BREWER, J.

A jury found defendant guilty of murder. The Court of Appeals reversed defendant's conviction and remanded the case based on its conclusion that the trial court had erroneously admitted eyewitness testimony of two witnesses who identified defendant as the perpetrator. On review, we conclude that the trial court properly admitted the challenged identification testimony of one of the witnesses. We also conclude that any error in admitting the identification testimony of the other witness under OEC 403 was harmless. Accordingly, we reverse the Court of Appeals' decision and affirm the judgment of the trial court.

I. FACTUAL AND PROCEDURAL BACKGROUND

We begin with an overview of pertinent evidence the admission of which is not challenged on review. On December 31, 2007, a number of men, including defendant and another man, Porter, attended a party at a house in Portland. When Porter arrived at the party, he observed a fight in progress, during which a man ran into the house. That man was Christopher Monette, who was later shot and killed. Soon thereafter, "[w]ords were exchanged" between Porter and Monette. The exchange was sufficiently heated to cause two other people to intervene. Porter pulled out a pocketknife, because Monette was "a big individual." Shortly thereafter, defendant arrived, and he also exchanged words with Monette. The argument stopped and defendant walked away.

Porter, a convicted felon who testified with the hope of receiving lenient treatment on criminal charges that were pending against him, testified that defendant then grabbed a ski mask out of Porter's back pocket. According to Porter, defendant put on the ski mask, approached Monette, and shot him four times with a handgun in front of several eyewitnesses. Monette died at the scene. Porter testified that, after shooting Monette, defendant walked into the street and fired several shots in the air.

Defendant's uncle, Miller, another convicted felon who testified with the hope of receiving leniency on an unrelated criminal charge, also was an eyewitness to the shooting. Miller, too, testified that defendant was the shooter. Miller stated that, after shooting Monette, defendant took off the ski mask and left the scene. Three other people, Anderson, Grant, and Pskar; none of whom could specifically identify defendant as the shooter, each provided eyewitness testimony that the shooter was an African-American male, approximately 5'7" tall, and with a stocky build. Defendant is 5'6" tall and has a stocky build. Porter is 6'1" tall.

After the shooting, people fled from the party on foot and in cars. The police arrived within minutes of the shooting. Officer Mast approached Porter and defendant, who were walking away from the scene. Porter stopped to talk to Mast, but defendant continued walking away. At that point, a woman, G, ran up yelling and screaming; she claimed that the shooter was getting away in a car. Hearing for the first time that someone had been shot, Mast went with G to the driveway and found Monette's body. G identified the shooter as "Cello." The car that she identified the shooter getting into was stopped. Moncello James, also known as Cello, was not in the car, but his identification was found there.

After his encounter with Officer Mast, defendant fled from the area. During his flight, both of his shoes came off, and he lost his watch when he jumped over a fence. As he approached a nearby golf course, he jumped over another fence and fell on the other side, breaking his leg.1

At the crime scene, the police found the handgun and the ski mask. They submitted the ski mask to the crime lab for DNA testing. The lab found DNA from three people on the ski mask. The lab further determined that defendant was the primary source of the DNA.

Shortly before the shooting, two women, D (19 years old) and N (18 years old), had arrived by car at the house party. D and N are both white. The east side of Portland was "out of [D's] element." D and N were in the back seat of the car. D told a police investigator on the night of the shooting that "she didn't see the shooting and really couldn't describe much. Knew that there was an argument occurring, but could not give specific descriptions of who was involved." D also told the investigator that another man, who identified himself as "Corey," jumped into the car as it left the scene.

N told police on the night of the crime that she witnessed the shooting and that the perpetrator was a "black male, stocky, in his mid-twenties, and wearing a do-rag."

D was interviewed by a defense investigator a few weeks before defendant's trial. In that conversation, D told the investigator that she could describe the men in the altercation only as "big black men." According to the investigator, D explained that "all black men look the same" to her. At trial, D denied making that statement. D told the investigator that the shooter had a "big Afro," but could give no further details about the shooter's hair. A day later, in an interview with the prosecutor, D stated that the shooter had "twisties" with "close black hair." In that interview, D told the prosecutor that she was not certain that she could identify the shooter. In response, the prosecutor proposed that, at trial, D should signal him with a "look in the eye" if she recognized the shooter while on the witness stand. The prosecutor told D, "If you do [recognize the perpetrator], then let the Court know — let the trier of fact know. If you don't, then you don't."

Between the night of the crime and defendant's trial, 23 months passed. During that time, the state made no attempt to have D or N identify the perpetrator, nor did the state inform defense counsel that it intended to ask D or N to make an in-court identification of the perpetrator at trial.

D testified on the third day of defendant's trial. At that time, defendant was present in the courtroom and seated next to his counsel. Defendant was the only African-American in the well of the courtroom, although there were six to 12 African-American men seated in the back of the courtroom. D was aware that defendant was the person charged with Monette's murder. Shortly after the state began its direct examination of D, an equipment malfunction occurred in the courtroom, and the court recessed. As the jurors left the courtroom, everyone, including defendant, stood up. The court staff cleared the public from the courtroom, but defendant remained in the courtroom with his counsel. D left the stand and walked past defendant into the hallway. One of the prosecutors accompanied D as she left the courtroom and noticed that she was hyperventilating. D said to the prosecutor: "Oh, my God, that's him, that's him, that's him." Without saying anything to her, the prosecutor sat D down next to D's mother. During the recess, D had no contact with any of the other witnesses.

After the court resolved the equipment malfunction, D resumed the witness stand. D testified that, before the shooting, she saw three or more African-American men fighting near the front door of the house. She stated that there were 25 to 50 other people in the yard. Monette wore a tank top, which he took off during the fight. The other people whom D noticed "pretty much all looked, like, the same. They were all wearing really baggy clothing and many of them were very husky gentlemen."

