*364OPINION OF THE COURT

Criminal trials naturally provoke an excess of emotion. This may lead to potential disruption by spectators at trial. Nonetheless, it is the obligation of trial courts to protect a defendant’s right to a fair trial, and to ensure that conduct by spectators does not impair that right. On this appeal, we conclude that although the trial court should have taken action when defense counsel objected to T-shirts worn by certain spectators that bore a photograph of the deceased victim, defendant was not deprived of a fair trial.

I.

Defendant’s conviction stems from the shooting of two roommates, Mark Maldonado and Leo Walton, in their Brooklyn apartment. Maldonado testified at trial that defendant was angry with him because after defendant and Maldonado had been arrested together for shoplifting, Maldonado was released on bail, but defendant remained in jail for several months. Maldonado attempted to explain to defendant that he had tried to secure defendant’s release, but defendant was skeptical.

On March 20, 2008, defendant asked Maldonado if he could stay the night at Maldonado’s apartment, and Maldonado agreed. When they arrived at the apartment, Maldonado’s roommate, Walton, was at home. Maldonado made defendant a drink and left defendant in the living room of the apartment with Walton. Maldonado went into the bedroom with his girlfriend and locked the bedroom door. After approximately 15 minutes, Maldonado heard three gunshots in the living room. Maldonado told his girlfriend to seek cover next to the bed. Defendant then kicked in the bedroom door and shot Maldonado once in the head. Maldonado attempted to seek shelter behind a closet door, but defendant shot him three more times, striking him in his chest and legs. Defendant then fled the apartment.

Walton had been shot three times in the back of the head and died from his injuries. Maldonado survived after receiving medical treatment. The People’s proffered motive at trial was that defendant shot Maldonado over his anger at being left in jail and killed Walton to eliminate a witness.

*365Defendant was apprehended two days later, and he gave oral, written, and videotaped statements to police. In those statements, he claimed that he had heard rumors that Maldonado believed him to be a “snitch” and wanted to kill him. When defendant confronted Maldonado about the rumors, Maldonado assured him that they were not true and invited defendant back to his apartment. Defendant stated that once they arrived there, Maldonado fired multiple shots at defendant with a .22 caliber handgun, but defendant ducked, and the shots hit Walton instead. According to defendant, he then got up from the floor, pulled from his waistband the .380 caliber handgun he was carrying, and followed Maldonado into the bedroom, where he fired at Maldonado four times.

At trial, defendant raised a justification defense. The People presented evidence at trial, however, that was inconsistent with that defense. According to that evidence, the bullets recovered from the bodies of Walton and Maldonado, as well as the bullets found at the scene, were all fired from a .22 caliber handgun and could not have been fired from a .380 caliber weapon. All seven shell casings found at the scene were also fired from the same .22 caliber handgun.

On the last day of trial, after defense counsel’s summation, counsel asked for a sidebar and noted that three members of Walton’s family who were observing the trial were wearing T-shirts bearing Walton’s photograph and the phrase “Remembering Leo Walton.” Counsel asked that the spectators be required to change their shirts and argued that they were trying to “inflame” or “influence” the jury. The prosecutor opposed the request.

The court refused to instruct Walton’s family members to remove the shirts. The court noted that the spectators were seated quietly and had not drawn attention to themselves or their shirts. The court also stated that Walton’s family members had worn the shirts on previous occasions during the trial but that counsel had not brought the shirts to the court’s attention or requested any relief on those dates. When counsel protested that the shirts had not been worn before that day, the court found “as a matter of fact that one of the females has worn this shirt for at least three court dates.” The court characterized defense counsel’s application as a “disingenuous” attempt to gain a strategic advantage before the People’s summation.

*366The jury found defendant guilty of murder in the second degree for the death of Walton and assault in the first degree for the shooting of Maldonado. Before sentencing, defendant moved pursuant to CPL 330.30 to set aside the verdict. Defendant argued, among other things, that by wearing the T-shirts, Walton’s family members had attempted to improperly influence the jury. The court denied the motion at sentencing, noting that the family members were seated in the second row of the gallery, that they had not called attention to themselves in any way, and that most of the family members were wearing an outer garment on top of the T-shirt.

