Appellants Jeremiah Mungo and Lamont A. Peete were tried jointly for the murders of William Powell and Norman Isaac, who were fatally shot on June 3, 1997, as they sat in a car in an alley behind 58th and Blaine Streets, N.E. After the jury in appellants’ first trial was unable to reach a verdict and the court declared a mistrial, appellants were retried during January and February 2000. The second jury found Peete guilty as charged of two counts of first-degree premeditated murder while armed, while convicting Mungo *1148of two counts of the lesser-included offense of second-degree murder while armed. Both appellants were convicted of weapons charges as well. Both filed direct appeals from their convictions and, thereafter, both filed motions under D.C.Code § 23-110 seeking relief on the ground of ineffective assistance of counsel. After conducting an evidentiary hearing during the summer of 2006, the trial court denied both 28-110 motions in a single written order dated February 11, 2008. Both Mungo and Peete appealed that ruling, and we consolidated those appeals with their direct appeals.
The direct and collateral appeals raise related issues. In their direct-appeal briefs, appellants contend variously that the trial court (1) erred by permitting the use of stun belts during jury voir dire; (2) erroneously exercised its discretion in ruling on the parties’ Batson challenges; (3) violated appellants’ rights under the Confrontation Clause in permitting a forensic pathologist to testify on the basis of autopsy documents prepared by a different medical examiner, whom the government did not call to testify; (4) abused its discretion by allowing the government to introduce “other bad acts” testimony; (5) abused its discretion by not declaring a mistrial in light of the inappropriate demeanor of each appellant’s trial counsel; and (6) gave a legally erroneous and prejudicial aiding-and-abetting instruction. Appellants also contend that the trial court erred in rejecting their claims that their trial counsel provided ineffective assistance by failing to object to certain of the foregoing asserted trial-court errors, by their conduct that occasioned the errors, and by other omissions. We are unpersuaded that appellants are entitled to relief on any of these grounds, and we therefore affirm the judgments of conviction and the ruling of the trial court denying the 23-110 motions.
I. Background
The government presented several eyewitnesses who testified that they saw Mun-go and Peete, both carrying pistols, enter the alley and approach the driver’s side of a parked car occupied by Powell and Isaac. According to witnesses, after Mungo exchanged heated words with Powell, Peete fired multiple shots into the car from a distance of two to three feet away, turned to leave, and then fired a second round of shots into the car. Police found Powell slumped over the steering wheel and Isaac in the passenger’s seat, both having sustained multiple gunshot wounds.
II. Stun Belts
During jury selection, the prosecutor advised the court that he was having lower-back problems that made standing for long periods of time painful, and he raised the possibility of conducting voir dire in the jury room instead of in the courtroom. Mr. Beaman, trial counsel for appellant Peete, commented that the “only difficulty will be[,] if we go in the jury room, Your Honor, they will have to take time to strap our clients into harnesses. I’m not sure how visible those harnesses are.” The trial judge responded, “They have the stun belts.1 I don’t think they have to be very visible.” The court then asked, “Are you asking for [the defendants] to be present?” When counsel Bea-*1149man responded in the affirmative, the court said, “it may just be easier to do it that way. So I’ll talk with the deputies .... ” The record establishes that appellants did wear stun belts during two days of jury voir dire while they sat at one end of a table about fifteen feet away from where individual jurors were seated during their interviews. Acknowledging that their trial counsel neither objected to use of the stun belts nor asked the court to consider other security options, appellants now argue that the trial court plainly erred in directing or permitting use of the belts without making findings on the record to justify use of the devices. They also argue that their lawyers provided ineffective assistance by their failure to object to use of the stun belts, to propose alternatives, to ensure that the belts were not visible to jurors, to ensure that the court made relevant findings, and to ensure that the Marshal’s Service rules for use of stun belts were followed.