D testified that the overhead street lighting at the scene was "fluorescent." D further testified that, before the shooting, she focused her attention on the shooter and the victim because the two men were engaged in an argument. She stated that she "got a good view of both of the gentlemen." D explained that she was talking to N and not looking in the direction of the shooting when it occurred. D testified that, moments after the shooting, she saw one of the men who had been fighting fire several gunshots into the air. According to D, the shooter was then standing 12 feet away from her and under street lighting. D described the shooter as being black, in his 20s to early 30s, stocky, tall 5'7". to 6'), and having a "close" Afro hairstyle or braids. She also described his facial features. The prosecutor asked D if she saw that person in the courtroom, and D said that she did.

Before D identified anyone, however, defendant objected, citing the Due Process Clause of the Fourteenth Amendment to the United States Constitution and OEC 403. Outside the jury's presence, the trial court had a lengthy discussion with counsel and, ultimately, overruled defendant's objection. The state resumed its direct examination, and D identified defendant as the man whom she had seen firing gunshots into the air. She explained to the jury that, before entering the courtroom, she did not know whether she would recognize the shooter. She further explained that, after the equipment malfunction occurred and when she walked into the hallway, she became emotional and told the prosecutor "that that was the shooter, that [was] him." D testified that she was 95 percent certain of the accuracy of her identification.

D also testified that, after the shooting, people started "running westbound, jumping into cars, cars were leaving." D stated that, as the car she was riding in started to drive away, she saw a man run toward the car; according to D, the man tried to get in the car. She thought he was the shooter, but wasn't sure. The car that D and N were riding in was driven a few blocks from the house before the police stopped it.

N testified on the fourth day of trial. As with D, in the period of time between the shooting and her testimony at trial, N had not taken part in an out-of-court identification procedure and had not identified anyone as the perpetrator. N testified that she heard one or two gunshots and ducked down; she then looked up and saw the shooter, who was standing 20 to 25 feet away. N testified that, after the shooting, a man that she believed was the shooter came to the car; she wasn't sure whether he was trying to get into the car or was fighting with one of the passengers. However, that passenger repelled the man.2

N repeated her pretrial description that the shooter was a "black male, stocky, in his mid-twenties, and wearing a do-rag." In addition, N testified that the shooter was 5'7" tall and that his hair was about three inches long and "nappy." N stated that the shooter was not wearing a head covering when he came toward the car she was riding in and that she got a good look at him from close range. Although N had not given particular information about the perpetrator's hair to police during her initial interview, she had given it to the prosecutor during an interview about two weeks before trial. N testified that she did not give a more specific description of the perpetrator on the night of the crime because she was upset.

The prosecutor asked N if she saw the perpetrator in the courtroom, and N said that she did. Defendant objected on due process grounds and pursuant to OEC 403. The trial court overruled the objection, and N identified defendant as the perpetrator.

Defendant offered expert testimony that eyewitness identification testimony of the sort given by D and N, more than two years after a stressful event, was inherently unreliable.

The jury convicted defendant of Monette's murder. On appeal, the Court of Appeals reversed. The court concluded that, under this court's decision in State v. Lawson/James, 352 Or. 724, 765, 291 P.3d 673 (2012), D's and N's in-court identifications of defendant as the shooter were problematic; the court therefore remanded the case for a new hearing on the admissibility of that evidence, based on the considerations prescribed in Lawson/James. State v. Hickman, 255 Or.App. 688, 298 P.3d 619 (2013). On review, the state asserts that the challenged evidence was admissible and that, even if the trial court erred in admitting it, the error was harmless in light of other evidence of defendant's guilt.

II. ANALYSIS

A. The Lawson/James Framework

We begin with a brief summary of the analytical calculus prescribed in Lawson/James. Under that framework, "when a criminal defendant files a pretrial motion to exclude eyewitness identification evidence, the state — as the proponent of the eyewitness identification — must establish all preliminary facts necessary to establish admissibility" under generally applicable provisions of the Oregon Evidence Code (OEC). Lawson/James, 352 Or. at 761, 291 P.3d 673. In particular, the focus of the court's opinion — consistently with its presenting circumstances — was on the admissibility of eyewitness testimony that was asserted to have been tainted by suggestive pretrial police procedures. Thus, the court stated:

Id. at 741, 291 P.3d 673 (emphasis added).3

Where such a challenge implicates OEC 602 or OEC 701, the state must provide "proof under OEC 602 that the proffered eyewitness has personal knowledge of the matters to which the witness will testify, and proof under OEC 701 that any identification is both rationally based on the witness's first-hand perceptions and helpful to the trier of fact." Lawson/James, 352 Or. at 761-62, 291 P.3d 673. If the state satisfies its burden, the burden shifts to the defendant to prove "under OEC 403 that, although the eyewitness evidence is otherwise admissible, the probative value of the evidence is substantially outweighed by the danger of unfair prejudice, confusion of the issues, misleading the jury, or by considerations of undue delay or needless presentation of cumulative evidence." Id. at 762, 291 P.3d 673.

The court in Lawson/James identified two categories of factors that affect the reliability and, thus, the probative value, of eyewitness identifications: so-called "estimator variables" and "system variables." The term estimator variables refers to "characteristics of the witness, the alleged perpetrator, and the environmental conditions of the event that cannot be manipulated or adjusted by state actors." Id. at 740, 291 P.3d 673.4 In Lawson, the court concluded that a number of estimator variables undermined the reliability of the identification evidence. In that case, the eyewitness, who had sustained a critical gunshot wound, was under "tremendous stress and in poor physical and mental condition," which the court noted would "tend to impair a witness's ability to encode information into memory." Id. at 763, 291 P.3d 673. Further, the environmental conditions under which the eyewitness viewed the perpetrator were poor: it was dark and the perpetrator had covered the eyewitness's face with a pillow to obscure her view. The eyewitness viewed the perpetrator for only a few seconds, and the perpetrator wore a hat that concealed his hair, a key identifying feature. Finally, the eyewitness's in-court identification of the defendant occurred more than two years after the incident. By contrast, the court was less concerned with the estimator variables in James, which involved a robbery in which the eyewitnesses came face-to-face with the perpetrators and observed them for an extended period of time. Additionally, the eyewitnesses in James described the appearances of the perpetrators in detail, noting their "race, height, weight, and clothing." Id. at 765, 291 P.3d 673.

