People v. Lerma

Annotate this Case

In 2008 Gill was shot to death while sitting outside his Chicago home with Clark. The evidence of defendant’s guilt consisted solely of two eyewitness identifications. The first identification, made by the victim, was admitted into evidence under the excited utterance exception to the hearsay rule. Before Gill died, Gill’s father (Johnson) asked Gill who had shot him. Both Clark and Johnson testified that Gill responded that “Lucky” had shot him. Multiple witnesses testified that defendant, who lived across the street from Gill, was known by the nickname “Lucky.” Gill’s mother testified that Gill and defendant had been friends for years, that defendant often spent time in Gill’s home, and that defendant recently had been fighting with a member of Gill’s family. From photographs and during an in-person show-up, Clark identified defendant as the shooter. Clark admitted she had seen defendant only “[l]ike once or twice” before the shooting, had not spoken to him, and did not know him. Clark apparently contradicted herself about whether defendant wore a hood. Defendant filed a motion in limine to allow an attorney/licensed psychologist, to testify as an expert on the topic of memory and eyewitness identification. The Illinois Supreme Court reversed denial of the motion, finding that the requested testimony was relevant and the error was not harmless.

Download PDF

2016 IL 118496 IN THE SUPREME COURT OF THE STATE OF ILLINOIS (Docket No. 118496) THE PEOPLE OF THE STATE OF ILLINOIS, Appellant, v. EDUARDO LERMA, Appellee. Opinion filed January 22, 2016. JUSTICE THOMAS delivered the judgment of the court, with opinion. Chief Justice Garman and Justices Freeman, Kilbride, Karmeier, Burke, and Theis concurred in the judgment and opinion. OPINION ¶1 The issue is whether, in light of the specific facts and circumstances of this case, the circuit court of Cook County abused its discretion when it denied defendant’s motion to allow expert testimony concerning the reliability of eyewitness identifications. For the reasons that follow, we hold that it did. ¶2 BACKGROUND ¶3 This case arises from the murder of Jason Gill, who in May 2008 was shot to death while sitting on the front steps of his Chicago home. The history of this case is set forth fully in the appellate court’s opinion below, and we need not repeat the entirety of that history here. Instead, we will set forth only those facts pertinent to the issue presently before the court. ¶4 The Eyewitness Identifications ¶5 The evidence of defendant’s guilt consists solely of two eyewitness identifications. The first eyewitness identification was made by the victim, Jason Gill, and was admitted into evidence under the excited utterance exception to the hearsay rule. 1 The evidence at trial established that, on the night of the shooting, Gill was sitting on the front steps of his home with a friend, Lydia Clark. At approximately 11:20 p.m., a gunman approached the porch and opened fire at Gill and Clark. Gill was struck several times, and Clark dragged Gill into the house. Inside the house, Gill’s father, Bill Johnson, asked Gill who had shot him. Both Clark and Johnson testified that Gill responded that “Lucky” had shot him. Multiple witnesses, including both Clark and Gill’s mother, testified that defendant, who lived across the street from Gill, was commonly known by the nickname “Lucky.” In addition, Gill’s mother testified that Gill and defendant had been friends for several years, that defendant often spent time in Gill’s home, and that defendant recently had been fighting with a member of Gill’s family. ¶6 The other eyewitness identification was made by Clark. Clark testified that, on the evening of the shooting, she was with Gill on the unlit front steps of Gill’s Chicago home. At approximately 11:20 p.m., a man dressed all in black approached Gill’s house, pulled a gun, and began shooting at Gill and Clark. Though the man was wearing a hooded sweatshirt, Clark testified that the hood was down at the time of the shooting. 2 Gill covered Clark’s body with his, and the two of them fell to the ground together. When the shooting stopped, Clark saw that Gill had been shot several times. Clark dragged Gill into the house, where she called 911. By this point, Bill Johnson had come downstairs and was asking Gill what had 1 The excited utterance exception allows the substantive admission of an otherwise inadmissible hearsay statement where the proponent of that statement is able to demonstrate (1) the occurrence of an event or condition sufficiently startling to produce a spontaneous and unreflecting statement; (2) absence of time to fabricate; and (3) a statement relating to the circumstances of the occurrence. People v. Smith, 152 Ill. 2d 229, 258 (1992). Statements admitted under this exception are admitted because such statements “tend to be reliable.” People v. Nevitt, 135 Ill. 2d 423, 442 (1990). 