We are not persuaded that the trial court plainly erred in permitting the use of stun belts under the circumstances here.2 To be sure, the Supreme Court has held that “absent a trial court determination, in the exercise of its discretion” that their use is “justified by a state interest specific to a particular trial,” the use of visible physical restraints during the guilt phase of a criminal trial violates due process because it “undermines the presumption of innocence and the related fairness of the factfinding process.” Deck v. Missouri, 544 U.S. 622, 629, 630, 635, 125 S.Ct. 2007, 161 L.Ed.2d 953 (2005). The Court has also recognized that the use of physical restraints can interfere with a defendant’s ability to communicate with his counsel. Id. at 631, 125 S.Ct. 2007 (“[Sjhackles ‘impose physical burdens, pains, and restraints ..., tend to confuse and embarrass’ defendants’ ‘mental faculties,’ and thereby tend ‘materially to abridge and prejudicially affect his constitutional rights.’ ”) (citing People v. Harrington, 42 Cal. 165, 168 (1871)). Several courts of appeals in other jurisdictions have reasoned that stun belts amount to physical restraints, see, e.g., Gonzalez v. Pliler, 341 F.3d 897, 904 (9th Cir.2003); can potentially chill a criminal defendant’s consultation with counsel because of the danger of accidental activation of an electrical charge, see, e.g., Durham, 287 F.3d at 1302 n. 2, 1305, 1308; and should not be used unless the trial court makes findings on the record that their use is warranted. See, e.g., United States v. Wardell, 581 F.3d 1272, 1287-88 (10th Cir.2009); see also United States v. Edelin, 175 F.Supp.2d 1, 4-5 (D.D.C.2001). But, critical to our analysis, neither the United States Supreme Court nor this court has ever held that a stun belt qualifies as a type of physical restraint whose use is subject to the strictures that the Supreme Court set out in Deck, that the reasoning of Deck (which addressed the use of physical restraints during the guilt phase of a trial or during the penalty phase of a capital trial) applies equally to the limited use of restraints during jury selection; or that on-the-record findings are required before stun belts may be used. Accordingly, we cannot conclude *1150that the trial court plainly erred in allowing the use of stun belts during the jury voir dire proceedings.3
We also can find no error in the trial court’s denial of appellants’ ineffective-assistance claims premised on their counsels’ failure to object to the use of stun belts. The cases on which appellants rely to establish that stun belts should be used only where a court has made on-the-record findings justifying their use and that stun belts constitute physical restraints, not only are from other jurisdictions, but also were decided after appellants’ trial, which took place in 2000.4 We cannot conclude that appellants’ trial counsel were deficient for failing to anticipate the holding of those cases. Moreover, as the trial court implied on the basis of testimony by Mr. Tun (appellant Mungo’s trial counsel) at the 23-110 hearing, defense counsel’s willingness to accept the use of stun belts in conjunction with voir due in the jury room was a tactical decision by counsel, whose objective was “for [appellant Mungo] to be able to participate in the jury selection process” instead of remaining at counsel table in the courtroom during the voir dire process. Peete testified at the 23-110 hearing that he had the same objective, telling his counsel that he “wanted to be involved” and “wanted to hear what was being said” during jury selection.
Even if arguendo appellants’ trial counsel’s failure to object to use of the stun belts or to insist on procedural safeguards amounted to deficient performance, “[standing alone, the attorneys’ failure to request an inquiry into the justification for the stun belt is not ineffective assistance. Some prejudice is required before a trial counsel’s performance falls below the constitutional minimum.” Wrinkles v. Buss, 537 F.3d 804, 815 (7th Cir.2008) (citing Strickland v. Washington, 466 U.S. 668, 691-92, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984)). On the issue of whether they were prejudiced by their counsel’s failure to object, at the 23-110 hearing, appellants presented testimony by Jay Scott, a cor*1151rectional officer who served on the jury that convicted appellants, who happened to meet appellant Mungo while Mungo was incarcerated, and who told the court that during the voir dire process, he noticed appellants wearing belts that he thought were stun belts.5 Scott testified that this caused him to think that appellants were “some bad criminals” who “had definitely been in the system before.” Asked whether “what happened in the jury room during voir dire affected [his] verdict in any way,” Scott answered, “Yes.” The trial court, however, gave little weight to Scott’s testimony, finding that his “failures of recall” about other details of what he saw during voir dire and where he had seen a stun belt previously “decreased the value of’ his testimony. We have no basis to disturb the court’s credibility finding as to Scott’s testimony, particularly in light of Mr. Tun’s testimony that Mungo’s jacket covered the stun belt he was wearing and that Tun took steps to ensure that the stun belt was not seen because he understood that it could be prejudicial. Moreover, as the trial court emphasized, although Scott answered “yes” to the question from Mungo’s counsel about whether his verdict was affected by what he saw during the voir dire process, he testified on cross-examination that his view was that “[i]f you don’t have the evidence, ... then it’s no way you should go to jail, whoever you are,” and that he did not make up his mind about appellants’ guilt or innocence until he heard the evidence. In light of that testimony, the trial court did not abuse its discretion in concluding that appellants failed to show prejudice from their counsel’s failure to object to the “possible visibility of the stun belts.”