System variables, by contrast, relate to "circumstances surrounding the identification procedure itself that are generally within the control of those administering the procedure."5 Id. at 740, 291 P.3d 673. In Lawson, the court concluded that a number of system variables were implicated. Specifically, when the police first interviewed the eyewitness, she was hospitalized and in a "fragile physical and mental condition." Id. at 764, 291 P.3d 673. At that time, the police asked leading questions and "implicitly communicated their belief that [the] defendant was the shooter." Id. Moreover, before identifying the defendant as the perpetrator, the eyewitness viewed the defendant in two photographic lineups, in a newspaper article photo, and at a hearing to which police had brought her. Finally, the court considered significant the alterations in the eyewitness's statements over time. She initially told police that she had not seen the perpetrator's face, but she later identified the defendant as the perpetrator "[a]fter a series of leading questions" by the police. Id. at 765, 291 P.3d 673.

Conversely, in James, this court concluded that system variables did not require exclusion of the eyewitness identification evidence, despite the court's determination that the police had conducted a suggestive showup. Central to that conclusion was the accuracy with which the eyewitnesses described the perpetrators' "unique features" before the suggestive showup. Id. at 767, 291 P.3d 673. Those unique features included the perpetrators' clothing and a particular bottle of beer found in the defendant's backpack.

Ultimately, "[t]he decision whether to admit, exclude, or fashion an appropriate intermediate remedy short of exclusion is committed to the sound exercise of the trial court's discretion." Id. at 762, 291 P.3d 673. In Lawson/James, the court noted that "it is doubtful that issues concerning one or more of the estimator variables that we have identified will, without more, be enough to support an inference of unreliability * * *." Id. However,

Id. at 763, 291 P.3d 673. In such cases, the trial court assumes a "heightened role as an evidentiary gatekeeper because `traditional' methods of testing reliability — like cross-examination — can be ineffective at discrediting unreliable or inaccurate eyewitness identification evidence." Id. at 758, 291 P.3d 673.

B. Framing the Issues — The Parties' Arguments

Unlike in Lawson/James, the challenged identifications in this case occurred at trial in the absence of any preceding attempts by the state to have the witnesses identify Monette's murderer. That is, no suggestive pretrial police procedures were administered to either D or N. Instead, with the possible exception of a single pretrial interaction between D and the prosecutor that we discuss in greater detail below, no suggestive state procedures of the sort with which the court was concerned in Lawson/James were administered in this case; the entire identification process occurred in the courtroom setting. The question before us is whether and, if so, how, those variations in identification procedures implicate and affect the analysis under Lawson/James.

The state asserts that, where no suggestive out-of-court identification process preceded an in-court identification, several of the factors discussed in Lawson/James are inapplicable to the analysis under OEC 403, and the resulting risk of unfair prejudice to the defendant is necessarily low. That is so, the state reasons, because the in-court identification is itself sworn `testimony that occurs in the presence of the jury and the defense team. Thus, according to the state, and in contrast to a pretrial identification that follows suggestive police procedures, any unreliable aspects of the in-court identification are exposed to the view and evaluation of everyone present. It follows, the state urges, that the trial court's evidentiary gatekeeping function in such circumstances is minimal, and the credibility of the in-court identification is for the jury to assess.

Defendant responds that the logic of Lawson/James is even more compelling when first-time identifications occur in front of a jury. As defendant sees things, the safeguards intended to prevent unreliable identifications are almost entirely absent in such circumstances, thereby increasing the likelihood that unreliable evidence will be admitted. In addition, defendant asserts, any exception to Lawson/James for first time in-court identifications would create a powerful incentive — and the means — for prosecutors to circumvent the protections designed to reduce wrongful convictions. According to defendant, nothing about the administration of an eyewitness identification procedure in the presence of a jury justifies the application of less than a complete Lawson/James analysis. Moreover, defendant argues, in this case only a small portion of the identification process involving D occurred in front of the jury. Defendant notes that the plan whereby D — once on the witness stand — would signal the prosecutor with a look if she could identify the shooter, was unknown to jurors and defense counsel who, therefore, could not watch for and evaluate the reliability of the identification if and when it occurred. Defendant also points out that the instant when D claimed to recognize defendant as the perpetrator appears to have occurred when the jury was out of the courtroom and that D's accompanying emotional reaction occurred in the hallway when neither jurors nor defense counsel were present.

C. First Step of Lawson/James Analysis — A Foundational Inquiry

Unlike cases relying on federal constitutional principles, this court's decision in Lawson/James is rooted in evidentiary considerations that are governed by the Oregon Evidence Code. See Lawson/James, 352 Or. at 746-48, 291 P.3d 673. The first step is foundational, and it provides the "minimum baseline of reliability" for eyewitness identifications. Id. at 758, 291 P.3d 673. That step implicates three interrelated evidentiary concepts: relevance under OEC 401, personal knowledge under OEC 602, and lay opinion under OEC 701. Id. at 752, 291 P.3d 673.

To be relevant, the identification must have some "tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." OEC 401. In addition, to satisfy the first step of the analysis, the witness must have the personal knowledge necessary to make an identification, OEC 602, and the identification must be rationally based on that knowledge and be helpful to the jury. OEC 701.