2 Detective Michael Hughes, who interviewed Clark after the shooting, testified that Clark reported on the night of the shooting that the assailant’s hood was up. -2- happened. Clark heard Gill say, “Lucky shot me.” When the police arrived at the scene, Clark told them that “Lucky” had shot Gill. The next morning, at 1:25 a.m., Clark went to the police station, where she was shown a photo lineup of six Hispanic males, one of whom was defendant. From the photographs, Clark identified defendant as the shooter. One day later, in a one-person show-up, Clark again identified defendant as the shooter. In open court, and no less than five times, Clark pointed to and specifically identified defendant as the man who shot Gill. Clark testified that she knew defendant only by his nickname, “Lucky.” On direct examination, Clark testified that, in the six months to a year before the shooting, she had seen defendant on the porch across the street from Gill’s house approximately ten times. On cross-examination, however, Clark admitted that, in her grand jury testimony, she testified that she had seen defendant only “[l]ike once or twice” before the shooting. Either way, Clark had never talked to or had a conversation with defendant, and she had never been in the same room or the same house with defendant. When asked directly how long she had known defendant prior to the shooting, Clark responded, “I did not know him.” ¶7 Defendant’s Motion in Limine ¶8 In anticipation of these eyewitness identifications, defendant filed a pretrial motion in limine to allow Dr. Solomon Fulero, an attorney and licensed psychologist, to testify as an expert on the topic of memory and eyewitness identification. Defendant argued that Dr. Fulero’s testimony would aid the jury by identifying and explaining several “common misperceptions” that exist concerning the accuracy and reliability of eyewitness identifications. According to defendant’s motion, which included a report authored by Dr. Fulero, Dr. Fulero’s testimony would include the following scientifically documented findings, all of which are beyond the common knowledge of the average layperson: that the witness’s level of confidence does not necessarily correlate to the accuracy of the identification; that numerous factors can undermine the accuracy of an eyewitness’s identification, including the stress of the event itself, the presence of a weapon, the passage of time, the “forgetting curve,” the wearing of partial disguises such as hoods, exposure to postevent information, nighttime viewing, and suggestive police identification procedures; that eyewitnesses tend to overestimate time frames; and that cross-racial identifications tend to be less reliable than same-race identifications. -3- ¶9 The State opposed defendant’s motion on three principle grounds. First, citing this court’s decision in People v. Enis, 139 Ill. 2d 264 (1990) and the First District’s decision in People v. Tisdel, 316 Ill. App. 3d 1143 (2000), the State argued that “Illinois courts have consistently upheld a trial court’s decision to bar expert testimony regarding witness identification.” Second, the State argued that the matters about which Dr. Fulero intended to testify are well within the common knowledge of the average layperson and therefore could be addressed adequately through cross-examination, closing arguments, and jury instructions. Third, the State argued that the data and conclusions contained in Dr. Fulero’s report “do not fit the facts of this case” because they deal solely with stranger identifications, whereas in this case both Gill and Clark knew defendant prior to the crime. To bolster this point, the State presented the trial court with an unpublished decision from the Ohio Court of Appeals containing a summary description of Dr. Fulero’s testimony from a 1999 murder trial. See State v. Nickleberry, No. 77516, 2000 WL 1738356, at *3 (Ohio Ct. App. Nov. 22, 2000). According to that summary, while Dr. Fulero “testified that eyewitness identification may be unreliable because of a variety of factors,” he also “admitted *** that these factors are considered applicable where the eyewitness is viewing a stranger and not someone he or she has met before.” Id. In addition, the State presented the trial court with an article that it “pulled off the internet yesterday,” in which a “Mr. Mark Green who has a Ph.D. in psychology” wrote that “[t]here are some situations where identification is more likely accurate *** [f]or example, if the suspect is someone previously known to the victim, then high accuracy is more probable.” ¶ 10 Following oral arguments, the trial court denied defendant’s motion to allow Dr. Fulero’s expert witness testimony on the reliability of eyewitness identifications. In so ruling, the trial court emphasized that the one fact that distinguishes this case from other identification cases is that the eyewitnesses here knew defendant prior to the shooting. According to the trial court: “[I]t is not a circumstance that requires the testimony of an expert to establish what pretty much everybody knows, which *** it is a fact that persons *** are less likely to misidentify someone they have met or know or seen before than a stranger. That’s not a function of psychology or expert opinion testimony. It is a function