BECKWITH, Associate Judge:

Appellants Michael Robinson, Steven Edwards, and Davone Kellibrew appeal from numerous convictions arising from an incident in which the complainant, Donna Terry, was shot several times. A jury convicted all three appellants of assault with intent to kill while armed (AWIKWA), aggravated assault while armed (AAWA), assault with a dangerous weapon (ADW), and various gun-related charges. The jury also convicted Davone Kellibrew and Michael Robinson of kidnapping while armed and threatening to injure a person, found Mr. Kellibrew guilty of the separate charge of first-degree sexual abuse, and found Mr. Robinson guilty of the separate charge of simple assault. The jury acquitted Steven Edwards of kidnapping while armed, the associated count of possession of a firearm during the commission of a crime of violence (PFCV), carrying a pistol without a license (CPWL), unlawful possession of ammunition (UA), and possession of an unregistered firearm (UF). All three appellants were sentenced to lengthy prison terms.

Appellants raise myriad claims, but primarily contend that they were prejudiced by the trial court's exclusion of expert testimony regarding the effects of phencyclidine (PCP) — testimony appellants hoped would cast doubt upon the accuracy of the perceptions of the complaining witness. Appellants argue that the prejudicial exclusion of the expert testimony resulted either from the government's violation of its duties under Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), for failing to disclose to the defense that Donna Terry had used PCP on the morning of the offense, or from the trial court's error in excluding two different expert witnesses the defense lawyers scrambled to secure during the course of the trial. We conclude that the government did not violate its Brady obligations because it did not suppress the evidence that Ms. Terry had smoked PCP that morning. And while we conclude that the trial court erred in excluding relevant expert testimony on the effects of PCP, we find this error to be harmless. We affirm as to these and appellants' remaining claims, but remand the case so that the trial court may vacate certain duplicative convictions.

I. Facts and Proceedings

The government's theory in this case was that in the early morning hours of September 14, 2006, Donna Terry, who had spent several hours hanging out and selling crack in the area of 20th and Rosedale Streets, N.E., was subjected to a prolonged ordeal in which she was forced into Michael Robinson's sport utility vehicle (SUV), driven to various locations around D.C., sexually assaulted by Davone Kellibrew, and eventually pushed out of the vehicle, shot several times by Mr. Kellibrew and Steven Edwards, and left for dead in an alley. The government contended that this series of events began when the three men, having returned from a Maryland nightclub to the 20th and Rosedale neighborhood to find Ms. Terry waiting for her friend Steven Edwards, decided to punish Ms. Terry for allegedly stealing money from Mr. Edwards. It was undisputed that Ms. Terry had known both Mr. Edwards and Mr. Robinson for several years.

Davone Kellibrew's theory at trial was that he was not the third man in the car — the man the others called "Karim" — and that Ms. Terry mistakenly identified him in a photo array months after the incident. Michael Robinson contended that he was smoking crack and drinking throughout the incident, that he was uninvolved in the attacks on Ms. Terry, and that he was unaware that Karim intended to shoot her. Steven Edwards testified at trial that Ms. Terry, despite being a good friend of many years, was lying when she said he fired some of the shots at her, and that his only wrongdoing was lacking the courage to stop Karim's rampage because he feared the volatile man with a gun.

The Government's Evidence

When the government called its main witness, Donna Terry, the prosecutor asked Ms. Terry at the outset whether she had used drugs prior to the offense in this case. She acknowledged that she had smoked PCP and marijuana on the morning of September 13, 2006 — something the government had not disclosed to appellants prior to trial. Ms. Terry contended that the effects of these drugs typically lasted "about an hour" and would have worn off by that night.

Late that night, Ms. Terry took the bus to the corner of 20th and Rosedale, where she hoped she would find her friend, Steven Edwards, and that he would help her sell some crack. She hung out in the area by herself, selling a little crack, from about 12:30 a.m. to around 3 a.m., when she saw an SUV driven by Michael Robinson, a man she had known for a couple of years and was friendly with. Another man, whom she identified at trial as Davone Kellibrew, got out of the back seat of the SUV, greeted her by her nickname and reminded her that he had seen her before when he picked up a friend from her house. Finding Steven Edwards drunk and asleep in the front passenger seat of the SUV, Ms. Terry tried to wake him up and asked him to give her money for a cab ride home. On Mr. Edwards's instructions, she took some money out of his pocket and began counting out bills when Mr. Robinson approached her, accused her of trying to rob Mr. Edwards, then slammed her onto the ground, banged her head into the concrete, and punched her in the lip.

When Mr. Robinson told Mr. Kellibrew that Ms. Terry had "tried [to] rob little Steve," Mr. Kellibrew told Mr. Robinson to put her in the SUV, and said "I know what we going to do, we going to slump her," a term Ms. Terry took to mean "kill." According to Ms. Terry, the two men threw her in the back seat of the SUV and Mr. Robinson locked the vehicle's doors with child locks that prevented anyone from getting out. Ms. Terry woke up Steven Edwards — her friend since she was a young teenager1 — and told him what had happened. Mr. Edwards said he knew she would not rob him and crawled into the back seat with her.

Ms. Terry testified that Mr. Robinson drove, with Mr. Kellibrew directing, to what she thought was a crack house, and Mr. Kellibrew put a gun to her hip and pushed her while Mr. Robinson led the way in. Once inside, Mr. Robinson indicated that she was going to have sex for the fifty dollars he was throwing on the floor. Ms. Terry was crying throughout these events, told the men she was "not doing nothing," and when she began crying more loudly, Mr. Kellibrew put the gun to her head and said they were "going to have to take her out of [t]here" because she was being too loud. The four got back in the SUV, with Mr. Kellibrew pushing Ms. Terry with the gun and forcing her into the back seat between him and Mr. Edwards.