We take this opportunity to clarify how OEC 602 applies, because the court in Lawson/James did not fully elucidate its foundational nature. Oregon has adopted a liberal standard for determining who may be a witness. Equitable Life Assurance v. McKay, 306 Or. 493, 760 P.2d 871 (1988). OEC 601 provides:

The only exceptions to OEC 601 are found in OEC 602 (requiring proof that a witness has personal knowledge of the matter to which he or she testifies), OEC 603 (requiring an oath or affirmation), OEC 604 (providing for the use of interpreters), and OEC 605 and 606 (prohibiting the presiding judge or any member of the jury from testifying). Otherwise, "[f]erreting out and discounting biased testimony is treated as a question of believability for the jury, not admissibility for the court." Equitable Life Assurance, 306 Or. at 498, 760 P.2d 871.

OEC 602 provides the basis for the first step of the Lawson/James "personal knowledge" criterion:

(Emphasis added.) "[W]hether the witness has personal knowledge is a matter of conditional relevancy" under OEC 104(2). Laird C. Kirkpatrick, Oregon Evidence § 601.03[3] (6th ed 2013). OEC 104(2) provides:

Under that rule, the proponent of evidence need only present evidence from which a juror could find the fact. In that regard, the Legislative Commentary to OEC 104(2) warns:

As this court noted in Lawson/James, OEC 602 expressly allows for proof of personal knowledge to consist of the witness's own testimony. Lawson/James, 352 Or. at 753, 291 P.3d 673. This court also noted that, pursuant to ORS 44.370, a "witness is presumed to speak the truth." Id. at 752 n. 8, 291 P.3d 673. In addition, the Legislative Commentary to OEC 602 provides:

Legislative Commentary to OEC 602, reprinted in Kirkpatrick, Oregon Evidence § 602.02.

Given those principles, an identification satisfies OEC 602 if the eyewitness testifies to facts that, if believed, would permit a reasonable juror to find that the eyewitness observed the facts necessary to make the identification. Whether the eyewitness actually did observe those facts is a credibility determination for the jury. So-called "system variables" do not apply to the trial court's OEC 602 determination, because they affect the witness's recollection of her observations, not the observations themselves. See Lawson/James, 352 Or. at 741-44, 291 P.3d 673 (describing system variables).

OEC 701 provides the basis for the final two criteria. That rule limits a witness's testimony regarding inferences and lay opinions to those that are "[r]ationally based on the perception of the witness" and "[h]elpful to a clear understanding of testimony of the witness or the determination of a fact in issue." In describing the first of those two criteria, this court stated in Lawson/James:

352 Or. at 754-55, 291 P.3d 673.6 The last criterion — that the identification is helpful — merely requires that the identification "communicates more to the jury than the sum of the witness's describable perceptions." Id. at 756, 291 P.3d 673.

D. Application of OEC 602 and OEC 701

There can be little doubt that D's and N's identifications of defendant were logically relevant under OEC 401. But, as noted, whether a witness has the requisite personal knowledge to testify under OEC 602 is a matter of conditional relevance under OEC 104(2). Therefore, the state was required to adduce evidence from which a rational juror could find that D and N were able to make observations sufficient to support their identifications of defendant. If the state adduced such evidence, whether D and N actually had sufficient personal knowledge was a credibility issue that ultimately was for the jury to resolve.

The in-court identifications in this case satisfied that foundational standard for admissibility. D testified that she was able to view the perpetrator immediately after the shooting and that she observed his facial features at that time. Immediately before the shooting, she focused her attention on the perpetrator and the victim because the two men were engaged in an argument. She testified that she "got a good view of both of the gentlemen." Moments after the shooting, D testified, she saw the perpetrator standing in the street firing shots into the air. At that time, he was illuminated by overhead fluorescent lighting. D was about 12 feet away from the perpetrator when she saw him. At trial, D testified that, at the time of the shooting, she noticed that the perpetrator was African-American, in his 20s to early 30s, had a "stocky" build, his hair was in braids, and had particular facial features. D was relatively young, and there was no evidence that she had any sensory deficits or that her senses were otherwise impaired on the night of the shooting.

N, in turn, testified that she was able to observe the perpetrator at the time of the shooting. Immediately beforehand, her attention was focused on the perpetrator because he was arguing with the victim in the front yard of the house. When the shooting occurred, the perpetrator was 20 to 25 feet away from N. She indicated that, after the shooting, he came up to her car window and tried to get in the car and that she got a good look at him at that time. N observed that the perpetrator was African-American, "stocky," about 5'7". tall, and wearing jeans and a t-shirt. N also was relatively young, had no reported sensory deficits, and there was no evidence that her senses were impaired when she made her observations. Based on that evidentiary record, a reasonable juror could find that both D and N had sufficient personal knowledge to make the in-court identifications.7

The challenged identifications also satisfied the OEC 701 inquiry:

Lawson/James, 352 Or. at 755, 291 P.3d 673. Here, both witnesses testified at trial that they had a clear view of defendant's face immediately after the murder. D testified that she observed defendant's facial features while he was standing 12 feet away from her and under street lighting. N testified that she got a good look at defendant at close range when he ran up to her car window immediately after the shooting. In admitting the evidence, the trial court specifically found that the witnesses were situated "in such a way as to have a direct view, fairly close, of the person she purport[ed] to identify." In short, the evidence showed, and the trial court found, that the witnesses perceived sufficient facts to support inferences of identification.

Moreover, there were no suggestive pretrial police procedures that created a competing inference that the in-court identifications were derived from an impermissible source. Although the prosecutor told D to give him an eye signal if she recognized the perpetrator while testifying at trial, the trial court determined that D's in-court identification of defendant was "routine" and that there was no suggestiveness in the process beyond that inherent in a courtroom setting. Under those circumstances, the identifications — buttressed by the trial court's findings — satisfied the first part of the analysis under OEC 701.

The identifications also satisfied the second part of the OEC 701 analysis. Although the court in Lawson/James anticipated that the helpfulness requirement "will be easily satisfied in nearly all cases," it gave the following example of an exception:

352 Or. at 755-56, 291 P.3d 673. No similar deficiency existed with respect to the challenged identifications in this case. They were helpful to the trier of fact because they conveyed more "than the sum of the witness's describable perceptions." See Lawson/James, 352 Or. at 756, 291 P.3d 673.