of human nature, and it is not something that would require the application of expert opinion testimony because it is not beyond the ken of an ordinary juror.” -4- In addition, the trial court reasoned that, because the eyewitnesses in this case knew defendant prior to the crime, “the claims Dr. Fulero speaks to regarding cross-racial difficulties, certainty or confidence, stress, weapons focus, multiple witness identifications, suggestibility, they are necessarily made less relevant than they would be if these persons had never seen the shooter before the night in question.” The trial court then explained that this decrease in relevance “leads to a concomitant increase of potential prejudice at least in the relative sense.” More specifically, the trial court explained that, given its relative lack of relevance, Dr. Fulero’s testimony ran the risk of both “generat[ing] *** a referendum on the efficacy of identification testimony generally” and “operating as his opinion on the credibility” of the eyewitnesses. ¶ 11 Defendant’s First Motion to Reconsider ¶ 12 One month later, and before his trial began, defendant filed a brief motion to reconsider the denial of his motion in limine. In support, defendant cited and attached the New Jersey Supreme Court’s then-recent decision in State v. Henderson, 27 A.3d 872 (N.J. 2011), which comprehensively reviewed the current state of scientific research concerning the reliability of eyewitness identifications. In addition, at the hearing on the motion to reconsider, defense counsel informed the trial court that, if the motion was allowed, Dr. Fulero would “testify that misidentifications have occurred with people who the witness knew beforehand.” At the conclusion of the hearing, the trial court denied defendant’s motion to reconsider. In doing so, the trial court began by observing that “in the years that I have been doing this in one capacity or another there is always an issue or two du jour” and that “this seems to be one that is coming on strong.” From there, the trial court explained that it was denying the motion to reconsider “for the reasons previously expressed at some considerable length,” the “glaring one” of which is that “the persons who identify Mr. Lerma *** all claim to have known him.” According to the trial court, “[t]hat factor makes this [case] sufficiently distinct from those cases where identification procedures are of special moment and subject to the dangers which arise when strangers are called upon to identify persons [who are] also strangers.” The trial court then pointed out that its decision was -5- “supported by the defense’s own witness, Dr. Fulero,” who is reported to have offered a similar opinion in Nickleberry. 3 ¶ 13 Defendant’s Second Motion to Reconsider ¶ 14 Midway through trial, and after the State had presented the eyewitness testimony set forth above, defense counsel renewed defendant’s motion to call an eyewitness identification expert. Because Dr. Fulero had since passed away, defense counsel this time tendered to the trial court a report authored by Dr. Geoffrey Loftus, a professor of psychology at the University of Washington and widely-published and globally-recognized expert in the field of human perception and memory. The data and conclusions contained in Dr. Loftus’s report largely tracked with the contents of Dr. Fulero’s report, with two significant exceptions. First, Dr. Loftus’s report stressed that he would not “issue judgments about whether a particular witness’s memory and assertions *** are correct or incorrect” and that “any testimony on [his] part which implies unreliability on the part of eyewitness(es) who identify a defendant should not, ipso facto, be taken to imply that the defendant is innocent—it implies only that the eyewitness evidence should be viewed with appropriate caution.” Second, and more importantly, unlike Dr. Fulero’s report, which was silent on the subject of acquaintance identifications, Dr. Loftus’s report specifically stated that “[i]t would seem intuitive to a jury that if a witness identifies a suspect with whom he or she is acquainted, the witness’s identification would likely be accurate. However, this is not necessarily true.” Rather, the report explained, “if circumstances are poor for a witness’s ability to perceive a person,” and “the situation fosters a witness’s expectations that he or she will see a particular acquaintance[,] *** then the witness will tend to perceive the person as the expected acquaintance even if the person is in fact someone else.” According to Dr. Loftus’s report, such poor circumstances include low lighting; viewing longer distances in the dark; divided attention of the witness, including a focus on a weapon; time duration, with less time leading to less available information, and a witness’s tendency to overestimate time durations; cross-racial 3 In making this observation, the trial court acknowledged that Dr. Fulero in fact disputed the accuracy of Nickleberry’s description of his testimony in that case. Nevertheless, the trial court stated that “when an appellate court justice acting on the record