Once in the SUV, Mr. Kellibrew forced Ms. Terry to perform oral sex while he pointed the gun at her head and directed Mr. Robinson where to drive. According to Ms. Terry, Mr. Kellibrew told Mr. Edwards that they were taking her to the woods by some train tracks to kill her. Eventually Mr. Kellibrew told Mr. Robinson to stop the SUV, told Mr. Edwards to get Ms. Terry out, and then shot her twice, in her stomach and her chest. Ms. Terry testified that Mr. Kellibrew then passed the gun to Mr. Edwards and told him that "he better finish [her] off." She was still standing when Mr. Edwards took the gun and shot her twice more. Ms. Terry fell down, the men got back in the SUV, and Mr. Robinson began to drive away. Instead of leaving, however, the vehicle pulled back around right beside her and somebody shot her about four more times in her back. After the men were gone, Ms. Terry got up and tried to walk and call for help, but soon collapsed near a dumpster.

When police arrived at around 4:30 a.m. in response to a 911 call they found Ms. Terry lying on her back and bleeding.2 The officer who found Ms. Terry said that she told him that "Steve and Mike" had shot her and described a gold SUV.3 At the hospital, where she was treated for seven or eight gunshot wounds, Ms. Terry told a detective that three people were involved in the incident — Steven Edwards; a man named Mike, who drove a gold SUV; and another man whose name she did not know but whom she had seen before when he picked up her friend. She identified that man and Mr. Edwards as the actual shooters.

When police located Steven Edwards and interviewed him about the events of the previous day, he acknowledged having been in the area of 20th and Rosedale after returning from a club called Tradewinds with two friends in a tan SUV. Mr. Edwards said he was drunk and passed out, but that he remembered seeing a man in the back seat of the SUV forcing Donna Terry to perform oral sex at gunpoint. He said the vehicle eventually pulled into an alley, where someone told the driver to stop, Ms. Terry either ran or was pushed out of the truck, and Mr. Edwards heard a gunshot.

More than six months later, police found and questioned Davone Kellibrew after Ms. Terry identified him from an array of nine photographs. During a three-hour interview, Mr. Kellibrew told police that he had been at a club called Tradewinds on the night in question and had returned around 3 a.m. to the area of 20th and Rosedale with two friends in a brown SUV.

Evidence for the Defense

In support of Davone Kellibrew's defense that he was not the third man involved in the offenses, his girlfriend, Devin Waters, provided testimony designed to show that Mr. Kellibrew could not have been the third man involved in the offense.4 In Steven Edwards's defense, Sterling Holton, who said he was friends with both Steven Edwards and Donna Terry, testified that Ms. Terry told him several months after the incident that she did not know why she had told people that Steven Edwards had shot her, that she "didn't know the boy that shot her," that Mr. Edwards was asleep throughout the incident, and that she "was just mad at the time."

Steven Edwards himself testified that although he was present in the SUV with Ms. Terry and the other two men on the night of the offense, he was not involved in the crimes, he did not shoot Ms. Terry, and he was guilty only of being a coward who failed to stop these events. Mr. Edwards also testified that the third man in the vehicle who committed the sexual assault and the shooting was not Davone Kellibrew, but a man named Karim whom he did not know well. Mr. Edwards got so drunk at Tradewinds that he passed out in the front seat of Michael Robinson's SUV and remembered little of what happened afterwards, though he did recall looking in the back seat and seeing Karim holding a gun to Donna Terry's head while she performed oral sex. Mr. Edwards said nothing because he was afraid Karim would "bust [his] head." He testified that the group never made the intervening stop at a crack house that Ms. Terry described, but that they drove straight to an alley where Karim pulled Ms. Terry out of the car and shot her.

Jury Deliberations and Verdict

In the afternoon of its second day of deliberations, the jury returned a verdict convicting Davone Kellibrew and Michael Robinson of all counts against them but submitted a note indicating the jurors "disagree[d] about the other defendant." The jury continued to deliberate and eventually acquitted Steven Edwards of kidnapping while armed and the associated PFCV, CPWL, UF, and UA, while finding him guilty of AWIKWA, AAWA, ADW, and three counts of PFCV.

II. The Exclusion of Expert Testimony and Appellants' Brady Claims

We first address appellants' claim that the trial court erroneously excluded their expert testimony regarding the effects Donna Terry's recent PCP use may have had upon the accuracy of her perceptions, and the related claim that the government's failure to disclose that Ms. Terry had used PCP that morning caused their failure to give timely notice of their intent to call a drug expert at trial.

A. The Parties' Arguments and the Trial Court's Rulings

Following the revelation that Ms. Terry had smoked PCP and marijuana that day, appellants first responded to the unexpected testimony by seeking to call a drug expert, Detective Mark Stone, to explain that PCP has a hallucinogenic and reality-clouding effect within a 24-hour period after it is used and to undercut Donna Terry's insistence that its effects lasted no more than an hour. Appellants proffered that Detective Stone would testify that the perception-distorting effects of PCP are triggered by the kind of exertion and stress Ms. Terry experienced during her ordeal, and that PCP also had a tranquilizing effect that was arguably manifested by Ms. Terry's ability to remain standing even after being shot. Detective Stone had been qualified as an expert hundreds of times in Superior Court and federal court and had repeatedly testified about the effects of PCP, usually on behalf of the government. Appellants argued that the court should excuse the lack of pretrial notice of the drug expert because they had "no way of knowing that it was even going to be relevant" until Ms. Terry testified that she had used PCP the morning of the offense.

The government objected to the admission of Detective Stone's testimony because of the lack of notice and because the proffered expert testimony was "a collateral matter" unless the defense called "a pharmacologist who can testify with certainty as to what the effects were at the time of the alleged offense." The prosecutor nevertheless acknowledged that Detective Stone was a good and experienced expert with expertise in the indicia and primary effects of drug use. The trial court sustained the government's objection on timeliness grounds and because the proffer was, in the court's view, insufficient to establish that Detective Stone was "competent to testify with respect to the specific effects of PCP on Miss Terry on that occasion during the times that we've indicated."

When trial resumed, appellants asked the court to rule that the government had violated Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), in failing to timely disclose that Ms. Terry had used PCP on the morning of these offenses, thus hampering the defense from getting an expert to counter the government's claim that the PCP had no effect on Ms. Terry's perceptions 15 hours later. Counsel argued that even if the government did not violate Brady, the late disclosure of this information hamstrung defense counsel and prevented them from giving notice and securing an expert witness the court would accept. The trial court ruled that there was no Brady violation because the prosecutor was entitled to rely on Ms. Terry's pretrial representation that she was not under the influence of any drug during the offense and because appellants knew from a statement that Steven Edwards had given to police — and that the government had provided to appellants — that Ms. Terry was a PCP user.