Appellants also argue that they were prejudiced by their counsel’s omissions because the stun belts instilled fear in them, distracted them, and chilled them ability to participate in their defense. They testified that they were told that, if the stun belts were activated, electrical volts would cause them to defecate and urinate on themselves. Counsel Tun testified at the 23-110 hearing, however, that appellant Mun-go did talk with him about which jurors to select and which jurors he did not like.6 Similarly, the transcript of the jury-selection proceedings contains explanations to the court by appellant Peete’s counsel, Mr. Beaman, that he sought to strike a juror because of the “feelings of our clients with respect to their having viewed him and their having made a gut instinct call” and by Mungo’s counsel that he struck the same juror at his “client’s insistence.” Further, at the 23-110 hearing, Peete was able to describe in detail why jurors were stricken during the voir dire process, evidencing his ability to concentrate and follow the proceedings despite having been outfitted with the stun belt. Because the record shows that appellants were able to consult with their lawyers while wearing the stun belts, we cannot conclude that the trial court erred in concluding that appellants failed to show prejudice from their counsel’s acquiescence in use of the stun belts.7 Cf. Martin v. Sec’y, Dep’t of Corr., *1152No. 08-14308, 2009 WL 3073253, at *7, 2009 U.S.App. LEXIS 21386, at *22-*25 (11th Cir. Sept. 28, 2009) (holding that trial court correctly found that Martin failed to meet the prejudice prong of Strickland where, although Martin testified that wearing a stun belt affected his decision about whether to testify at his trial and prevented him from being able to communicate with counsel during trial, he also admitted that his decision not to testify was “based largely on his lengthy criminal record” and his counsel testified that Martin “communicated regularly with him throughout trial and would often offer suggestions or point out something about a witness’s testimony”).
III. Batson Challenges
During jury selection, Peete’s counsel objected that the government had used eight of its ten peremptory challenges to strike African-American jurors. Noting that the strikes were “consistent with the panel” — the jury venire was about 75 to 80% African-American and about 20% white — the prosecutor argued that the defense had failed to make out a prima facie case of discrimination. The court agreed. Appellants now argue that the trial court erred in not requiring the government to provide race-neutral reasons for its strikes. On this record, we cannot agree. Before a party can be required to justify its peremptory strikes, the party bringing the Batson challenge must establish a prima facie case of discrimination by citing strikes of jurors of a particular race that are disproportionate to their presence in the venire or by pointing to disparate treatment of jurors who are similarly situated except as to race. See, e.g., Miller-El v. Dretke, 545 U.S. 231, 239, 125 S.Ct. 2317, 162 L.Ed.2d 196 (2005); Robinson v. United States, 890 A.2d 674, 682-85 (D.C. 2006). Here, as in Robinson, the defense “cited the fact that eight of the prosecutor’s ten initial strikes were against potential jurors who were black, but that statistic signifies nothing unless it is compared to the proportionate representation of black jurors in the entire panel.” Id. at 685. We agree with the government that the court was not plainly wrong in finding that appellants’ bare-bones challenge to “every peremptory strike of a black juror” did not raise an inference of purposeful racial discrimination and that the government therefore was not required to provide race-neutral reasons to justify its strikes. See Epps v. United States, 683 A.2d 749, 753 (D.C.1996); Jefferson v. United States, 631 A.2d 13, 20 (D.C.1993) (“[I]f the defendant’s only supporting contention is that the resulting jury was disproportionately composed, the defendant must show the composition of the venire or jury.”).