in an Ohio appeals court case claims that [Dr. Fulero] did in fact [state that opinion], I am not going any further down that road.” -6- identification; stress; and a partially obscured face. Dr. Loftus stated such situations may lead to misidentification because: “In such circumstances, the witness’s acquaintance with the expected—and hence perceived—person works against accurate identification for two reasons: First, it would be natural and easy for the witness to subsequently pick the acquaintance in an identification procedure *** (because the witness already knows whom she is seeking in a lineup procedure, she could immediately rule out all the fillers, and zero in on the acquaintance/suspect). Second, the witness could use his or her prior knowledge of the acquaintance’s appearance to reconstruct his or her memory of the original events—the crime—such that the in fact poor original memory of the actual criminal is replaced with a stronger and more confidence-evoking memory of the acquaintance ***.” ¶ 15 In response to defendant’s renewed motion, the State argued simply that “[n]othing has changed from the previous situation when you heard these motions.” Rather, the State argued: “[Clark] still knows the defendant, still always maintained she knew him, and as we discussed before these studies and these issues and expert testimony on identification issue has primarily focused on individuals who did not know the person that they were identifying prior to the incident that occurred. Nothing with regard to that has changed.” ¶ 16 After hearing from both sides, the trial court stated that it was denying defendant’s motion to present Dr. Loftus’s testimony “consistent with the reasons [the court] set forth in detail when [the court] made the ruling on your similar motion with respect to Dr. Fulero.” ¶ 17 Conviction and Sentence ¶ 18 In the end, the jury convicted defendant of first degree murder, personally discharging the firearm that caused Jason Gill’s death, and aggravated discharge of a weapon. The trial court sentenced defendant to 45 years in prison. -7- ¶ 19 Defendant’s Appeal ¶ 20 Defendant appealed, and the appellate court reversed and remanded. 2014 IL App (1st) 121880. In doing so, the appellate court reasoned that, under this court’s decision in People v. Enis, 139 Ill. 2d 264 (1990), when deciding whether to admit expert testimony on the subject of eyewitness identification, a trial court is required to “carefully consider the necessity and relevance of the expert testimony in light of the facts in the case before him.” Id. at 290; see also People v. Tisdel, 338 Ill. App. 3d 465, 468 (2003) (holding that “[t]rial courts should carefully scrutinize the proffered testimony to determine its relevance—that is, whether there is a logical connection between the testimony and the facts of the case”). According to the appellate court, the trial court here clearly failed to do this, as the reasons the trial court gave for denying the admission of Dr. Fulero’s testimony (i.e., that “it is a fact” that acquaintance identifications are reliable and that expert testimony in this case would be “operating as *** opinion on the credibility” of the eyewitnesses) could not possibly serve as a basis for denying the admission of Dr. Loftus’s testimony, as Dr. Loftus’s report directly refutes both of these points. 2014 IL App (1st) 121880, ¶¶ 36, 38. Nevertheless, in denying the admission of Dr. Loftus’s testimony, the trial court stated that its ruling was for reasons “consistent with the reasons” it gave for denying the admission of Dr. Fulero’s testimony. According to the appellate court, this demonstrates that “the trial court *** did not carefully consider or scrutinize Dr. Loftus’s report where [that] report directly contradicted the court’s prior finding that it is common knowledge that an eyewitness is less likely to misidentify an acquaintance.” Id. ¶ 36. Moreover, the appellate court explained, the trial court’s initial ruling denying the admission of Dr. Fulero’s testimony was itself an abuse of discretion because the reasons the trial court gave for that ruling amounted to “little more than a series of conclusions based on its personal belief” that acquaintance identifications are accurate and therefore not a proper subject for expert testimony. Id. ¶ 38. Thus, when it later invoked those same reasons as the basis for its decision denying the admission of Dr. Loftus’s testimony, the trial court was not only invoking reasons that were now factually inapposite but also reasserting its own “subjective value judgments” as the basis for rejecting the considered opinion of a qualified expert. Id. Accordingly, the appellate court reversed the trial court’s ruling denying the admission of Dr. Loftus’s testimony and remanded the cause for a new trial that includes the admission of expert testimony on the matter of eyewitness identification. Id. ¶ 40. -8- ¶ 21 The State appealed to this court, and we allowed its petition for leave to appeal. Ill. S. Ct. R. 315 (eff. July 1, 2013). ¶ 22 ANALYSIS ¶ 23 A criminal defendant’s right to due process and a fundamentally