In light of a concern the trial court had expressed that Detective Stone was not a pharmacologist, appellants informed the court prior to the government's rebuttal case that they had located a forensic pharmacologist who was a full professor at George Washington University and who taught forensic science focusing on drug issues. They asked the court to permit them to call the professor, Nicholas Lappis, as an expert witness to testify about the effects of PCP — specifically, how it acts as an anesthetic; how it causes hallucinations, sensory distortions, and grandiose delusions; how it impairs concentration and the ability to perceive things accurately; and how one who has used PCP can experience flashbacks under stress. The prosecutor opposed the request on grounds of timeliness and because the expert would not testify about the specific effect PCP had on Donna Terry that night. The court ruled that even assuming Professor Lappis was qualified, the request was untimely. The court also ruled that the expert was not able to "give an opinion within a reasonable degree of scientific certainty that 15 to 20 hours later [Ms. Terry's] ability to see, perceive, recall, understand were significantly affected by her use of PCP[.]"

After the jury returned its guilty verdicts, Davone Kellibrew filed (and his codefendants joined) a motion for new trial raising the Brady claim and arguing that the expert testimony was wrongly precluded. The trial court denied the motion at Mr. Kellibrew's sentencing.5

Michael Robinson also separately raised a claim under Brady v. Maryland that the government failed to disclose evidence that Steven Edwards had an alternative motive to commit the crimes against Donna Terry — namely, that Ms. Terry's boyfriend had previously cooperated with the government in a criminal prosecution against Mr. Edwards's brother. In Mr. Robinson's view, this evidence could have countered the government's theory that Ms. Terry's so-called theft from Mr. Edwards triggered Mr. Robinson's involvement in the offenses. The trial court denied that motion at Mr. Robinson's sentencing.

B. The Brady Claims

Appellants claim that the trial court erred in rejecting their contention both at trial and in the new trial motion that the government violated Brady v. Maryland in failing to disclose that Ms. Terry had smoked PCP on the morning of September 13, 2006. Had they known about Ms. Terry's recent PCP use, appellants argue, they would have given proper notice of expert testimony and they would have secured a competent expert witness to cast doubt upon Ms. Terry's ability to clearly perceive events 15 hours after she smoked PCP. The government does not seriously dispute that the evidence was favorable to appellants for Brady purposes, but contends instead that it did not suppress evidence about Ms. Terry's drug use and that in any event the evidence was not material.

The Due Process Clause of the Fifth Amendment to the U.S. Constitution requires the government to disclose material evidence — including impeachment evidence — that is favorable to the accused. Brady v. Maryland, 373 U.S. at 87, 83 S.Ct. 1194; Giglio v. United States, 405 U.S. 150, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972); Kyles v. Whitley, 514 U.S. 419, 433, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995). The government's failure to do so violates the directive of Brady if three factors exist: the evidence must be favorable to the accused; it "must have been suppressed by the State, either willfully or inadvertently"; and "prejudice must have ensued." Strickler v. Greene, 527 U.S. 263, 282, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999). When the government has failed to turn over favorable evidence in a timely way, that suppression is material — and this court must reverse the defendant's convictions — where there is "a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different." United States v. Bagley, 473 U.S. 667, 682, 105 S.Ct. 3375, 87 L.Ed.2d 481 (1985); see also Kyles, 514 U.S. at 434, 115 S.Ct. 1555. A reasonable probability of a different result is shown when the suppression of evidence "undermines confidence in the outcome of the trial." Id. (quoting Bagley, 473 U.S. at 678, 105 S.Ct. 3375).

At the outset, we disagree with the trial court's ruling that the government could not have suppressed the evidence of Ms. Terry's PCP use that morning because it turned over enough other information — primarily, Steven Edwards's statements to police about Ms. Terry's penchant for PCP in the past — from which appellants could have gleaned that Ms. Terry might have been under the influence of PCP on the day of the offense.6 Specific evidence that a key eyewitness used drugs on the day of the offense is the kind of bread-and-butter impeachment evidence that the government should disclose under Brady, particularly in a case, such as this one, that turns on the reliability of that witness's story and the accuracy of her perceptions. See Jackson v. United States, 377 A.2d 1151, 1154 (D.C.1977) (stating that the "use of narcotics is a proper subject of inquiry going to the credibility of the witness in his recollection of the events in question"); Perez v. United States, 968 A.2d 39, 65-66 (D.C.2009) (concluding that "[t]here is no question" the government should have disclosed a witness's grand jury testimony that he was drunk at the time of the assaults in that case). Disclosure of a defendant's statement mentioning the witness's prior drug use did not satisfy the government's obligation to disclose the more specific favorable information that she had used PCP that very morning. That the prosecutor himself had initially assumed that Ms. Terry had not used drugs on the day of the offense undercut the government's claim that appellants had every reason to expect such evidence to emerge at trial because they knew Ms. Terry had used PCP in the past and that she was selling crack that night.7

We nevertheless conclude that the government did not suppress the evidence for Brady purposes because there is no basis in the record for concluding that the government knew about Ms. Terry's recent PCP use prior to trial. At trial, the prosecutor stated that Ms. Terry had repeatedly told him that she had not used any drugs or alcohol that affected her perceptions at the time of the incident; he said he only learned the day before she testified, after asking her "more specifically... when was the last time you used drugs," that she had used PCP and marijuana on the morning of September 13. The government cannot have disclosed to the defense what it did not know itself.8 Sanchez v. United States, 50 F.3d 1448, 1453 (9th Cir.1995) ("The government has no obligation to produce information which it does not possess or of which it is unaware."); Reyes v. United States, 933 A.2d 785, 794 (D.C.2007).