Because the state presented an adequate foundation under OEC 401, OEC 602, and OEC 701, for both courtroom identifications, they satisfied the first prong of the Lawson/James test.

E. OEC 403 and First Time In-Court Identifications

OEC 403 provides:

If the state establishes a foundation for the admission of eyewitness identification evidence under OEC 602 and OEC 701, "the burden shifts to [the] defendant" to demonstrate that the danger of unfair prejudice substantially outweighs the identification's probative value. Lawson/James, 352 Or. at 762, 291 P.3d 673.

So-called system variables are particularly important under OEC 403 when "an eyewitness has been exposed to suggestive police procedures." Id. at 758, 291 P.3d 673. For eyewitness identifications involving that "discrete evidentiary class," the trial court has a heightened role as an evidentiary gatekeeper because the danger of unfair prejudice increases. Id. However, as this court cautioned,

Id. at 763, 291 P.3d 673 (emphasis added). That is, under OEC 403, exclusion requires that the state-administered identification procedures have been suggestive in one or more ways. Although this court used the term "suggestive police procedures" in describing the type of actions that produce an unreliable identification, the state acknowledges, and properly so, that suggestive procedures also can be administered by other state actors. The key issue in this case is whether the courtroom setting itself was so inherently suggestive as to make the challenged first time in-court identifications unfairly prejudicial within the meaning of OEC 403.

Although — insofar as it involves in-court identifications not preceded by suggestive pretrial identification procedures — this case presents an issue of first impression under the Oregon Evidence Code, other courts have addressed the problem. The concerns with in-court eyewitness identification, where suggestive pretrial identification procedures were administered by the state, are essentially those that this court described in Lawson/James. For example, the witness may identify the person in court, not based on his or her recollection of observations at the time of the crime charged, but as a result of the suggestive pretrial identification. Because the factfinder was not present to observe the pretrial identification, the factfinder is unable to observe the witness making that initial identification. Thus, variables such as indications of witness certainty or hesitation during the identification process, including facial expression, voice inflection, and body language, and any other observations pertinent to assessing the reliability of a person's statements, are unavailable to the factfinder for evaluative purposes. Moreover, where an in-court identification confirms an earlier identification, there is a risk that the later identification will be expressed with greater certainty than the earlier identification. See Lawson, 352 Or. at 741-44, 291 P.3d 673.

On the other hand, when a first-time eyewitness identification occurs in court and no suggestive pretrial identification procedures were administered by the state, courts generally have concluded that the factfinder is better able to evaluate the reliability of the identification because he or she can observe the witness's demeanor and hear the witness's statements during the identification process. See Byrd v. State, 25 A.3d 761, 766 (Del.2011); United States v. Domina, 784 F.2d 1361, 1368 (9th Cir.1986) (stating that, when a witness identifies a defendant at trial, that "testimony has generally been held admissible unless tainted by the prior suggestive identification process"). In addition to affording the factfinder an opportunity to observe and evaluate the identification itself, a first-time in-court identification is subject to immediate challenge through cross-examination. "Where a witness first identifies the defendant at trial, defense counsel may test the perceptions, memory and bias of the witness, contemporaneously exposing weaknesses and adding perspective in order to lessen the hazards of undue weight or mistake." People v. Rodriguez, 134 Ill.App.3d 582, 89 Ill.Dec. 404, 480 N.E.2d 1147, 1151 (1985), cert. den, 475 U.S. 1089, 106 S.Ct. 1476, 89 L.Ed.2d 731 (1986). In Domina, the Ninth Circuit noted that, although there "can be little doubt that the initial in-court identification is suggestive, * * * procedures could be used to lessen the suggestiveness." 784 F.2d at 1368-69. The court explained that,

Id. (Internal quotations and citations omitted).

The foregoing decisions reflect the mainstream of jurisprudence addressing the admissibility of first-time in-court eyewitness identification evidence. The principles on which those decisions rest are embedded in the profound respect that our system of justice holds for the role of juries in the adjudicative process. It is true that, as additional studies have examined the fallibility and tendency to decay of human memory of events, a widespread public understanding of those issues has not kept pace. See Lawson/James, 352 Or. at 760 n. 10, 769-79, 291 P.3d 673 (discussing in detail such research). Because of that gap in common understanding, over time, courts increasingly have permitted the use of safeguards such as expert testimony to help reduce the risk of jury "overvaluation" of eyewitness identification testimony. See, e.g., State v. Guilbert, 306 Conn. 218, 49 A.3d 705 (2012). In Lawson/James, this court took judicial notice of "the data contained in those various sources as legislative facts that we may consult for assistance in determining the effectiveness of our existing test for the admission of eyewitness identification evidence." Lawson/James, 352 Or. at 739-40, 291 P.3d 673. In doing so, however, the court took care to state:

Id. at 741, 291 P.3d 673.

Based on the research of which it took judicial notice, this court in Lawson/James ultimately concluded that, where suggestive pretrial police procedures are asserted to have tainted eyewitness memory, the existing test for determining the admissibility of the witness's identification testimony was inadequate for purposes of applying OEC 403. As an illustration of such a suggestive procedure, the court explained:

Id. at 742-43, 291 P.3d 673. Based in part on that concern, the court said:

Id. at 758, 291 P.3d 673 (emphases added).