fair trial includes the right to present witnesses on his or her own behalf. People v. Wheeler, 151 Ill. 2d 298, 305 (1992). “In Illinois, generally, an individual will be permitted to testify as an expert if his experience and qualifications afford him knowledge which is not common to lay persons and where such testimony will aid the trier of fact in reaching its conclusion.” Enis, 139 Ill. 2d at 288. In addressing the admission of expert testimony, the trial court should balance the probative value of the evidence against its prejudicial effect to determine the reliability of the testimony. Id. at 290. In addition, in the exercise of its discretion, the trial court should carefully consider the necessity and relevance of the expert testimony in light of the particular facts of the case before admitting that testimony for the jury’s consideration. Id. This court has held that expert testimony is only necessary when the subject is both particularly within the witness’s experience and qualifications and beyond that of the average juror’s, and when it will aid the jury in reaching its conclusion. People v. Cloutier, 156 Ill. 2d 483, 501 (1993). Expert testimony addressing matters of common knowledge is not admissible “unless the subject is difficult to understand and explain.” People v. Becker, 239 Ill. 2d 215, 235 (2010). When determining the reliability of an expert witness, a trial court is given broad discretion. Enis, 139 Ill. 2d at 290. Therefore, we review the trial court’s decision to admit evidence, including expert witness testimony, for an abuse of that discretion. Becker, 239 Ill. 2d at 234. An abuse of discretion occurs only where the trial court’s decision is “arbitrary, fanciful, or unreasonable to the degree that no reasonable person would agree with it.” People v. Rivera, 2013 IL 112467, ¶ 37. ¶ 24 The issue in this case is whether the trial court abused its discretion in denying defendant’s request to allow Dr. Loftus’s expert testimony on the reliability of eyewitness identifications. Before addressing that question directly, we would like briefly to comment on the current state of jurisprudence concerning the admission of eyewitness expert testimony, which contrary to the trial court’s belief, cannot possibly be dismissed as a mere “issue du jour.” The last time this court addressed the admission of such testimony was in Enis, which was decided more than 25 -9- years ago when the relevant research was in its relative infancy. Even then, this court recognized that “in the past decade a number of courts have held that expert testimony concerning eyewitness identification should be admissible in certain circumstances.” Enis, 139 Ill. 2d at 286-87 (collecting cases). Nevertheless, this court also expressed skepticism and caution against the overuse of such testimony (id. at 289), such that the exclusion of such testimony remains the common practice in Illinois to this day. See, e.g., People v. McGhee, 2012 IL App (1st) 093404, ¶ 55 (observing that “Illinois continues to reject, at least in practice, expert testimony on the reliability of eyewitnesses”). The decades since Enis, however, have seen a dramatic shift in the legal landscape, as expert testimony concerning the reliability of eyewitness testimony has moved from novel and uncertain to settled and widely accepted. Indeed, as the Supreme Court of Pennsylvania recently noted, there is now “a clear trend among state and federal courts permitting the admission of eyewitness expert testimony, at the discretion of the trial court, for the purpose of aiding the trier of fact in understanding the characteristics of eyewitness identification.” Commonwealth v. Walker, 92 A.3d 766, 782-83 (Pa. 2014) (collecting demonstrative cases from 44 states, the District of Columbia, and 10 federal circuit courts). The reason for this trend is that, although findings of the sort described in Dr. Fulero’s and Dr. Loftus’s reports are now “widely accepted by scientists,” those same findings “are largely unfamiliar to the average person, and, in fact, many of the findings are counterintuitive.” State v. Guilbert, 49 A.3d 705, 723-24 (Conn. 2012) (collecting cases and studies demonstrating this point). At the same time, advances in DNA testing have confirmed that “eyewitness misidentification is now the single greatest source of wrongful convictions in the United States, and responsible for more wrongful convictions than all other causes combined.” State v. Dubose, 699 N.W.2d 582, 591-92 (Wis. 2005) (collecting relevant studies). In other words, in the 25 years since Enis, we not only have seen that eyewitness identifications are not always as reliable as they appear, but we also have learned, from a scientific standpoint, why this is often the case. Accordingly, whereas Enis allowed for but expressed caution toward the developing research concerning eyewitness identifications, today we are able to recognize that such research is well settled, well supported, and in appropriate cases a perfectly proper subject for expert testimony. ¶ 25 With that in mind, we now turn to the issue before us: whether, given