There remains some question whether, once it learned about Ms. Terry's recent drug use the day before Ms. Terry testified, the government suppressed the information for critical hours during appellants' last-minute trial preparation — an interval that takes on greater importance in light of the trial court's subsequent rejection, partly on timeliness grounds, of the expert witnesses appellants scrambled to produce during the heat of trial. The government must disclose favorable evidence "at such a time as to allow the defense to use the favorable material effectively in the preparation and presentation of its case," and compliance with the Jencks Act is not enough. Edelen v. United States, 627 A.2d 968, 970 (D.C.1993).9

The record is less than clear, both with respect to exactly when the prosecutor learned about Ms. Terry's September 13 PCP use and with respect to when, if ever, the prosecutor informed the defense of this fact. At trial, the prosecutor stated that he learned about the September 13 PCP use the day before Ms. Terry testified. In its opposition to Mr. Kellibrew's motion for new trial, however, the government stated that it learned the information "shortly before she testified" and immediately informed defense counsel. In its brief, the government reiterates that it first learned this information on the day of Ms. Terry's testimony. Counsel for Mr. Kellibrew indicated at trial and in his motion for new trial, and now reiterates on appeal, that the defense did not learn of Ms. Terry's September 13 PCP use until she acknowledged it on the stand on January 18, 2008.10 At another point, Steven Edwards's attorney referred to the prosecutor's "proffer that as soon as he realized that the complaining witness had used PCP within the 24-hour period, it was disclosed to us in as timely a fashion as it could," and added that defense counsel "didn't know that that was even relevant until a short time after the government knew it was relevant." In the absence of any more concrete allegation by appellants — and any clearer indication in the record — that the prosecutor kept Ms. Terry's eleventh-hour admission to recent PCP use to himself, we are unwilling to say that the government suppressed it during the short, albeit important, time period just prior to her testimony. We conclude, therefore, that there is insufficient basis for finding that the government suppressed the evidence of Ms. Terry's recent PCP use.

We turn to Michael Robinson's separate Brady claim that the government failed to turn over evidence of Steven Edwards's alternative motive to harm Donna Terry. Specifically, Mr. Robinson contends that evidence that the father of Ms. Terry's child had helped the government obtain a conviction and harsh sentence against Steven Edwards's half brother would have undermined the government's theory that Ms. Terry's alleged theft of money from Steven Edwards was what motivated Mr. Robinson to take part in the ongoing assault. This claim fails to satisfy the requirements of Brady. First, while the government can be assumed to have known that it had prosecuted Mr. Edwards's half brother, Jesse Edwards, and that the father of Ms. Terry's child, Eric McDaniels, had implicated Jesse Edwards in a crime,11 there is no indication that the government knew of any fallout from that case that would connect it to the assault Steven Edwards was alleged to have committed against Eric McDaniels's child's mother in the present case.12 We disagree with the government's assertion that because motive is not an element of any of the offenses and was not part of the government's theory, evidence of this sort "is of no consequence." On the contrary, evidence tending to bolster a codefendant's motive in an offense could turn out to be critically important to another defendant's defense, regardless of whether — or perhaps because — the prosecutor is downplaying motive. Moreover, what is favorable for Brady purposes must take into account how such evidence would play in the hands of competent defense counsel in an unfolding trial. Kyles, 514 U.S. at 439-40, 115 S.Ct. 1555 (admonishing prosecutors to resolve close questions in favor of disclosure). But in this case, the proffered evidence of the relationships among the various players in these two half-brothers' prosecutions is too attenuated and inconclusive for us to deem it favorable evidence that the government had a duty to disclose — assuming the government even grasped these family connections and their potential to suggest a motive on the part of Steven Edwards to avenge his brother's conviction and sentence by hurting Ms. Terry.

C. The Exclusion of Expert Testimony Regarding the Effects of PCP

Appellants argue that whether or not the government violated its obligations under Brady v. Maryland, the trial court still erred in precluding them from presenting expert testimony regarding the general effects of PCP in an effort to cast doubt upon the accuracy of Donna Terry's perceptions of the events that night and to undermine her claim that the PCP affected her for only one hour.13

The admission of expert testimony falls within the discretion of the trial judge, and we will sustain the court's decision to admit or exclude such testimony unless it is an abuse of discretion. Benn v. United States, 978 A.2d 1257, 1273 (D.C. 2009). We have nevertheless "cautioned that because the right to confront witnesses and to present a defense are constitutionally protected," the trial court, in exercising its discretion, "must be guided by the principles that the defense should be free to introduce appropriate expert testimony." Benn, 978 A.2d at 1269 (citation and internal quotations omitted); see also Johnson v. United States, 398 A.2d 354, 365 (D.C.1979) ("Discretion without a criteria for its exercise is authorization of arbitrariness.") (quoting Brown v. Allen, 344 U.S. 443, 496, 73 S.Ct. 397, 97 L.Ed. 469 (1953)). In general, expert testimony should be admitted if it is relevant and is likely to help the trier of fact in its search for the truth, Smith v. United States, 27 A.3d 1189, 1195 (D.C.2011) — "that is to say, in understanding the evidence, determining the facts that must be found and rendering its verdict." Steele v. D.C. Tiger Market, 854 A.2d 175, 181 (D.C.2004). Whether expert testimony is helpful to the jury is determined by the well-established criteria for the admission of expert testimony set forth in Dyas v. United States, 376 A.2d 827, 832 (D.C.1977),14 and ultimately turns on "the relevance and probative value of the proposed scientific evidence." Benn, 978 A.2d at 1278; see also Ibn-Tamas v. United States, 407 A.2d 626, 632, 639 (D.C.1979) (identifying "two levels of analysis" in the decision whether to admit expert testimony: the "three-fold test" of Dyas and the weighing of "probative value versus prejudicial impact"). A trial court may exclude outright speculation, but short of speculation, a particular expert witness's degree of certainty in proffering an opinion goes to the weight of the testimony, not its admissibility, and "the weight to be given an expert opinion is for the jury to decide." Clifford v. United States, 532 A.2d 628, 639 (D.C. 1987).

In this case, the admissibility of the evidence turned not upon the Dyas criteria but upon appellants' failure to provide notice of the expert testimony and upon the proffered experts' inability to provide what the trial court viewed as relevant testimony. The trial court excluded the testimony on notice grounds, but ruled that it would have excluded it anyway on the ground that neither Detective Stone nor Professor Lappis could testify as to the actual effects the PCP had upon Donna Terry at the time of the offense in this case. We conclude that the trial court erred in excluding this testimony on notice grounds and in misapprehending the testimony's relevance and ability to assist the jury.