The Appellate Division affirmed, with one Justice dissenting (.People v Nelson, 125 AD3d 58 [2d Dept 2014]). The Appellate Division was troubled by the trial court’s failure to alert counsel to the issue when it first noticed the shirts, and stated that the “better course would have been to immediately inform Walton’s family members that their conduct could potentially imperil the legitimacy of the trial, and give them an opportunity to voluntarily acquiesce to defense counsel’s request, thus obviating the need for explicit direction from the trial court” (id. at 63). The Court nevertheless declined to create a per se rule requiring reversal “whenever a spectator brings a depiction of a deceased victim into a courtroom” because “each particular instance of challenged conduct calls for a sui generis determination of its potential effect on the jury, made in light of the particular circumstances of the case” (id.). The Court held that, under the particular circumstances of the case, the trial court’s determination “that the spectator conduct did not threaten the ability of the jury to remain impartial” was not error (id. at 64).

The dissenting Justice agreed “that a per se rule compelling reversal in every case involving such a display is not tenable” (id. at 67 [Dickerson, J., dissenting]). The dissent disagreed, however, with the conclusion that defendant was not deprived of his fundamental right to a fair trial under the circumstances (see id. at 70-71).

The dissenting Justice granted defendant leave to appeal to this Court (2015 NY Slip Op 68346[U] [2d Dept 2015]). We now affirm.

II.

We first address the threshold issue whether defendant’s contention is properly preserved for our review. Defendant *367contends that the trial court should have taken action not only upon defense counsel’s objection, but also when the trial court first noticed the shirts, before counsel objected. Defendant asserts that this latter part of his contention is preserved for appellate review pursuant to CPL 470.05 (2), which provides that a question of law is presented when, “in re[s]ponse to a protest by a party, the court expressly decide [s] the question raised on appeal.”

Defendant confuses the trial court’s factual observations with a legal ruling. Upon defense counsel’s objection to the shirts, the trial court made a factual observation that at least one member of Walton’s family had worn the shirt on previous days. The trial court did not make a legal ruling that it had no obligation to act on those previous dates, nor did defendant argue that the trial court was obligated to take action sua sponte. Defendant did not, for example, move for a mistrial on the ground that the wearing of the shirts on previous days had deprived him of a fair trial. The language of CPL 470.05 (2) upon which defendant relies therefore is inapplicable.

Trial courts have the inherent authority and the affirmative obligation to control conduct and decorum in the courtroom, in order to promote the fair administration of justice for all (see generally Matter of Katz v Murtagh, 28 NY2d 234, 238-240 [1971]; People v Mendola, 2 NY2d 270, 276 [1957]; People v Jelke, 308 NY 56, 63 [1954]; 22 NYCRR 100.3 [B] [2]). Furthermore, “one accused of a crime is entitled to have his guilt or innocence determined solely on the basis of the evidence introduced at trial” (Taylor v Kentucky, 436 US 478, 485 [1978]). It is the duty of the trial court to protect the defendant’s right to a fair trial, and to ensure that spectator conduct does not impair that right, regardless of whether defense counsel has noticed or objected to such conduct. Nevertheless, even where the trial court has an affirmative obligation to take certain action in order to protect a fundamental constitutional right, we have required defendants to preserve any alleged error for appellate review (see People v Alvarez, 20 NY3d 75, 80-81 [2012], cert denied 569 US —, 133 S Ct 2004 [2013], and cert denied sub nom. George v New York, 569 US —, 133 S Ct 1736 [2013]). Application of the preservation rule to spectator conduct provides the trial court with a timely opportunity to correct a problem of which it may not be aware, and also facilitates appellate review of any alleged error, inasmuch as spectator conduct often will not appear on the *368record. Here, defense counsel did not move for a mistrial upon learning that the spectators had worn the shirts on previous occasions, or otherwise argue that the trial court should have taken action sua sponte (cf. id. at 79, 81). This part of defendant’s appellate contention therefore is unpreserved for our review. Defendant did partially preserve his contention for appellate review, however, by objecting to the shirts during summations and requesting that the trial court take action. We therefore review only that part of his contention.

III.

The United States Supreme Court has declined to create, a federal standard for evaluation of spectator conduct claims. In Carey v Musladin (549 US 70 [2006]), the Supreme Court held that “the effect on a defendant’s fair-trial rights of. . . spectator conduct . . . is an open question in our jurisprudence” (id. at 76). The Court reasoned that the test it had established in Estelle v Williams (425 US 501 [1976], reh denied 426 US 954 [1976]) and Holbrook v Flynn (475 US 560 [1986]) for the effect of potentially prejudicial courtroom practices on defendants’ fair-trial rights — whether the practice presents an “unacceptable risk ... of impermissible factors coming into play” — had been applied only to “state-sponsored courtroom practices” (Musladin, 549 US at 75-76 [internal quotation marks omitted]). The Supreme Court observed that it had never applied the Williams and Flynn framework to spectator conduct (see id. at 76). In other words, the Supreme Court left resolution of spectator conduct issues to the state courts.