the specific facts presented, the trial court abused its discretion in denying defendant’s - 10 - request to allow Dr. Loftus’s expert testimony in this case. For the following reasons, we hold that it did. ¶ 26 To begin with, there is no question that this is the type of case for which expert eyewitness testimony is both relevant and appropriate. The only evidence of defendant’s guilt in this case is the eyewitness identifications made by Clark and Gill. There is no physical evidence tying defendant to the crime, and defendant neither confessed nor made any other type of incriminating statement. In other words, the State’s case against defendant hangs 100% on the reliability of its eyewitness identifications. In addition, of the several factors that both Dr. Fulero and Dr. Loftus identified as potentially contributing to the unreliability of eyewitness testimony, most are either present or possibly present in this case. These include the stress of the event itself, the use and presence of a weapon, the wearing of a partial disguise, exposure to postevent information, nighttime viewing, and cross-racial identification. 4 Next, of the State’s two eyewitness identifications, only one was subject to adversarial testing and cross-examination at trial, as Gill’s identification was admitted through the testimony of other witnesses under the excited utterance exception to the hearsay rule. And finally, while both the trial court and the State insist that both Gill and Clark “knew” defendant prior to the crime and that their identifications of defendant therefore are presumptively reliable, the record is far from clear on this point as it relates to Clark. Indeed, while Clark testified at trial that she had seen defendant from across the street approximately ten times in the six months to a year prior to the shooting, she testified before the grand jury that she had seen him only “[l]ike once or twice” before the shooting. Either way, Clark had only ever seen him standing on a porch from across the street. She expressly testified that she had never spoken with him before, had never been in the same room with him, and had never even been in the same house as him before. And when asked directly how long she had known defendant prior to the shooting, Clark’s response was, “I did not know him.” Even construed in the State’s favor, this degree of “knowledge” is limited at best and nothing like the State’s representation to the trial court that Clark “knows the defendant” and “always maintained she knew him.” Again, Clark expressly denied knowing him. Taken together, these considerations convince us that this is exactly the type of case in which expert eyewitness testimony would be both probative and admissible. 4 The record shows that Clark is African-American, whereas defendant is Hispanic. - 11 - ¶ 27 The next question, therefore, is whether the trial court abused its discretion in denying defendant’s request to admit Dr. Loftus’s expert testimony in this case. We hold that it did. The trial court denied defendant’s request for reasons “consistent with the reasons [it] set forth in detail when [it] made the ruling on [defendant’s] similar motion with respect to Dr. Fulero.” As discussed above, those reasons consisted primarily of the trial court’s personal conviction that “it is a fact that persons *** are less likely to misidentify someone they have met or know or seen before than a stranger,” as well as the trial court’s belief that expert testimony of this sort would both “generate *** a referendum on the efficacy of identification testimony generally” and “operat[e] as [an] opinion on the credibility” of the eyewitnesses themselves. As to the first of those points, the trial court emphasized at the hearing on defendant’s first motion to reconsider that its reasoning was “supported by the defense’s own witness, Dr. Fulero,” who is reported to have offered the same opinion in Nickleberry. ¶ 28 The problem with the trial court’s reasoning is that, even if it is defensible as to Dr. Fulero’s expected testimony, it is not defensible as to Dr. Loftus’s expected testimony. To be sure, Dr. Fulero’s report did not specifically address identifications made by witnesses who were acquainted with the accused prior to the crime. Neither did Dr. Fulero’s report specifically address the scope of his testimony and whether he would comment or offer an opinion on the credibility of Clark’s identification in this case. But Dr. Loftus’s report did address both of these matters directly, and it contradicted both of the trial court’s previous assumptions about what Dr. Fulero would say on these points. Again, after reading Dr. Fulero’s report, the trial court concluded that expert eyewitness testimony was unnecessary in this case both because “everyone knows” that acquaintance identifications are reliable and because Dr. Fulero’s testimony would likely “operat[e] as [an] opinion on the credibility” of the eyewitnesses themselves. Dr. Loftus’s report, however, directly addressed both of these points and stated the very opposite of what the trial court had previously assumed. According