The government raised its own Batson challenge, noting that defense counsel used seven of their ten strikes against white jurors, who made up only about 20% of the venire. After the defense provided explanations for the strikes, the court overruled the government’s challenge as to all but one juror, juror number eight. In explaining why the defense struck juror number eight, Peete’s counsel cited the “feelings of our clients with respect to their having viewed him and their having made a gut instinct call....” Mun-go’s counsel explained that the juror “live[s] in the Georgetown area” where people are “upperly [sie]-mobile” and tend “to be more conservative than those are in the middle class.” Noting that Batson and its progeny make the test “not whether race is a predominant factor but whether *1153race played any significant role,” the court upheld the government’s challenge as to juror eight. Though observing that the matter was “close,” the court found that race did play a significant role in the strike and that juror number eight would not have been stricken if he were black. Thus, the court re-seated the juror. Appellants argue that this was error and that they are entitled to a new trial on this basis.
A trial court’s finding about discriminatory intent is entitled to “great deference,” Hernandez v. New York, 500 U.S. 852, 364, 111 S.Ct. 1859, 114 L.Ed.2d 395 (1991), and we will reverse only when it is plainly wrong or without evidence to support it. Epps, 683 A.2d at 753. Here, we cannot say that the trial court was plainly wrong in concluding that race was a factor, among other factors, in the defense strike of juror number eight. The court could reasonably interpret the defense explanations about “gut instinct” reactions to this juror and about the significance of his residence in Georgetown as proxies, for race. See id. at 753 (“As a direct observer of defense counsel’s conduct and demean- or, the trial judge was in a unique position to determine his credibility. The judge’s finding is one that this court cannot make and, given the record before us, cannot overturn”). Even though the defense set forth some race-neutral reasons, juror number eight spoke to the court again and acknowledged that he had done some radio commentaries in favor of stronger gun-control laws and was in favor of banning guns completely, and the defense renewed their request to strike him on that basis, the court could reject this belated explanation for a strike (particularly in light of the prosecutor’s comment that the defense had declined to strike other jurors who expressed strong feelings about gun control). See Miller-El, 545 U.S. at 246, 125 S.Ct. 2317 (reasoning that explanations that “reek[ ] of afterthought” are “evidence suggesting that the explanation is a sham and a pretext for discrimination”) (citations and internal quotations omitted). Moreover, juror number eight stated that he would be able to “listen to all the facts and make a judgment on the basis of what the facts are, what you learned in the case.”
IV. Autopsy Testimony
The autopsies of Powell and Isaac were performed by then-Deputy Medical Examiner Joseph Garceau. Prior to appellants’ trial, Garceau was fired from his job for practicing without a valid medical license; his Georgia license had expired several days before he began work for the Medical Examiner’s Office. Garceau did not testify at appellants’ trial. Instead, the government called Chief Medical Examiner Jonathan Arden to testify about the cause of death and to rebut the defense theory that the actual shooter was someone who was standing on the passenger side of the car in which the victims were seated (not on the driver’s side, as the eyewitnesses testified). Dr. Arden testified on the basis of death certificates, photographs, x-rays, police reports, and notes made by Dr. Garceau. Supported by amicus curiae Public Defender Service, appellants argue that the admission of Dr. Arden’s testimony and of the autopsy records, without the live testimony of the medical examiner who performed the autopsies and prepared the records, violated their rights under the Sixth Amendment Confrontation Clause. They characterize the autopsy report as a “pretrial statement ] that declarant ] would reasonably expect to be used prosecutorially,” Crawford v. Washington, 541 U.S. 36, 51, 124 S.Ct. 1354, 158 L.Ed.2d 177 (2004), and they argue that autopsy reports cannot be distinguished from drug-analysis reports that the Supreme Court has held are testimonial and subject to the requirements of the Confrontation Clause. See Melendez-*1154Diaz v. Massachusetts, - U.S. -, -, -, 129 S.Ct. 2527, 2532, 2538, 174 L.Ed.2d 314 (2009) (holding that certificates-of-analysis of drugs are within the “core class of testimonial statements” as to which the protections of the Confrontation Clause apply, and noting that “whatever the status of coroner’s reports at common law in England, they were not accorded any special status in American practice” such that they should be admissible without an opportunity for confrontation); Roberts v. United States, 916 A.2d 922, 938-39 (D.C.2007) (holding that appellant was erroneously denied the right to cross-examine witnesses against him where expert witness’s testimony that appellant could not be excluded as a DNA contributor rested on conclusions reached by a team of others who did not testify but performed the actual laboratory analysis set forth in a report that the expert reviewed). Appellants also argue that their counsels’ respective failures to object to Dr. Arden’s testimony constituted ineffective assistance of counsel, entitling them to relief.