As the Supreme Court recognized in Musladin, “[r]eflecting the lack of guidance from [that] Court, lower courts have diverged widely in their treatment of defendants’ spectator-conduct claims” (id. at 76). Some courts have applied the Williams and Flynn framework to claims that spectator conduct deprived the defendant of a fair trial (see e.g. United States v Farmer, 583 F3d 131, 150 [2d Cir 2009], cert denied 559 US 1058 [2010]; Norris v Risley, 918 F2d 828, 830-834 [9th Cir 1990]; Overstreet v State, 877 NE2d 144, 158-159 [Ind 2007], cert denied 555 US 972 [2008]; State v Lord, 161 Wash 2d 276, 289-290, 165 P3d 1251, 1258-1259 [2007]). Other courts have considered whether the trial court abused its discretion in responding to spectator conduct, whether the spectator conduct caused the defendant to suffer actual prejudice or deprived the defendant of a fair trial, or some combination of those *369considerations (see e.g. Commonwealth v Sanchez, 614 Pa 1, 39-40, 36 A3d 24, 47-48 [2011], cert denied 568 US —, 133 S Ct 122 [2012]; State v Iromuanya, 282 Neb 798, 827-829, 806 NW2d 404, 432-433 [2011]; Allen v Commonwealth, 286 SW3d 221, 229-230 [Ky 2009]; Lonergan v State, 281 Ga 637, 640, 641 SE2d 792, 794-795 [2007]; State v Speed, 265 Kan 26, 47-48, 961 P2d 13, 29-30 [1998]; State v Braxton, 344 NC 702, 709-710, 477 SE2d 172, 176-177 [1996]; State v Franklin, 174 W Va 469, 474-475, 327 SE2d 449, 454-455 [1985]).

Despite these divergent methods, the common thread in these cases is that courts have refused to apply any per se rule of reversal to spectator conduct. They have consistently declined to hold that any particular category of spectator conduct is so inherently prejudicial that it necessarily deprives the defendant of a fair trial. Even those courts that have relied upon the Williams /Flynn framework have evaluated the particular circumstances of each case in determining whether the spectator conduct was so inherently prejudicial as to deprive the defendant of a fair trial (see Farmer, 583 F3d at 149-150; Norris, 918 F2d at 831-832; Overstreet, 877 NE2d at 158-159; Lord, 161 Wash 2d at 289-291, 165 P3d at 1258-1259).

Whether the trial court should intervene, and what intervention is appropriate, must depend upon the facts and circumstances of each particular case. The trial court may consider such factors as: the particular nature of the spectator conduct at issue; how many spectators are involved; the duration of the conduct; whether the involved spectators have called attention to themselves in some way; where the spectators are seated in the courtroom; whether the jury can see or did see the spectator conduct; whether the involved spectators are part of some recognizable organization or group; whether the spectator conduct is the result of some intentional effort to influence the jury or merely an unintended display of emotion; and whether intervention will correct an ongoing problem or will simply serve to highlight a brief instance of misconduct for the jury. This list is not exhaustive, inasmuch as we do not presume to anticipate all of the various forms of spectator conduct that may occur during any given trial.

If the trial court decides to act, appropriate intervention may include such actions as a curative instruction to the jury, ordering the spectators to remove the display, removal of the offending spectators from the courtroom, or questioning of the jurors to determine whether they were influenced by the spectator *370conduct. If the trial court determines, in its discretion, that the spectator conduct was so prejudicial that no other form of curative action can ensure the defendant’s right to a fair trial, then a mistrial will be warranted. The appropriate action will, of course, be informed by any request from counsel. In deciding whether to intervene and what intervention is appropriate, the trial court’s paramount concerns must be the protection of the defendant’s fundamental right to a fair trial and the court’s obligation to preserve order and decorum in the courtroom.

Appellate courts evaluating a defendant’s contention that the trial court erred in refusing to intervene in spectator conduct, or did not intervene appropriately, should review the trial court’s action or inaction for abuse of discretion. The trial court is best situated to take all the circumstances into account and to determine the appropriate intervention (cf. People v Ming Li, 91 NY2d 913, 917 [1998]; Matter of Plummer v Rothwax, 63 NY2d 243, 250 [1984]).

One factor that the trial court should not consider, however, in deciding whether and how to intervene in spectator conduct, is any First Amendment rights of the spectators themselves.

“The court is not a public hall for the expression of views, nor is it a political arena or a street. It is a place for trial of defined issues in accordance with law and rules of evidence, with standards of demeanor for court, jurors, parties, witnesses and counsel. All others are absolutely silent nonactors with the right only to use their eyes and ears” (Katz v Murtagh, 28 NY2d at 240).