The trial court's principal reason for excluding the expert testimony — appellants' failure to give adequate notice of the testimony — was unwarranted given how late the parties became aware of Ms. Terry's use of PCP despite determined efforts prior to trial to obtain evidence of any recent drug use, and given the importance of appellants' right to present a defense and to meaningfully confront Ms. Terry about the PCP. Where the prosecutor himself acknowledged that evidence of Ms. Terry's recent drug use took him by surprise, and where the defendants quickly found an expert witness who could testify at trial, excluding that witness's testimony on grounds of untimeliness was error. Cf. Holmes v. South Carolina, 547 U.S. 319, 330-331, 126 S.Ct. 1727, 164 L.Ed.2d 503 (2006) (state evidentiary rule barring relevant evidence of third-party guilt arbitrarily infringed upon right to present a defense); Tucker v. United States, 571 A.2d 797, 799-800 (D.C.1990) (trial court's discretionary denial of appellant's reasonable request for a continuance was reversible error); Ungar v. Sarafite, 376 U.S. 575, 589, 84 S.Ct. 841, 11 L.Ed.2d 921 (1964) (stating that "insistence upon expeditiousness in the face of a justifiable request for delay can render the right to defend with counsel an empty formality"). The court's ruling in this regard stemmed from its view that the defendants knew enough about Ms. Terry's penchant for PCP to have anticipated the need for an expert witness. For the reasons discussed above, neither Steven Edwards's statement to police nor the fact that Ms. Terry was selling crack put appellants on notice that Ms. Terry's perceptions of the offense may have been distorted by her use of PCP hours beforehand.

In addition to ruling on notice grounds, the trial court indicated that the two experts' inability to testify with certainty about the PCP's precise effects on Donna Terry provided an alternative ground for excluding the evidence. We do not believe this concern provided a valid basis for excluding the expert testimony in the circumstances of this case. In its brief, the government casts this issue in terms of the experts' "qualifi[cations] to discuss the effects of Ms. Terry's PCP use." While the questions are intertwined, the trial court's alternative ruling seems to us more a question of the basic relevance of the proffered testimony and whether it would assist the jury in understanding the facts in issue.15 In the trial court's view, because the expert witnesses had not interviewed Ms. Terry, they had no basis for testifying that Ms. Terry's descriptions of the events of that night were unreliable, and their testimony was therefore not relevant. The court did not indicate that admission of the testimony would create a danger of unfair prejudice, but treated such evidence as invariably irrelevant unless the expert could explain the effects of PCP on the actual witness who smoked it.

In this case, Donna Terry took the stand and stated that she had smoked PCP and marijuana that morning when she woke up, that she had used PCP before, and that its effects typically lasted no more than an hour. When defense counsel tried to press her on this subject on cross-examination, she added that she had eaten breakfast and taken her son to a friend's house prior to smoking the drugs, and then reiterated that the PCP affected her for one hour. No one contests that the questions the prosecutor and defense lawyers posed to Ms. Terry about her drug use that day — whether she had taken drugs, what drugs she had taken, when she had taken them — were relevant inquiries. They were relevant to her perception of the events of that day and to her ability to remember them. See, e.g., Jackson, 377 A.2d at 1154 (stating that the "use of narcotics is a proper subject of inquiry going to the credibility of the witness in his recollection of the events in question"); Durant v. United States, 551 A.2d 1318, 1326 (D.C.1988) (stating that "a witness'[s] use of drugs is not a collateral issue when an evidentiary foundation can be established that the witness was using drugs at the time of the incident") (citation and internal quotation marks omitted). If the basic questions about Ms. Terry's PCP use that day were relevant, appellants' attempt to shore up their impeachment of Ms. Terry's responses with expert testimony explaining the basis for such impeachment was equally relevant. Cf. Smith v. United States, 27 A.3d 1189, 1198 (D.C.2011) (holding that the relevance of expert testimony the trial court erroneously excluded was heightened by the government's own expert's testimony on the same subject); Kigozi v. United States, 03-CF-1181, 07-CO-684, slip op. at 15-33 (D.C. June 14, 2012) (finding counsel ineffective for failing to present expert testimony regarding the possible effects of PCP on the reliability of a dying declaration).

In the circumstances of this case, the background provided by an expert witness about the general effects of PCP within 24 hours of use would put jurors "in a better position"16 to make sense of Ms. Terry's responses on the stand and of certain facts — such as the way Ms. Terry remained standing after being shot several times, and the various peculiarities in her accounts of what happened — that in appellants' view suggested the PCP was still affecting her. The proffers contained sufficient detail to signify that if the jurors credited the expert's testimony, it would assist them in drawing fair inferences from the various facts they heard at trial about the PCP Ms. Terry smoked, when she smoked it, her past experience in using PCP, her self-reporting about how long it affected her, and her behavior at the time of the offense. See Kigozi, supra, slip op. at 33 n. 16 (stating that a juror could infer from expert testimony about the effects of PCP "that there was more than a mere possibility that [the decedent] might have been under the effects of PCP when he made his accusatory statements, even if it could not be established as an actual fact"). On this record, we are comfortable concluding that the expert's testimony would not have left the jury open to speculate and that the trial court erred in imposing an unduly stringent standard of relevance. See Allen v. United States, 603 A.2d 1219, 1224 (D.C.1992) (en banc) ("Probative evidence should not be excluded because of crabbed notions of relevance or excessive mistrust of juries.").

We disagree with the government that our decision in Coates v. United States, 558 A.2d 1148 (D.C.1989), compels a contrary conclusion. The government relies on Coates to support its contention that "the proffered expert testimony was inadmissible because the proposed expert witnesses had not evaluated Ms. Terry, and could not opine about PCP's [e]ffects on her personally." Coates instead establishes that such testimony is admissible.