We review appellants’ direct-appeal claim only for plain error. We can assume without deciding that Dr. Garceau’s autopsy notes, which were admitted as substantive evidence without any limiting instruction, were testimonial and that the error in admitting them without Dr. Garceau’s live testimony is (now) plain.8 That assumption is unavailing to appellants, however, because they cannot satisfy the remaining prongs of the plain-error test. We are unpersuaded that Dr. Arden’s testimony based on the documents that we assume are testimonial affected appellants’ substantial rights, i.e., that there is “a reasonable probability that the Confrontation Clause violation had a prejudicial effect on the outcome of [appellants’] trial.” Thomas v. United States, 914 A.2d 1, 21 (D.C.2006). While Dr. Arden testified that the death certificates listed the cause of each victim’s death as multiple gunshot wounds, the cause of death was not in dispute, as counsel agreed on the record. Dr. Arden testified on the basis of a photograph (Government Ex. 403) that Powell (who was seated in the driver’s seat of the car) had entrance gunshot wounds in the left side of his neck and gunshot exit wounds on the right side of his neck, which is consistent with the shooter having been on the driver’s side of the car and inconsistent with bullets having gone through Isaac (who was seated in the front passenger seat) into Powell. Dr. Arden also testified from a photograph (Government Exhibit 224) that victim Isaac had an entrance gunshot wound on the back of his head. Although much of the rest of his testimony was on the basis of diagrams prepared by Dr. Garceau (that we assume are testimonial), the testimony was largely cumulative of the points that Dr. Arden made on the basis of the non-testimonial photographs.9 Indeed, the defense lawyers sought to preclude the government from introducing the graphic autopsy photographs, arguing that “[t]he information that can be demonstrated by the photographs is also demonstrated by the diagrams.” Cf. Lewis v. United States, 930 A.2d 1003, 1011 (D.C.2007) (out-of-court statement was not prejudicial to the defense where it was “merely cumulative” of other identification evidence).
Further, admission of the autopsy notes and Dr. Arden’s testimony based on them *1155did not undermine the fairness of the proceedings. Peete’s counsel elicited testimony from Dr. Arden about the absence of soot or stippling, which was based on Dr. Garceau’s autopsy notations. Far from questioning the reliability of the notations, Peete’s counsel sought to use them to defense advantage. He also sought from Dr. Arden testimony about the time of death shown on the death certificates to aid the defense. See Lewis, 930 A.2d at 1011 (rejecting Confrontation Clause challenge where “defense counsel made full, intentional use of the statement” in question in an attempt to discredit government witness). In addition, appellants could have subpoenaed Dr. Garceau to testify had they wanted to challenge his findings. In these circumstances, “appellants] cannot claim that fairness requires that [they] nonetheless be given a chance to contest [the autopsy reports] now.” Thomas, 914 A.2d at 23 (reaching that conclusion after observing that Thomas “could have subpoenaed and cross-examined the [DEA] chemist if he doubted her findings, qualifications, or methodology” and “never disputed the accuracy of the chemist’s report”); Veney v. United States, 936 A.2d 811, 831-32 (D.C.2007) (fourth prong of plain-error test not met where appellant “has not suggested how cross-examination ... would have bolstered his defense”).
We turn next to appellants’ ineffective-assistance-of-counsel claim based on their counsels’ failure to object to admission of the autopsy records and Dr. Arden’s testimony. As the trial court observed, appellants’ trial was pre-Crawford (and pre-Thomas). And, as we have held previously, “we cannot say that trial counsel’s failure to anticipate our decision in Thomas ... fell below prevailing professional norms, or was enough to overcome the presumption that counsel rendered reasonable professional assistance.” Otts v. United States, 952 A.2d 156, 165 (D.C.2008) (citation and internal quotation marks omitted) (amended Apr. 24, 2008). In addition, for reasons already discussed, appellants cannot show that but for their counsels’ failure to object to admission of the autopsy reports and Dr. Arden’s testimony based on them, “the result of the proceeding would have been different.” Strickland, 466 U.S. at 694, 104 S.Ct. 2052.
