an impact on accuracy. Can you tell us a little bit about the cognitive aspects of the work that police investigators do and how they may sometimes actually lead to mistaken conclusions?
Simon: Most of legal academia is quite fascinated with the adjudicatory phase of the process, right? As our, sort of, consumers of legal TV shows and such, what I think we fail to perceive is that the evidence that we consume, sort of in the courtroom and subsequent proceedings, it's an appeals and post-conviction proceedings, we sort of take the evidence for given. A witness said something, so it's a fact, or a confession was made and signed, so that becomes a fact. And, these facts are, by the time we look at them again in the courtroom and beyond, they're pretty reified. They're well established. We don't question them, per-se. We work around the law that sort of tries to make sense of those facts.
What we, I think, fail to appreciate or at least under-appreciate is that those facts aren't necessarily an accurate reflection of what they purport to be. And, there is quite a lot of latitude and discrepancies between what happened at that sort of fateful event that we are trying to reconstruct in a courtroom miles away, months away, different people, different times, different contexts, different understandings. It's a difficult task to reproduce all of that and to render an accurate decision. And, for the most part, that sort of reconstruction is facilitated through witnesses. Now, to some degree we rely on objective or scientific evidence; maybe it's a video from a surveillance camera, or some DNA test, and so on and so forth—all of which, by the way, are also to some degree subject to some level of error. But, for the most part, for the most sort of time consumed in court, it's witnesses reporting memories of that event that happened some other time, some other place, and so on.
Lawrence: Right, particularly if the event took place someplace where a video camera wouldn't be, if it was in somebody's home, if it was in the stairwell of an apartment building, that kind of thing. There might not be the video camera the way there would be on a street or in an elevator.
Simon: Right, and it's true that we're getting more and more of this sort of digital or scientific evidence that just becomes more available. But still the... Like, DNA for example, is relevant to perhaps 10% of violent felonies. And, in the majority, in the vast majority of cases, there's no biological evidence left on the scene or that could be collected and examined. Yes, so there are cameras in some places, but many places are not covered by cameras. And, even if there is a camera, it's not always obvious that we can draw that crucial inference that we are seeking.
Lawrence: So, then we're left with people trying to recollect what happened.
Simon: Exactly. As it turns out, there are some ways in which whatever the person actually witnessed does not get communicated, at least not in the official way that the law accepted as a fact, which is in testimony in court somewhere down the line. There could be both what we call spontaneous errors, that's the errors that we all make just as it's part for the course, we forget things or we fail to notice things, or we're at a dinner and we forget the name of the person sitting next to us, and we don't remember where we put the keys, or we think we recognize somebody in the street but it’s not her.
Lawrence: Is that error, do we think, greater in the situation where someone becomes aware of the fact that she or he is observing a crime? Your pulse starts to pick up, you might be more nervous, so you're more likely to be more focused and get the details right? Or, are you more likely to get the details wrong, because you're thinking about other things?
Simon: That's a great question that doesn't have an equally great answer, because-
Lawrence: Well, give us your best answer.
Simon: So, on one dimension, you have a certain increase in accuracy, but up to a point. Basically what you're kind of describing is the level of arousal or engagement.
Lawrence: Right.
Simon: And, the studies go back decades and decades that suggest sort of an inverse U-shape. It's called the Yerkes-Dodson curve. That is, if you plot performance on the vertical Y-axis, and arousal on the horizontal X-axis, you'll have an inverted U. That is, if you're not aroused, you're not engaged; you're daydreaming, you’re reading your news feed on your phone, when at the time that a robbery is happening just in front of your face, right?
Lawrence: Right. You're not paying attention, so you don't process at all.
Simon: So, you're not processing, or even if you're seeing it but not interested in it, because there's no reason that you think anything's interesting happening... so, your performance then is low. In this case, it would be encoding for the purpose of subsequently remembering. When you get to the extremely high sort of levels of arousal, you're, as you suggested, you're basically overworked, you're overexcited.
Lawrence: Right.
Simon: Physiological factors kick in. You might be concerned about safety, about planning routes of escape and whatnot, or just being shocked and horrified, which also sort of can decrease our performance. In the middle, actually is when you're at your peak, when you're sort of not overly aroused but still engaged. So, in that regard, it really depends on the level of arousal. The second point is-
Lawrence: Well, we had a major national conversation about that in this country, not all that long ago, during the confirmation hearing for Justice Kavanaugh. In the question of whether Christine Blasey Ford's recollection could or could not be trusted under the circumstance. Do you have any view on what might or might not be likely in terms of her ability to recollect?