Based on the passages quoted above and other statements in Lawson/James, the Court of Appeals in this case appears to have concluded that the research sources on which this court relied in that case — including the research pertaining to so-called "estimator variables" and "system variables" — had been incorporated into a general interpretive overlay for the application of OEC 403 to eyewitness identification testimony. Hickman, 255 Or.App. at 696-97, 298 P.3d 619. Consistently with that understanding, in applying the Lawson/James analysis, the Court of Appeals in effect treated the courtroom trial setting in this case as a "system variable" much like a police-administered showup that unfairly amplified the suggestiveness of D's and N's in-court identifications of defendant as the perpetrator of the crime. See Hickman, 255 Or.App. at 698, 298 P.3d 619 ("Most significantly, the procedure was similar to, but significantly more suggestive than, a `showup,' which is `inherently suggestive' because the witness is always aware of whom police officers have targeted as a suspect.").

The Court of Appeals' treatment of this case was understandable in light of this court's reliance in Lawson/James on the research of which it took judicial notice in determining that the Classen test was inadequate. However, in reviewing the Court of Appeals' conclusion, two cautionary factors merit emphasis. First, as discussed, Lawson/James was a case about "a discrete evidentiary class," eyewitness identifications subjected to suggestive out-of-court police procedures, which this court deemed to be "particularly susceptible to concerns of unfair prejudice." Lawson/James, 352 Or. at 758, 291 P.3d 673. This court in Lawson/James did not intimate — let alone hold — that admission of a first time in-court eyewitness identification of a defendant that is untainted by suggestive pretrial state-administered procedures is "unfairly prejudicial" under OEC 403 merely because it occurs in a courtroom setting where the identity of the accused is apparent to the witness.

Second, in taking judicial notice as "legislative facts" of the research that it considered, this court in Lawson/James relied on its earlier decision in State v. O'Key, 321 Or. 285, 309 n. 35, 899 P.2d 663 (1995), for the proposition that "[t]he validity of proffered scientific evidence * * * is a question of law" to be determined by judicial notice of legislative facts submitted to the court. Lawson/James, 352 Or. at 740, 291 P.3d 673 (quoting O'Key). In O'Key, however, unlike in Lawson/James and this case, the issue before the court was the admissibility of the scientific evidence itself, not whether and to what extent scientific research would be used to determine the admissibility of percipient lay testimony. In that respect, this court went further in Lawson/James in taking judicial notice of scientific data as so-called "legislative facts" than it previously had done in O'Key. In particular, although the court cautioned that its "acknowledgment of the existence of that research in these cases is not intended to preclude any party in a specific case from validating scientific acceptance of further research or from challenging particular aspects of the research described in this opinion," id. at 741, 291 P.3d 673, it nevertheless proceeded to use that research to disavow the Classen test and to remand the case for the purpose of determining the admissibility of eyewitness testimony, at least where suggestive pretrial police procedures may have tainted witness memory.8

In the absence of appropriate limitations, that action could have far-reaching and, perhaps, unforeseen implications. After all, decision-making biases affect all people alike, including juries, advocates, social scientists, and, we daresay, judges acting as evidentiary gatekeepers.9 Assume, for example, that some justice system observers believe that factfinders tend to overvalue the testimony of law enforcement officials, spiritual leaders, school teachers, and other categories of witnesses who, by demeanor or resume, may be perceived to possess an exceptional gravitas or aura of credibility. If social science studies were to persuasively support that belief and show that, despite appropriate admonitions, it is unduly difficult for juries to overcome such biases even where those witnesses give erroneous testimony, what do we do with that information? Will the trial judge then have a heightened screening role to perform under OEC 403 with respect to the admissibility of that evidence as well?

To avoid excessive encroachment on the factfinding role of juries in the adjudicative process, legislatures and courts have set important limits on the admission of expert opinion evidence. Thus, an expert may be permitted to give opinion testimony based on scientific data, theories, and rules, but the law — both historically and modernly — has resisted expert attempts to tell a trier of fact who, in particular, is telling the truth or giving accurate testimony, and who is not. See, e.g., State v. Lupoli, 348 Or. 346, 357, 234 P.3d 117 (2010) ("This court has long held that one witness may not give an opinion on whether he or she believes another witness is telling the truth."); ORS 44.370 ("Where the trial is by the jury, they are the exclusive judges of the credibility of the witness."); ORS 136.320 ("questions of fact" in criminal cases are generally for the jury to decide). When courts decide whether a percipient witness may testify before a jury based on the views of expert witnesses about the reliability of such testimony and the capacity of jurors to accurately assess its reliability, they embark on a course that requires an acute awareness of the fundamental assumptions on which our system of justice is delicately balanced.10

On the other hand, the risk of misidentification stemming from an in-court identification is elevated by the inherently suggestive circumstances of a procedure that, like the one that occurred here, can be analogized to a suspect show-up.11 And, of course, once an in-court identification occurs, it may be too late to construct and administer a non-prejudicial identification procedure. The dilemma is how to mitigate the risk of error in factfinding and still honor our fundamental belief that the trusted rule of law depends on broad citizen participation in the adjudicative process. To at least a degree, OEC 403 exists as a means to strike the balance between those competing values. We therefore turn to the application of OEC 403 to the circumstances of this case.

We begin with the pretrial statements and trial testimony of N. Unlike D, N described the shooter in some detail, including his age, race, build, and apparel, in her initial police interview. Although N's trial testimony was not wholly consistent with D's testimony and that of other witnesses, and N's testimony was more detailed than her initial statement, those factors do not demonstrate that N's testimony was unreliable as a matter of law. Witnesses can, and often do, have different memories of fast-moving events and, depending on the questions asked, may give more or less detailed information over the course of multiple interviews.

In addition, there was no evidence of any suggestive preidentification procedure that could have tainted N's in-court identification of defendant, nor did any portion of that identification occur outside the presence of the jury. In Lawson/James, this court indicated that suggestive pretrial police identification procedures, including "suggestive questioning, cowitness contamination, and other sources of post-event memory contamination," can raise particular concerns with respect to the source and reliability of subsequent recounted eyewitness memory:

352 Or. at 743, 291 P.3d 673. Those factors generally are either absent where, as here, a first-time identification of a perpetrator occurs in a courtroom setting, or, if they arise, are subject to court supervision and jury assessment.