Appellants contend, however, that their counsels’ omissions with respect to Dr. Arden went beyond a failure to raise Confrontation Clause objections. They rely on the fact that Dr. Garceau was not licensed at the time he performed the autopsies (a fact they assert their counsel should have known, as it was reported in the press), and they emphasize a statement by the prosecutor, at the 23-110 hearing, that Garceau’s work was sometimes “sloppy.” They argue that their counsel provided deficient representation by failing to cross-examine Dr. Arden about the circumstances of Dr. Garceau’s dismissal. Although appellants assert that this information was an “exculpatory fact” that would have led to a “devastating cross-examination,” they do not suggest how the cross-examination would have assisted their case. We agree with the trial court that appellants have not shown how the “lapse in [Garceau’s] license could have affected the credibility of the testimony ... to a degree that would undermine confidence in the verdict.” Moreover, as the trial court noted in its February 11, 2008 Order, the parties stipulated at the beginning of the 23-110 hearing that Dr. Garceau was a trained and experienced forensic pathologist whose work Dr. Arden found to be “professionally done.”
V. “Other Bad Acts” Evidence
Peete cites as error the trial court’s admission of “other crimes” and “bad acts” evidence, which he contends had a prejudicial effect far in excess of any probative value. Specifically, he contends that the *1156court erred by allowing (1) testimony that Peete was known to carry a nine millimeter semiautomatic gun -within the time frame of the murders at issue, and that Mungo was known to have a .357 revolver; (2) testimony about a shooting involving an associate of Peete’s that took place two years after the murders of Powell and Isaac; and (3) testimony that appellants failed to attend the decedents’ joint funeral. We discern no abuse of discretion in the court’s rulings allowing this evidence.
The court permitted the evidence about appellants carrying the described firearms only after the prosecutor explained that the evidence would be limited to the spring or summer of 1997, a time period close to the murders; that the evidence would come in through witnesses who would describe earlier gun incidents as related to the motive for the shootings of Powell and Isaac; and that the evidence would show that Peete had access to the same make and model gun that witnesses described him as having fired on the night of the murders, while Mungo had access to a revolver that would not have left any shell casings and could have been the source of bullet fragments of unidentified caliber found at the crime scene. The court did not abuse its discretion in finding that the evidence had strong probative value and was not propensity evidence prohibited by Drew v. United States, 331 F.2d 85, 89 (D.C.Cir.1964) (holding that evidence of another crime is inadmissible to prove disposition to commit the crime charged). “Drew does not apply at all where such evidence (1) is direct and substantial proof of the charged crime, (2) is closely intertwined with the evidence of the charged crime, or (3) is necessary to place the charged crime in an understandable context.” Jackson v. United States, 856 A.2d 1111, 1115 (D.C.2004) (citation omitted). Particularly relevant here, “[a]n accused person’s prior possession of the physical means of committing the crime is relevant and admissible because it offers some evidence of the probability of his guilt.” Gamble v. United States, 901 A.2d 159, 170 (D.C.2006) (citation and internal quotation marks omitted).