to Dr. Loftus’s report, the factors impacting the reliability of eyewitness identifications can operate even when the witness is previously acquainted with the accused, and Dr. Loftus’s testimony would not include any opinion as to the credibility of any specific witness or any specific identification. Nevertheless, in denying defendant’s request to allow Dr. Loftus’s testimony, the trial court simply invoked the reasons it gave for denying the admission of Dr. Fulero’s testimony, neither of which had any continued relevance in light of Dr. Loftus’s report. Indeed, Dr. Loftus’s report flatly - 12 - contradicted the trial court’s understanding. At this point, it is important to reiterate that, even in Enis, this court recognized that eyewitness identification is an appropriate subject for expert testimony. As importantly, both the State and the trial court in this case concede that Dr. Loftus is a qualified and highly respected expert in this field. Thus, in relying on its own personal beliefs about how eyewitness identifications function as the primary basis for denying the admission of Dr. Loftus’s testimony, the trial court in this case not only ignored the explicit contents of Dr. Loftus’s report but also effectively substituted its own opinion on a matter of uncommon knowledge for that of a respected and qualified expert. ¶ 29 At the same time, the trial court’s repeated insistence that the eyewitnesses in this case both “knew” defendant prior to the shooting is not nearly as certain and settled a fact as the trial court suggests. To be sure, there is no question that Gill knew defendant well and for several years prior to the shooting. As to Clark, however, the evidence is much less clear. Again, although Clark testified at trial that she had seen defendant from across the street approximately ten times in the six months to a year prior to the shooting, she testified before the grand jury that she had seen him only “[l]ike once or twice” before the shooting. Either way, Clark had only ever seen him standing on a porch from across the street. She expressly testified that she had never spoken with him before, had never been in the same room with him before, and had never even been in the same house as him before. And when asked directly how long she had known defendant prior to the shooting, Clark’s response was, “I did not know him.” Depending, then, upon which version of Clark’s testimony the jury chose to believe, the State’s only testifying eyewitness in this case is someone who had seen defendant from across the street possibly once but no more than ten times in the year before the shooting, had never spoken to defendant, had never been in the same room or in the same house as defendant, and expressly denied knowing defendant. Nevertheless, the trial court’s primary basis for denying defendant’s request to present expert eyewitness testimony in this case is the “glaring” fact that “the persons who identify Mr. Lerma *** all claim to have known him.” This is overstating the State’s evidence at best. ¶ 30 Finally, we note that, in denying defendant’s motion, the trial court expressly invoked the Ohio Court of Appeals’ decision in Nickleberry, claiming that the summary of Dr. Fulero’s testimony in that decision demonstrates that the trial court’s ruling was “supported by the defense’s own witness.” There are at least two problems with this. The first problem is that defense counsel informed the trial court on multiple occasions that the one-sentence summary contained in - 13 - Nickleberry does not accurately reflect Dr. Fulero’s full testimony in that case and that Dr. Fulero was fully prepared to clarify that if allowed to testify. In addition, defense counsel informed the trial court that Dr. Fulero was prepared to testify specifically in defendant’s case that “misidentifications have occurred with people who the witness knew beforehand.” Rather than hear from Dr. Fulero himself, however, the trial court chose to treat Nickleberry’s one-sentence summary of Dr. Fulero’s 1999 testimony not only as indisputably accurate but also as a binding and authoritative representation of Dr. Fulero’s opinion in 2012, such that the trial court cited Nickleberry as a valid basis for denying defendant’s motion. Needless to say, we find this troubling, as it is not difficult to imagine either that a one-sentence summary fails to capture the full complexity of an expert witness’s entire trial testimony, or that an expert opinion might have evolved over the course of 13 years. Unfortunately, the trial court’s approach foreclosed the discovery of either possibility. ¶ 31 The second problem with the trial court’s invocation of Nickleberry is that, even assuming that the one-sentence summary of Dr. Fulero’s testimony in that case is accurate, that summary in no way supports the trial court’s decision to exclude expert testimony in this case. According to the trial court, Dr Fulero’s testimony, as summarized in Nickleberry, supports the trial court’s belief that the factors impacting the reliability of eyewitness identifications are not present in a case such as