Simon: I would not venture to, sort of, judge her memory. Certainly, it seemed honest to start off with, though people sometimes lie about memories. That did not seem to me to be the case. But, again, I was just a consumer of her hearings through TV, nothing more than that. Mistaken memories, and this is taking us off on a whole different track, and as you probably know, I'd say in the 1990s I think, we started having lots of court cases revolving around what was called “repressed” memories at the time. By which-
Lawrence: Particularly in the area of sexual assault and domestic violence. That became a whole core element of those cases.
Simon: Right. I'm not trying to imply in the least that this was the case of Dr. Ford. But, just to say that issue at the time sort of sparked a pretty rigorous but also acrimonious split within the psychological community. Namely, between clinical psychologists on the one hand, who were mostly associated with these victims or the people who claim to have had these repressed memories, versus, on the other end, these sort of the cognitive psychologists, the people who study memory in a more experimental way, who had already been showing that memories are quite fungible and quite easy to distort through exposure of the person to post-event information, typically post-event misinformation.
Lawrence: What kind of evidence do we have on implicit bias that witnesses may be bringing to the entire situation that could affect what they see and therefore what they then remember?
Simon: Well, you're taking us very wisely even back one stage. And, that's asking the question, “To what degree does the witness perceive correctly what they saw?”
Lawrence: Exactly.
Simon: Which again—and I'll get to that in a second—but, what I want to point out to you is just how there's this whole chain of events which sometimes become a chain of unfortunate events, right?
Lawrence: Right.
Simon: But, in order to get an accurate outcome at the end of the process, that is for the jury to convict the true perpetrator and acquit the innocent person, lots of things have to fall in place, and any one of them actually can ruin the whole endeavor. So, for starters, we need the witness to actually perceive the event correctly or to encode the face of the suspect correctly. And, what we do know is that perception in of itself is very much driven by our knowledge schemas with which we enter the task. That is, the things such as beliefs about people, or groups of people, or professions, or skin color, or affiliation with other groups, gangs and such.
All of these can affect how we actually perceive the event in the first
place and quite easily—and this work dates back to the 1930s—is
that how the schemas actually affect our perception, that memory is to
some degree we're actually encoding. But, it's not a single ...
like a photo. Memories are really an aggregation of verbatim
details, sort of pieces, fragments that are visual or auditory and whatnot.
But, also our interpretations of them, and also emotions and
understandings and context as we mentioned before.
Lawrence: Let me shift from the identification by a witness now to the suspect him or herself. If identification by a witness is a strong piece of evidence, there's probably no more compelling piece of evidence than the confession. And, yet, we know that many times there are false confessions. What would possibly cause someone to confess falsely?
Simon: So, you're asking that, it's a million dollar question, which to some degree we all share, or at least surveys show that 90% of respondents in the US do not believe that innocent people would confess to a crime they didn't commit, short of torture and other kinds of harsh depravation. And this has, to my mind has led to over the course of history, to a great number of false confessions to actually have been misjudged and sent people to jail if not to death.
So, it is indeed not just counterintuitive, it's mind boggling when we think of why an innocent person might confess. One sort of quick little take on this is to actually flip the question around and to ask why would a guilty person confess?
If you think of it, a guilty person's probably got more to lose, and at least perceives that he or she has more to lose, because-
... an innocent person might say, "Okay, I'll get out of this horrible room with these people yelling at me, and I'll call my lawyer and we'll straighten all of this out." But, guilty people don’t - probably don't - feel that way, yet they do confess, too. We know that.
Now, we don't know the relative incidence of what percent of the confessions ultimately obtained are from truly guilty versus truly innocent people. We do know that roughly about 50% of the people subjected to interrogations will ultimately provide either a full blown confession or some partial admission of guilt. So, it's an effective tool. The question is: Is it also an accurate tool? Does it have diagnostic value in that it sort of sifts apart the guilty from the innocent?
Lawrence: Right, because clearly it's going to have enormous impact on either a juror who's hearing the confession as part of the case or as a judge who is doing the sentencing.
Simon: Absolutely. It's extremely powerful evidence. And, in addition to the fact that confessions just tend to be believed on their face, they actually also have sort of less noticeable or less noticed effects, in that once the police get a confession, it colors the rest of the investigation, meaning a witness is more likely going to pick out someone when they know that that person has confessed in a lineup.