In Coates, the defendant wanted to support his consent defense in a rape case by presenting expert testimony about the effects of two puffs of a marijuana cigarette laced with PCP on the behavior of the complaining witness. 558 A.2d at 1150. This court held that the trial court did not abuse its discretion in deeming too speculative a somewhat faltering proffer by the expert witness where that expert testified that the state of the scientific knowledge at the time (25 years ago) was that no one could predict a PCP user's behavior, where the trial court expressed "serious concerns about the witness'[s] expertise on the effect of street drugs on a person's behavior," where the amount of PCP was very small, where the expert acknowledged both the variability of many of the factors on which he relied and his limited experience in observing PCP users, and where the defense declined an opportunity to particularize its overly broad proffer by gathering from the complainant more information that might support a more reliable assessment of the effects PCP had on her behavior. Id. at 1151-55.

In recognizing that "observation of the individual who allegedly used the drugs is not necessarily a prerequisite for qualifying an expert to testify," our decision in Coates confirmed that expert testimony of the type proffered in this case is not inadmissible. Id. at 1154 (noting that it would not have been an abuse of discretion for the trial court to have admitted expert testimony on the effects of PCP). Nothing in Coates suggests the trial court could properly exclude such testimony in a case that did not feature the weak proffer and unusual factors in that case. Coates also rejected any standard that required an expert to hold her opinion to a "reasonable degree of scientific certainty" and confirmed that any uncertainty on behalf of the expert that falls short of speculation goes to the weight and not the admissibility of the testimony. Id. at 1152 & n. 6. "A rule of admissibility demanding a greater level of self-proclaimed certainty on the witness'[s] part would remove from the jury its role in weighing the evidence." Clifford, 532 A.2d at 639-40. The trial court's ruling here appeared to categorically exclude expert testimony based in part on the ground that it lacked such certainty.

As we have repeatedly emphasized, "the decision to admit or exclude expert testimony must be made on a case-by-case basis, grounded on the proffer made and on its potential to assist the jury in the particular case before the court." Benn, 978 A.2d at 1273 (citation omitted).17 Appellants' proffer here was clear cut and probative of the facts it was being offered to prove. The first expert witness appellants offered was a detective with decades of street experience handling drug cases who had testified as an expert witness hundreds of times in Superior Court. The prosecutor himself acknowledged Detective Stone's long history of testifying on the government's behalf on various issues that arise in drug cases, including the indicia of personal drug use and the general effects of various illegal drugs. The second expert was a forensic pharmacologist whom the trial court assumed (without deciding) was qualified. Appellants proffered that their expert would testify that PCP remains in the system for more than 24 hours, that it is released in concentrated doses in response to stressful situations like what Ms. Terry experienced, that a user's perceptions can be distorted when this happens, and that the tranquilizing effect of the PCP Ms. Terry smoked several hours earlier could have explained her ability to walk some distance without falling down after being shot several times. Given the evidence already in the record about Ms. Terry's use of drugs that day and her self-assessment of how long they affected her, this proffer was specific enough that if the jury credited it, it could draw fair inferences about the effects of the PCP upon her perceptions and her ability to recall events. This testimony could be useful to the jury's assessment of the witness's perceptions without the expert evaluating what the actual witness experienced.18 See, e.g., Benn, 978 A.2d at 1270 (discussing the admissibility, in appropriate cases, of "expert testimony concerning subtle psychological factors that might affect witnesses").

This court has stated that fairness dictates that criminal defendants, who have a right to present a defense, "should not be put at a disadvantage" in the use of expert testimony "comparable to that permitted to the government."19 Benn, 978 A.2d at 1270. Here, the proffered expert testimony could have lent credence to appellants' contention that PCP could substantially hinder a witness's ability to perceive and remember events many hours later. It would have made it more likely that the government's only eyewitness had distorted perceptions, and it would have provided a basis for questioning Ms. Terry's assessment of how much the PCP affected her. Curry v. United States, 520 A.2d 255, 268 (D.C.1987) (stating that evidence that has some logical tendency to prove or disprove a disputed material issue is relevant). We therefore conclude that in addition to erroneously precluding appellants' expert testimony on notice grounds, the trial court's alternative rejection of the testimony because the expert could not specifically say how PCP affected Ms. Terry misapprehended the purpose for which the evidence was offered and ran afoul of our case law indicating that expert testimony "should generally be admitted if it will assist the jury to understand the facts in issue." Clifford, 532 A.2d at 632.

Harmless Error Analysis

While we believe the trial court erred in excluding the expert testimony on these grounds, that error is not reversible if we can say "with fair assurance, after pondering all that happened without stripping the erroneous action from the whole, that the judgment was not substantially swayed by the error." Kotteakos v. United States, 328 U.S. 750, 765, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946). We take into account factors such as the closeness of the case, the centrality of the issue affected by the error, and any steps taken to mitigate the error, but "our focus is on the likely impact of the alleged error on the jury's verdict." Settles v. United States, 615 A.2d 1105, 1109 (D.C.1992). Given the evidence in this case, we are confident in holding that the erroneous exclusion of the expert testimony did not substantially influence the outcome of the trial.

It is beyond serious question that Donna Terry's ability to perceive the events and the accuracy of her identification of her alleged assailants were central to the government's case. But this is not like the exclusion of expert identification testimony in a case in which the government's case rests in large part upon the identification of witnesses who were total strangers to a defendant. See, e.g., Benn, 978 A.2d at 1282; Russell v. United States, 17 A.3d 581, 589 (D.C.2011). Here, because Ms. Terry knew two of the three perpetrators well and testified that the third man had reminded her that they had met before, Ms. Terry's identifications of these men were not the type that were particularly susceptible to mistakes because of her recent drug use.

The only appellant with an arguable misidentification defense — Davone Kellibrew — argues that the expert testimony could have convinced the jury that Ms. Terry had identified the wrong person. He contends that this theory is supported by Ms. Terry's failure to name him as the shooter in her first interviews with police, by evidence establishing various physical discrepancies between him and the third man Ms. Terry described, by evidence that he did not drive a car or even have a driver's license, and by Steven Edwards's testimony that Davone Kellibrew was not the third man in the SUV that night.