The trial judge initially sustained Peete’s objection to admission of testimony by government witness Calvin Bunch about a shooting incident between Bunch and Peete’s associate, David Thomas, after the murders of Powell and Isaac. The government called Bunch primarily to testify that, while he and Mungo were both incarcerated during 1999, Mungo asked him to murder eyewitness Marvin Kear-ney. The government explained that Bunch had been an associate of victim Powell, and after Powell’s murder, Bunch continued “beefing” with associates of Mungo. Anticipating that the defense would try to impeach Bunch by suggesting that it was not credible that Mungo would hire an enemy to kill a government witness for him, the government sought to have Bunch testify about Mungo’s efforts, through a phone call from the jail, to end the hostility between Bunch and Thomas that had led to the 1999 shooting. The court expressed skepticism about the relevance of the testimony and ruled that the government could not elicit the testimony during its case-in-chief. Later, satisfied that Peete had opened the door by attacking Bunch’s credibility in the way the government had anticipated, the court permitted the government to elicit testimony by Bunch that Thomas shot him and that Mungo arranged a telephone call between Bunch and Thomas to resolve the “beef’ at about the same time that Mungo asked Bunch to kill Kearney. As the government explained, the point of the testimony was to show that Mungo’s efforts to patch up the beef “allows [Bunch] to go out and do it [murder Kearney]. If the enemy comes at [Kearney] they will see him com*1157ing. If his Mends, they won’t see him coming.... It gives him at least a little more leeway to be around there and do what he has been asked to do.” The court — grasping the government’s reasoning even before the prosecutor gave the foregoing explanation — was persuaded that, in context, the disputed testimony was relevant to show that Mungo sought “to make it possible for [Bunch] to go out and do what he was asked to do.” The ruling on relevance was committed to the sound discretion of the court. Clayborne v. United States, 751 A.2d 956, 963-64 (D.C.2000). Especially given the trial court’s careful approach and its understanding of the evidence, we cannot find an abuse of discretion in the court’s assessment that the evidence in dispute was more probative than prejudicial. See Ebron v. United States, 838 A.2d 1140, 1152-53 (D.C.2003) (evidence tending to show that a defendant procured and participated in threats to a testifying witness is highly probative of consciousness of guilt).
The government argues that the evidence about appellants’ non-attendance at the victims’ funerals evinced consciousness of guilt. Appellants contend that this evidence had little probative value, but painted a picture of them as thoughtless, uncaring individuals. Even assuming that appellants have the better of this argument, in light of the eyewitness testimony the government presented (eyewitness testimony that the trial court described as “[s]trong”), we can say with “fair assurance ... that the error did not sway the verdict.” Id. at 1150.
VI. Defense Counsel’s Demeanor
During the trial, the jury sent a note to the judge stating, “Please ask the defense attornies [sic] to watch their court demeanors when focusing on the jury.”10 The prosecutor had previously alerted the court that both defense counsel were making faces in response to government -witness testimony and exhibiting other inappropriate behavior. Appellants argue that the court erred by failing to conduct an inquiry of the jurors to determine whether the attorneys’ behavior had prejudiced appellants, by giving no curative instruction, and, in the alternative, by failing sua sponte to declare a mistrial.11 Appellants also cite their attorneys’ behavior as an additional basis of their ineffective-assistance-of-counsel claim, arguing that their counsels’ lack of self-control throughout the trial so prejudiced them that they were denied a fair trial. They contend that the trial court abused its discretion in rejecting their ineffective-assistance claim as premised on their counsels’ demeanor.
“The decision to grant a mistrial has always been committed to the sound discretion of the trial court and as such, on appeal, a decision should be reversed only in extreme situations threatening a miscarriage of justice.” Lewis, 930 A.2d at 1008 (citation and internal quotation marks omitted). Here, we can find no such extreme situation warranting reversal. Testifying at the 23-110 hearing, the prosecutor explained that he interviewed jurors after the trial and learned that “they seemed to be very impressed with Mr. Beaman” and that Beaman “seemed to *1158resonate with a lot of the jurors.” The prosecutor further explained that, during the trial, he approached the bench about Beaman’s gesturing because he felt that Beaman “had a way of connecting with jurors and [the prosecutor] did not want the jurors to be drawn to his reactions.” In his testimony at the 23-110 hearing, appellant Peete recalled that the judge in appellants’ first trial had also warned Mr. Beaman about his conduct. That tidal ended in a hung jury (the jurors’ note explaining to the court that some jurors “firmly believe that [appellants] should be acquitted”). We cannot say that a miscarriage of justice was threatened because counsel continued such demonstrative behavior during the second trial, or that appellants were prejudiced by the court’s not having given a curative instruction12 or inquired into the effect of the attorneys’ demeanor on the jury.
As to appellants’ ineffective-assistance-of-counsel claim, counsel Tun explained in his testimony at the 23-110 hearing that the defense attorneys’ gesturing to the jury was tactical, a way to convey to jurors a “lack of belief of credibility in the witness’s answer.” Tun explained that “some judges believe that [such gestures] won’t help the defense. Some judges believe that it will help.” He testified that he had won three out of four previous cases in which a judge directed him to refrain from making further facial expressions. Although we can readily agree with appellants that the defense attorneys’ deliberate demeanor was highly improper and unprofessional as trial strategy, the trial court reasonably concluded that it did not amount to deficient representation.