It also bears emphasis that N's pretrial description of the perpetrator, which defendant does not contend was unavailable to him through discovery, was sufficiently particular to put defendant on notice that N might be asked to make an in-court identification. Courts considering the admissibility of first-time in-court identifications generally have placed the burden of seeking a prophylactic remedy on the defendant. See, e.g., United States v. Brown, 699 F.2d 585, 594 (2d Cir. 1983) ("[W]hen a defendant is sufficiently aware in advance that identification testimony will be presented at trial and fears irreparable suggestivity, as was the case here, his remedy is to move for a line-up order to assure that the identification witness will first view the suspect with others of like description rather than in the courtroom sitting alone at the defense table"); Domina, 784 F.2d at 1369 (concluding that "procedures could be used in court to lessen the suggestiveness [of first time in-court identifications], such as an in-court line-up, or having the defendant sit somewhere in the courtroom other than the defense table"). That allocation is consistent with the burden that this court has imposed on an opponent of eyewitness identification testimony under OEC 403 to show that its admission would be unfairly prejudicial. Lawson/James, 352 Or. at 762, 291 P.3d 673. Defendant sought none of those precautionary procedures here.12

In short, insofar as N's identification testimony was concerned, no suggestiveness was in play beyond that inherent in a normal courtroom setting, and defendant did not show that other factors tipped the balance under OEC 403 in favor of exclusion. Under those circumstances, application of the Lawson/James analysis would not have resulted in the exclusion of N's testimony. Id. at 765-68, 291 P.3d 673.

D's identification testimony presents a different calculus, both in terms of its lower probative value and potentially greater prejudicial effect. Minutes after the shooting, D told police that she "didn't see the shooting and couldn't describe much." When she met with a defense investigator about two weeks before trial, D told the investigator that she could describe the men in the altercation only as "big black men." She further told the investigator that the shooter had a "big Afro," but gave no further details about his hair. The next day, in an unrecorded interview with the prosecutor, D stated that the shooter had "twisties" with "close black hair." D also told the prosecutor that she was not certain that she could identify the perpetrator.

When she first took the stand at trial, D was aware that the man sitting between his two lawyers at counsel table was charged with Monette's murder. And, as noted, although there were a half dozen or more African-American men in the gallery of the courtroom, defendant was the only African-American in the well of the courtroom. When defendant objected to the prosecutor's question whether D could identify the perpetrator in the courtroom, the court excused the jury and considered the objection. During the course of the ensuing colloquy between the court and counsel, the prosecutor represented to the court that, when he met with D before trial, and she told him that she did not know whether she could identify the perpetrator, the prosecutor had proposed that, at trial, D should signal him with a "look in the eye" if she recognized the shooter when she took the stand. The prosecutor represented that he told D, "[i]f you do [recognize the perpetrator], then let the Court know — let the trier of fact know. If you don't, then you don't."

The prosecutor told the court that, in the brief period of questioning that took place before the court recessed due to a recording equipment malfunction, D had not signaled to the prosecutor that she recognized the perpetrator in the courtroom. The prosecutor further stated that, after the jury was excused, D walked past defendant and that she began hyperventilating when, at that moment, she recognized him as the perpetrator.

Defense counsel responded that none of D's emotional reaction was observed by the jury and that the process leading to D's identification of defendant was unfairly suggestive because "the courtroom [was] cleared during the break, so that there is one lone black man seated between several white people before she walks by him yet again, and that's when she has this reaction." Counsel then argued that the court should exclude D's testimony because the identification occurred outside the presence of the jury under unfairly suggestive circumstances.

After further discussion, the trial court ultimately overruled defendant's objection. The court stated that, because the malfunction of the court's recording equipment was an unforeseen event, it did not believe that the prosecutor had "set this up to make it suggestive." The court also stated that it was not going to rule that an in-court identification is inadmissible on the ground of unfair suggestiveness merely because "there is only one obvious defendant in the courtroom and other suspects are not lined up in the courtroom." The court concluded that, although there is an inherent aspect of suggestiveness to any in-court identification, the process leading to D's identification of defendant was not unfairly prejudicial so as to require the exclusion of her testimony. In its ruling, the court noted that the defense had an opportunity to engage in "powerful cross examination," and the court further invited defendant to call an expert witness on the reliability of eyewitness identification, which defendant did.

In cross-examining D after she identified defendant as the perpetrator, defense counsel did not ask any questions about events that had occurred during the recess. In redirect examination, however, D testified without objection that she had identified defendant as the perpetrator when she walked by him during the recess, that the prosecutor had then walked with D to the lobby where she sat with her mother, and that nobody else had since talked to her about that episode. Defense counsel did not inquire about the subject on re-cross-examination.

On review, defendant argues:

According to defendant,

Finally, defendant urges,

The force of those arguments is not insubstantial. Here, no suggestive police procedures preceded D's in-court identification of defendant. However, when, before trial, the prosecutor told D to give him a look if she recognized the perpetrator in the courtroom, two pertinent suggestions were made: First, that there was a distinct possibility that the perpetrator would be present in the courtroom; second, that the person that the state had charged with the crime was the African-American man seated at counsel table. Under the circumstances, the suggestion of defendant's identity as the perpetrator was substantial. See United States v. Rogers, 126 F.3d 655, 659 (5th Cir.1997) ("Even the best intentioned among us cannot be sure that our recollection is not influenced by the fact that we are looking at a person we know the Government has charged with a crime."). The jury was not present when D's initial identification of defendant and her ensuing emotional response occurred, and, therefore, it was unable to evaluate the significance of those events.

After D returned to the stand following the recess, she was able for the first time to give a detailed description of the perpetrator. D described the perpetrator as being in his 20s to early 30s, stocky, tall 5'7". to 6'), as having a close Afro or braids, and as having certain facial characteristics. That description closely matched defendant's appearance at trial, except that the general height description ("tall") was inaccurate. That sudden "improvement" in D's recollection of detail — in light of the other described circumstances — permitted an inference that her in-court identification of defendant may have been influenced by the suggestiveness of the courtroom setting.