The nature of the excluded evidence and the strength of other aspects of the government's case against Mr. Kellibrew — particularly Mr. Kellibrew's statement placing himself with two men in an SUV in the same place and at the same time the offense began that night — convince us that the trial court's rejection of appellants' expert testimony was harmless as to Davone Kellibrew. Ms. Terry testified unequivocally that the third man in the SUV was a man she had met before.20 She told police the same thing, she picked Mr. Kellibrew's photo out of a photo array, and she identified him in court. Her story had problems, and Mr. Kellibrew exposed these problems to the jury.21 But for all its potential weaknesses, Ms. Terry's testimony was corroborated by, among other things, Mr. Kellibrew's own police interview, during which he acknowledged being at 20th and Rosedale at around midnight the night of the offense, admitted going to a club called Tradewinds with two friends in a champagne-colored SUV, and admitted returning to 20th and Rosedale with those same two friends in that same SUV at around 3 a.m. when the club closed. These admissions by no means definitively established that Mr. Kellibrew was the third man, but the glaring coincidence — and the extent to which Mr. Kellibrew's statement dovetails with Steven Edwards's own testimony that he and Mr. Robinson met at 20th and Rosedale around midnight and went to Tradewinds in the SUV with "Karim" — belies Mr. Kellibrew's claim that Ms. Terry's PCP use that morning may have caused her to mistakenly identify him as the third man. Under these circumstances, expert testimony that Ms. Terry's perceptions may have been affected by the PCP she smoked when she woke up that day would not have appreciably undermined the rest of the evidence.

Steven Edwards and Michael Robinson focus more upon the extent to which the expert testimony could have made the jury doubt Ms. Terry's descriptions of each man's level of participation in the offenses and bolstered their contentions that they were present but not actively engaged in the assault on Ms. Terry. Even if the expert testimony came in very powerfully for the defendants, however, it would not have been likely to neutralize the evidence that Mr. Edwards and Mr. Robinson remained very involved in the incident over a prolonged period of time and that Mr. Robinson continued to drive the SUV from place to place as the situation worsened for Ms. Terry. It would not have significantly supported appellants' theory that Ms. Terry had fundamentally misperceived what was happening to her and misconstrued appellants' respective roles. While the proffered expert testimony was relevant and should have been admitted, the quality and nature of Ms. Terry's testimony was not such that it would have been undermined significantly more by this expert testimony than it already was by the various lines of attack appellants employed to impeach her credibility. Appellants' arguments to the jury on the subject of Ms. Terry's PCP use may have been frustrated by the absence of a concrete evidentiary basis for their suggestion that Ms. Terry's perceptions were distorted, but they were not nullified, as the jury knew that Ms. Terry had started her day by smoking PCP and marijuana, it heard the defense's impeachment of her statement that she was only high for an hour thereafter, and it heard appellants' theory in closing argument that the PCP had affected her perception of events.

With or without the expert testimony, the more plausible defense theory — and one pursued aggressively by appellants — was that Ms. Terry was fabricating certain details of her story that implicated Mr. Edwards and Mr. Robinson in offenses that they claimed were perpetrated primarily by the third man. Thus, Steven Edwards impeached Ms. Terry's allegation that Mr. Edwards had fired the second round of shots at Ms. Terry by calling a witness who testified that she admitted lying about that. Appellants also cast doubt upon Ms. Terry's testimony about how she spent all those hours at the corner of 20th and Rosedale and questioned the logic of her claim that she was merely waiting for Steven Edwards to arrive so she could borrow cab fare from him to get home. If Ms. Terry was dishonest about details like these, her testimony was suspect overall, and the jury could not trust her portrayal of the incident as a planned and coordinated event in which all three men actively participated. The jurors were not ultimately persuaded by this defense, but there is little basis for concluding that they would have found it any more convincing had appellants presented expert testimony that Ms. Terry may have experienced PCP-related hallucinations, distorted perceptions, or memory problems.

III. The Prosecutor's Rebuttal Argument

During his final argument in rebuttal of appellants' closing arguments, the prosecutor made the following statement about Donna Terry:

When appellants objected to this argument and moved for a mistrial, the trial court overruled the objection and denied the request for a curative instruction. In the trial court's view, the argument did not shift the burden of proof, and it was "not inappropriate for counsel to say if you want to argue some issue there needs to be some evidence on that issue."

Appellants contend on appeal that this argument improperly capitalized upon the trial court's exclusion of expert testimony by impermissibly leading the jury to believe that no evidence could possibly have demonstrated that Ms. Terry's PCP use earlier that day had affected her ability to accurately identify her attackers and describe the roles they played in the ordeal. They also contend that this argument improperly suggested that appellants had a burden to disprove the charges against them.

When evaluating a claim of improper prosecutorial argument, we consider first whether the comments in question were, in fact, improper, and if they were, we evaluate a number of factors in determining whether the improper comments substantially prejudiced appellants and therefore warrant reversal. Burgess v. United States, 786 A.2d 561, 570 (D.C. 2001); Carpenter v. United States, 635 A.2d 1289, 1295-96 (D.C.1993). These considerations include the gravity of the impropriety, its relationship to the issue of appellants' guilt, the effect of any corrective action on the part of the trial judge, and the strength of the government's case. Id. at 1296. In considering claims of improper argument, "it is our function to review the record for legal error or abuse of discretion by the trial judge, not by counsel." Gilliam v. United States, 46 A.3d 360, 366 (D.C.2012) (quoting Irick v. United States, 565 A.2d 26, 33 (D.C.1989)).

To the extent the trial court believed it was proper for the prosecutor to comment upon appellants' failure to "bring some evidence" to support their argument that Ms. Terry's PCP use earlier that day affected her perception of events, we disagree.22 The suggestion inherent in the comment was that no such evidence existed: appellants "need to bring some evidence" but in fact "[t]here is none." That was untrue, as appellants proffered exactly such evidence. Moreover, even though the prosecutor himself acknowledged that he had been surprised by Ms. Terry's revelation of recent drug use, and even though he implicitly acknowledged the relevance of her drug use when he questioned her about it on direct examination, he opposed appellants' request to present expert testimony to aid the jury in understanding Ms. Terry's testimony largely on timeliness grounds, and persisted in defending against the Brady claim on the ground that appellants knew enough about Ms. Terry's drug history to have anticipated their need for an expert. In these circumstances, suggesting to the jury that it should give weight to the fact that "there is no[]" such evidence is unfair and inaccurate.