A forensic examiner is more likely going to declare a match of, let's say, a fingerprint or a shoe print match knowing that that particular... whoever provided that sample actually also confessed. So, there are a lot of these sort of subterranean counter-normative, but unfortunately sort of, you know, quite realistic, added effects to the inculpatory power of confessions. But, to get to the core of your question is: Why on earth would an innocent—or as I suggest, anybody—confess? It cannot be about incentives, right? Normally, we say, "Oh, it's quite impossible that someone would confess, because it's just so detrimental to their better interests."
And, our intuition is really driven by incentives, but as I mentioned, the incentives are the same for innocent people and for guilty people. Both are really sticking a stake through the heart of their defense. So, it can't be about incentives, because in both cases we're not permitting torture or physical abuse, and-
Lawrence: Yeah. Putting coercion aside, assume... Or, putting undue coercion—I mean, obviously the entire context of an interrogation is coercive to a certain extent—but, putting that undue coercion aside, we do think that innocent people will in certain instances still confess to a crime they did not commit. Isn't that right?
Simon: Absolutely. It's not just that we think; we know that. We have today almost 2,500 exonerations that have been recorded in the past couple of decades. All of these are accessible on a website called theexonerationregistry.org.
And, you've got a break-down there by crimes, and years, and such. Out of that number of exonerations, we've got almost 1,000 exonerations, that is, people who were convicted - in this case for murder - and were subsequently, their conviction was overturned primarily because of factual sort of mistakes. That is, we got it wrong. So out of those almost 1,000 murder exonerations, we've got about 23% of them were... these are cases in which the supposed murder actually provided a confession. That usually-
Lawrence: That's astonishing.
Simon: Yeah. In that respect, we've got 200 plus known exonerations that were based on false confessions, but we know that there are many more. I'm a subscriber to the belief that whatever exonerations we have, it is the tip of an iceberg. So, think of all the confessions that led to convictions that will never be overturned, because there is no DNA, or there is no witness who comes... walks back into court 20 years later and wants to repent for falsely framing someone. So, it's exceedingly hard to actually overturn a conviction on factual basis. Exonerations are somewhat... There's no doubt in my mind that they vastly undercount the actual false convictions. These are just the lucky ones who managed to get enough evidence and enough legal aid and whatnot to overturn.
Lawrence: So, Dan, with all of the challenges your research has found, suppose that Senator Grassley, The Chair of the Senate Judiciary Committee, and Congressman Nadler, The Chair of the House Judiciary Committee, reached out to you and they said, "We've got criminal justice reform legislation in place, and we're going to give you one provision to add." What's that one provision? What would you add?
Simon: Can I haggle for two?
Lawrence: Yeah, sure, you can have two.
Simon: Follow best practices, which are available. They're science-based, and that was one. And the second is tape everything, that is every encounter with every witness. And, share the tapes with everybody concerned, that is with the jurors, the judges, the lawyers and everything. I think, we—going back to where we started, that sort of light is a very powerful disinfectant—we would gain a lot by actually seeing exactly who said what to whom. How did the lineup take place? What was said? When did the witness express confidence: before or after they got a high five from the detective? And so on and so forth. Who suggested the crucial detail in the interview? Might it have been the detective actually asking a leading question, oftentimes inadvertently, and the witness agreeing, and then that becomes the fact, even though the witness really never recalled it? How voluntary, and more importantly, how reliable was the confession rendered? It's a difficult-
Lawrence: And, if we have that tape, we've got a better shot at being able to actually going back and determining that.
Simon: Yes, yes, it's not an easy task, but we're in much better position to begin to assess that.
Lawrence: Well, Dan, thank you so much for coming in today. You've raised some very thought-provoking, problematic challenges, but it does seem to me that as criminal justice reform issues continue to occupy us, your work is going to play an important role in that. Thank you for being a Phi Beta Kappa Visiting Scholar in this past year, and thank you for the work that you're engaged in.
Simon: It's been my honor to be a Phi Beta Kappa Scholar, and it's my honor to be with you in this interview, so thank you very much for including me.
Musical interlude
Lawrence: This podcast is produced by Lantigua Williams & Co., Paolo Mardo was our sound designer, Hadley White is the Phi Beta Kappa producer on the show. Emma Forbes is our intern, and our theme song is Back to Back by Yan Perchuk. To learn more about the Phi Beta Kappa Visiting Scholar Program, please visit pbk.org. Thanks for listening. I'm Fred Lawrence. Until next time.
CITATION:
Lawrence, Fred, host. “USC Professor Dan Simon on the Intersection of Law and Psychology.” Key Conversations with Phi Beta Kappa, The Phi Beta Kappa Society, April 24, 2020. www.pbk.org.