this, where “the persons who identify [the defendant] *** all claim to have known him.” (Emphasis added.) Likewise, in defending the trial court’s decision before this court, the State points out that when the trial court denied defendant’s request to present Dr. Loftus’s expert testimony, it was aware that Dr. Loftus’s claim that “misidentification can occur when a witness knows the subject” was “directly contradicted by” Dr. Fulero’s testimony, as summarized in Nickleberry. (Emphasis added.) The problem with both the trial court’s and the State’s positions (aside from the fact that Clark expressly disavowed knowing defendant), is that Nickleberry’s summary of Dr. Fulero’s testimony says absolutely nothing about when a witness “knows” the person he or she is identifying. On the contrary, according to that summary, Dr. Fulero testified in Nickleberry that the factors impacting the reliability of eyewitness identifications are not present when the witness is identifying “someone he or she has met before.” (Emphasis added.) Nickleberry, 2000 WL 1738356, at *3. This distinction is very important because, in this case, Clark’s testimony was unequivocal that she had never met defendant prior to the shooting. Again, Clark denied ever having spoken - 14 - to defendant, ever having been in same room or house as defendant, and for that matter even “knowing” defendant. Given Clark’s testimony on this point, Dr. Fulero’s prior opinion that identifications are more reliable when the witness has “met” the accused is not only irrelevant to the actual facts of this case, it cannot possibly support the trial court’s decision denying defendant’s request to present expert testimony on the reliability of eyewitness identifications. Nevertheless, according to both the trial court and the State, Dr. Fulero’s prior opinion did exactly that, serving as one of the primary justifications for the exclusion of both Dr. Fulero’s and Dr. Loftus’s testimony in this case. ¶ 32 Abuse of discretion is a highly deferential standard of review, and we reiterate that we will find such abuse only when the trial court’s decision is “arbitrary, fanciful, or unreasonable to the degree that no reasonable person would agree with it.” Rivera, 2013 IL 112467, ¶ 37. We find that to be the case here. As discussed above, what we have in this case is the trial court denying defendant’s request to present relevant and probative testimony from a qualified expert that speaks directly to the State’s only evidence against him, and doing so for reasons that are both expressly contradicted by the expert’s report and inconsistent with the actual facts of the case. A decision of that nature rises to the level of both arbitrary and unreasonable to an unacceptable degree, and we therefore find that the trial court’s decision denying defendant’s request to admit Dr. Loftus’s expert testimony was an abuse of discretion. ¶ 33 The only remaining question is whether the trial court’s error was harmless. We hold that it was not. This court has recognized three approaches to determine whether an error such as this is harmless beyond a reasonable doubt: (1) whether the error contributed to the defendant’s conviction; (2) whether the other evidence in the case overwhelmingly supported the defendant’s conviction; and (3) whether the excluded evidence would have been duplicative or cumulative. People v. Blue, 205 Ill. 2d 1, 26 (2001). In this case, each of these approaches establishes that the trial court’s decision excluding Dr. Loftus’s testimony was not harmless beyond a reasonable doubt. First, there is no question that the error contributed to the defendant’s conviction, as the exclusion of Dr. Loftus’s testimony prevented the jury from hearing relevant and probative expert testimony relating to the State’s sole testifying eyewitness, in a case lacking any physical evidence linking defendant to the crime. Second, it cannot be said that the other evidence in the case overwhelmingly supported the defendant’s conviction, as other than Clark’s testimony, the only evidence of defendant’s guilt is a hearsay excited utterance - 15 - from a nontestifying witness. And while such evidence is certainly sufficient to support defendant’s conviction, we cannot say that it does so “overwhelmingly.” Third, the excluded testimony from Dr. Loftus was neither duplicative nor cumulative of other evidence, as the jury in this case heard precisely nothing in the nature of expert eyewitness testimony. ¶ 34 CONCLUSION ¶ 35 For the foregoing reasons, we affirm the judgment of the appellate court, which reversed the judgment of the trial court and remanded the cause for a new trial with directions to allow expert testimony on eyewitness identification subject to the provisions of Rule 702 of the Illinois Rules of Evidence (eff. Jan. 1, 2011). In doing so, we reiterate that the totality of the evidence presented at defendant’s trial was sufficient to prove defendant’s guilt beyond a reasonable doubt, such that no double jeopardy violation will occur on retrial. People v. Ward, 2011 IL 108690, ¶ 50. ¶ 36 Appellate court judgment affirmed. - 16 -