VIL Aiding-and-Abetting Instruction
During its deliberations, the jury sent a note asking “[h]ow does aiding and abetting fit into first or second degree murder?” In response, the court repeated the legally erroneous 13 “natural and probable consequences” instruction that it had given earlier, i.e.: “It’s not necessary that the defendant ... have had the same intent as the principal offender had when the crime was committed or that he have intended to commit the particular crime committed by the principal offender. An aider and abettor is legally responsible for the acts of the other persons that are the natural and probable consequences of the crime in which he intentionally participates.” Peete argues that the erroneous instruction permitted the jury to convict him of first-degree murder without proof that he acted with premeditation to kill the victims. He asserts that the jury “obviously was confused by the aiding and abetting instruction” since it sent a note asking for clarification and then reached its verdict about an hour after it was re-instructed.
Peete did not object to the instruction at trial. To the contrary, both defense counsel urged the court to re-read it in response to the jury’s question. Thus, our plain-error review standard applies. Although it is now plain that the instruction was erroneous, Peete cannot satisfy the remaining prongs of the plain error test. We agree with the government that “[a]ny impartial trier of fact who credited the prosecution’s evidence would ... be bound *1159to conclude” that Peete — who had been “beefing” with the victims, who brought a semiautomatic pistol to the scene, and whom each eyewitness identified as the shooter — acted as a principal rather than as an aider and abettor and did so with premeditation, deliberation and a specific intent to kill. See Wilson-Bey, 903 A.2d at 846-47. The jury, which convicted Mungo of only second-degree murder, obviously found that Mungo played a lesser role in the murders. Peete cannot show that the erroneous instruction affected his substantial rights,14 because there is no reasonable possibility that the jury relied on the aiding-and-abetting instruction to convict the principal Peete of first-degree murder. See Fortson v. United States, 979 A.2d 643, 659 (D.C.2009).
VIII. Other Bases for Appellants’ Ineffective-Assistance Claims
Appellants’ 23-110 motions raised a plethora of other complaints about their trial counsels’ performance: counsel Beaman’s alleged failure to visit Peete in jail with sufficient frequency, to prepare adequately for trial, and to do adequate investigation; Beaman’s failure to file certain pre-trial severance and suppression motions; Bea-man’s failure to call certain exculpatory witnesses; statements Beaman made that let the jury know that Peete was incarcerated during trial; Beaman’s having been suspended from the practice of law for non-payment of dues during the trial; and counsel Tun’s failure to cross-examine Dr. Arden.
The trial court discussed each of these claims in detail in its February 11, 2008 Order and found that appellants failed to specify any exculpatory information that counsel failed to elicit; that appellants’ own testimony showed that Beaman did conduct pre-trial investigation; that Bea-man was not ineffective for failure to file meritless motions; that the record revealed numerous reasons why counsel could reasonably have decided, as a tactical matter, not to call the purported “exculpatory” witnesses15; that Beaman’s elicitation of cross-examination testimony that revealed that Peete was incarcerated was a tactical decision to attack the credibility of government witness Yolanda Epps’s testimony that she conspired with Peete to commit perjury with the fact that she knew that a telephone call from Peete was being recorded because it was placed from the D.C. Jail; that appellants made no showing as to how Beaman’s suspension influenced the outcome of the trial; that Tun’s decision not to cross-examine Dr. Arden reflected a tactical decision to get the medical examiner “off the witness stand as fast as possible;” and that there was no reasonable possibility that the outcome of the trial would have been different if the alleged omissions by counsel had not occurred. As the court’s thoughtful findings are supported by the record testimony and by the court’s own observations at the trial and at the 23-110 hearing, and as they reflect the court’s “well nigh unassailable” credibility determinations, McCraney v. United States, 983 A.2d 1041, 1061 (D.C.2009), we have no basis to disturb them.
*1160For all the foregoing reasons, the judgments of conviction and the order denying appellants’ 23-110 motions are
Affirmed.