In addition, where, as here, defendant had no reason to expect that D would be asked to make an in-court identification (after all, D had never told anyone before trial that she could identify the shooter nor had she given any pretrial description that would indicate that she could), defense counsel had little reason to make a precautionary request for pretrial or in-trial steps to test D's recollection with a fairly constructed and administered identification procedure.

On the other side of the scale, defendant had an opportunity — but did not avail himself of it — to cross-examine D about her "signal" arrangement with the prosecutor, her failure to give a signal before the recess, and her identification of D under more suggestive circumstances when she walked past him in the courtroom during the recess. However, the evidence permitted an inference that the signal arrangement — unlike many suggestive pretrial police procedures — did not result in a positive identification. Instead, D did not profess to identify D until she walked past him at counsel table. That sequence of events raised a cross-cutting inference that the ultimate identification was not feigned or the product of suggestion but, rather, was authentic. The risk of such an adverse inference may, in part, explain why defense counsel did not bring the matter to the jury's attention. Furthermore, the identification that D made outside the jury's presence was not the result of a state-designed or engineered process. As the trial court explained, it occurred, albeit unexpectedly, during the course of the regular trial process.13 Finally, the inherent suggestiveness of a trial setting may be prejudicial in a general sense, but, as the trial court properly observed, that does not necessarily make a first time in-court identification unfairly prejudicial in the sense required for exclusion under OEC 403.

In sum — although they cut both ways — the foregoing array of circumstances renders D's in-court identification of defendant more troubling under OEC 403 than N's identification. Those circumstances also demonstrate that practitioners are well advised, before embarking on such a venture, to contemplate the risks associated with adducing first time in-court eyewitness identifications, so as to mitigate the prospect of unnecessary and unfair prejudice. However, we need not decide whether the trial court erred in admitting D's eyewitness testimony because we conclude that the error, if any, was harmless in light of N's identification testimony.

As discussed, N's version of events was more detailed than D's from the outset and was more consistent over time and more accurate; in addition, N did not purport to spontaneously recognize defendant outside the presence of the jury. Her identification of defendant occurred entirely in the jury's presence. Finally, there was no evidence of a pretrial signal arrangement between N and the prosecutor; thus, unlike with D, there were no suggestive pretrial or in-trial procedures at play with regard to N's identification that the jury was unable to observe and assess for itself.

Moreover, two other eyewitnesses (although subject to impeachment for bias) positively identified defendant as the perpetrator, three more eyewitnesses described the perpetrator as having a similar physical appearance to defendant (although they were unable to make a more positive identification because the shooter wore a ski mask), and defendant's DNA was prominently found on the ski mask that the perpetrator wore during the shooting. In those circumstances, there is little likelihood that D's relatively weaker identification testimony affected the jury's verdict. See State v. Davis, 336 Or. 19, 32, 77 P.3d 1111 (2003) ("Oregon's constitutional test for affirmance despite error consists of a single inquiry: Is there little likelihood that the particular error affected the verdict?"); see also State v. Camarena, 344 Or. 28, 41-42, 176 P.3d 380 (2008) (improper admission of complainant's testimonial statements in violation of confrontation requirements was harmless error where such statements were cumulative of properly admitted nontestimonial statements sufficient to establish defendant's guilt). Accordingly, any error in admitting D's identification testimony under OEC 403 was harmless.

III. DUE PROCESS

On review, defendant primarily relies on this court's evidentiary code analysis in Lawson/James. Defendant does not focus in detail on the due process arguments that he advanced before the trial court and the Court of Appeals. However, to complete our analysis, we briefly consider them here.

The United States Supreme Court has not extended constitutional protections to in-court identifications that are untainted by a prior identification resulting from unduly suggestive procedures. Domina, 784 F.2d at 1369 (noting that "[t]here is no constitutional entitlement to an in-court line-up or other particular methods of lessening the suggestiveness of in-court identification, such as seating the defendant elsewhere in the room."); see also Byrd, 25 A3d at 767 ("Based upon the different considerations involved in pretrial and in-court identifications, we join the majority of courts in concluding that [due process concerns do] not apply to in-court identifications [and] * * * that the remedy for any alleged suggestiveness of an in-court identification is cross-examination and argument."); State v. King, 156 N.H. 371, 934 A.2d 556 (2007) (same); State v. Lewis, 363 S.C. 37, 609 S.E.2d 515, 518 (2005) ("[Due process concerns do] not apply to a first-time in-court identification because the judge is present and can adequately address relevant problems; the jury is physically present to witness the identification, rather than merely hearing testimony about it; and cross-examination offers defendants an adequate safeguard or remedy against suggestive [identifications].").

Further, the Supreme Court has recently made clear that due process rights of defendants identified in the courtroom under suggestive circumstances are generally met through the ordinary protections in trial. Perry v. New Hampshire, ___ U.S. ___, ___, 132 S.Ct. 716, 728-29, 181 L.Ed.2d 694 (2012). Those protections include the right to confront witnesses; the right to representation of counsel, who may expose flaws in identification testimony on cross-examination and closing argument; the right to jury instructions advising use of care in appraising identification testimony; and the requirement of proof beyond a reasonable doubt. Id.; see also United States v. Thompson, 524 F.3d 1126, 1136 (10th Cir.2008) (holding in-court identification procedure not unconstitutionally suggestive where robber was an African-American male and defendant was the only African-American male in the courtroom); United States v. Davis, 103 F.3d 660, 670 (8th Cir.1996) (same).

In light of those authorities, we cannot hold that the in-court identification procedure complained of was so impermissibly suggestive as to violate defendant's due process rights.

The decision of the Court of Appeals is reversed. The judgment of the circuit court is affirmed.