"[A] prosecutor's misleading statements during closing argument, especially rebuttal argument, may `so infect[] [a] trial with unfairness as to make the resulting conviction a denial of due process.'" Woodard v. United States, 2012 WL 1888131, at *2 (D.C. May 24, 2012) (quoting Darden v. Wainwright, 477 U.S. 168, 181, 106 S.Ct. 2464, 91 L.Ed.2d 144 (1986) (internal quotation and citation omitted)). In Woodard v. United States, this court recently found reversible error in a case in which the prosecutor's rebuttal argument to the jury created a misimpression about facts bolstering a witness's credibility. The prosecutor in that case urged the jury to draw an inference that a witness had been afraid to identify a shooter face-to-face at the first trial in that case, all the while knowing what the jury did not know — that at the prior trial the witness had in fact made just such a face-to-face identification of another defendant as one of the shooters in the case. Id. at *2-3. Woodard held that the trial court erred in failing to correct the prosecutor's misleading statement, noting that regardless of a prosecutor's intent, "the law's aim is `to ensure [that the] jury is not misled by falsehoods.'" Id. (quoting Woodall v. United States, 842 A.2d 690, 697 (D.C. 2004)).

More akin to the case before us is our decision in Williams v. United States, 805 A.2d 919 (D.C.2002), where we affirmed the trial court's ruling precluding defense counsel from highlighting in closing argument the government's failure to present evidence of blood at the crime scene when that very evidence had been excluded at the request of the defense. Id. at 922-26. In that case, the prosecutor "vigorously" objected to the argument, asserting that the only reason the blood evidence "didn't come out in trial [was that] we were precluded from [introducing] it." Id. at 924. This court held that whether the evidence was excluded by an evidentiary ruling or as a sanction, "the result was the same" and "[t]he government was expressly prohibited from introducing any evidence about blood at the scene of the crime, even though such evidence existed." Id. at 926.

The same is true regarding appellants' proffered expert testimony on the effects of PCP 15 hours after use. "The prosecution's duty of candor ... extends to facts not in the record but known to the government." Woodard, at *2 (citing Powell v. United States, 880 A.2d 248, 258 n. 23 (D.C.2005) (stating that prosecutors must "guard against inviting inferences of fact arguably contrary to evidence of which they are aware")). As in Williams, the argument suggesting no such expert evidence existed was "contrary to fact, misleading, and inherently unfair," id. at 926, and it highlighted a defect in their case that appellants were powerless to correct not because the evidence was nonexistent, but because the trial court precluded it at the government's behest.23

The trial court's ruling on the improper argument did not prejudice appellants, however.24 The comment was undeniably closely related to the issue of guilt, as the success of the government's case depended largely upon Donna Terry's reliability, and the prosecutor's comment created the erroneous impression that appellants could never have presented proof that PCP could distort a person's memory of events 15 hours later. But for many of the same reasons we have held that the erroneous exclusion of the expert testimony was harmless in this case — most fundamentally, the unlikelihood that even the full-fledged and unhampered expert testimony appellants sought to present would have significantly vitiated the essential story Ms. Terry told — we do not believe the prosecutor's improper references to the related facts the defense "need[ed] to prove" had a "substantial and injurious effect or influence in determining the jury's verdict." Kotteakos, 328 U.S. at 776, 66 S.Ct. 1239. It was an unfair argument, but it was not a particularly powerful one.

IV. Merger of Counts

Finally, appellants argue that various counts should be merged.25 We review issues of merger de novo. Hampleton v. United States, 10 A.3d 137, 146 (D.C.2010). At the outset, as "ADW is a lesser included offense of aggravated assault while armed," Gathy v. United States, 754 A.2d 912, 919 (D.C.2000); see also McClary v. United States, 3 A.3d 346, 355 (D.C.2010), appellants' convictions on these two counts merge.26 Appellants' convictions for AWIKWA and AAWA do not merge, however, as aggravated assault requires a showing of serious bodily injury, and AWIK requires a showing of specific intent. Tolbert v. United States, 905 A.2d 186, 190 (D.C.2006); see also McClary, 3 A.3d at 355 n. 6. Nor do Mr. Robinson's kidnapping and assault convictions merge, as they are not the "same act" for purposes of Blockburger v. United States, 284 U.S. 299, 304, 52 S.Ct. 180, 76 L.Ed. 306 (1932),27 and as each required proof of an element the other did not.28

Turning to the PFCV counts, while a defendant can be convicted of separate PFCV counts stemming from predicate armed offenses, Stevenson v. United States, 760 A.2d 1034, 1035 (D.C.2000), "multiple PFCV convictions will merge, even if the predicate felony offenses do not merge, if they arise out of a defendant's uninterrupted possession of a single weapon during a single act of violence." Matthews v. United States, 892 A.2d 1100, 1106 (D.C.2006). We have adopted the "fresh impulse" or "fork-in-the-road" test for determining whether two PFCV convictions stem from a single act of violence or distinct acts. Hampleton, 10 A.3d at 146.

The government presented evidence at trial that Mr. Kellibrew and Mr. Edwards shot Ms. Terry and that Mr. Robinson aided and abetted that shooting. For that specific conduct, the government successfully prosecuted appellants for AWIKWA, AAWA, and ADW, and the associated PFCV convictions for these counts.29 These charges "overlapped substantially and were not independent of each other," Matthews, 892 A.2d at 1107, and indeed, we have already decided that ADW merges with AAWA. Two of appellants' three PFCV counts must therefore be vacated as duplicative. The PFCV counts stemming from Mr. Kellibrew's and Mr. Robinson's convictions for kidnapping while armed do not merge with the other PFCV counts, however, as the kidnapping and shooting were distinct acts between which appellants faced an unmistakable fork in the road.

V. Conclusion

For the reasons stated in this opinion, we affirm the judgments on appeal, except that we remand for the trial court to vacate appellants' convictions on the ADW count and two of each appellants' PFCV counts.

So ordered.