No court should tolerate a vocal outburst by a spectator on the ground that the spectator had a First Amendment right to express his or her views on the proceedings. The court similarly should not entertain such concerns when the spectator conduct is nonverbal (see Musladin, 549 US at 79 [Stevens, J., concurring]).

IV.

We now turn to the specific spectator conduct at issue in this case: spectator displays of a deceased victim’s photograph. We have held that portraits or photographs of a deceased victim, taken while the victim was alive, are generally inadmissible at trial unless “relevant to a material fact to be proved at trial” (People v Stevens, 76 NY2d 833, 835 [1990]). This is because such photographs may “arouse the jury’s emotions” (id.).

*371A similar risk is presented by images of a deceased victim displayed by spectators in the courtroom, either on their clothing or by some other method. Such depictions may be viewed by the jury as an appeal to sympathy for the deceased victim and the spectators wearing the display, and perhaps as a request to hold the defendant responsible for their loss (see Musladin, 549 US at 83 [Souter, J., concurring]; Meghan E. Lind, Hearts on Their Sleeves: Symbolic Displays of Emotion by Spectators in Criminal Trials, 98 J Crim L & Criminology 1147, 1153-1154 [2008]). We therefore conclude that spectator displays of a deceased victim’s portrait or photograph should be prohibited in the courtroom during trial. Here, the trial court erred in failing to instruct the spectators to remove the shirts or cover them upon defense counsel’s objection.

V.

Nevertheless, we agree with the Appellate Division that a per se rule requiring reversal whenever a spectator displays a photograph of a deceased victim during trial is untenable (see Nelson, 125 AD3d at 63; id. at 67-68 [Dickerson, J., dissenting]). We further decline to apply the Williams and Flynn framework to hold that such displays are necessarily so inherently prejudicial that they require reversal and a new trial in every case (see Musladin, 549 US at 75-76). That framework, at least as traditionally applied, suggests that certain courtroom conduct is so inherently prejudicial that it requires reversal and a new trial whenever such conduct occurs during trial (see Williams, 425 US at 504-505). In other words, that framework presumes both that there was error and that the error cannot be harmless because the defendant has been deprived of a fair trial thereby (see People v Crimmins, 36 NY2d 230, 238 [1975]).

We conclude, however, that although spectator displays depicting a deceased victim should be prohibited in the courtroom during trial, and although the trial court here erred in refusing to intervene upon defense counsel’s request, the error is subject to harmless error analysis. Defendant contends that the deprivation of his right to a fair trial can never be considered harmless. We agree only insofar as there can be no harmless error analysis if an appellate court concludes that spectator misconduct was so egregious and the trial court’s response so inadequate that the defendant was deprived of a fair trial. Where “there has been such error of a trial court. . . *372or such other wrong as to have operated to deny any individual defendant his fundamental right to a fair trial, the reviewing court must reverse the conviction and grant a new trial,” without regard to whether the proof of guilt was overwhelming or whether “the errors contributed to the defendant’s conviction” (Crimmins, 36 NY2d at 238). Here, however, the spectator conduct was not so egregious that defendant was deprived of a fair trial.

A per se rule of reversal is inappropriate in the context of spectator displays of a deceased victim’s image because such displays may vary widely. For example, the display could range from a small button worn on a spectator’s clothing to a life-size image. A trial court’s refusal to intervene in every such display upon defense counsel’s objection is error. However, not every such display requires the drastic remedy of a mistrial, or an appellate reversal. The trial court or the appellate court, respectively, must make that determination based on the unique circumstances of each case.

Under the particular circumstances of this case, we conclude that the trial court’s error in failing to instruct the spectators to remove or cover the shirts upon defense counsel’s objection is harmless. Consequently, defendant was not deprived of a fair trial.

The evidence of defendant’s guilt is overwhelming. In his statements to the police, defendant admitted that he had shot Maldonado. The People presented evidence that defendant did not shoot Maldonado in self-defense. Defendant admitted that he followed Maldonado into the bedroom after Maldonado had retreated. Moreover, the lock on Maldonado’s bedroom door was damaged, consistent with its having been kicked in, and the bullet holes in and near the closet door corroborated Maldonado’s testimony that he had sought shelter from defendant in the bedroom closet. The forensic evidence corroborated Maldonado’s testimony that it was defendant, not Maldonado, who killed Walton. The bullets recovered from Walton’s body, Maldonado’s body, and the apartment were all fired from a .22 caliber handgun. All seven shell casings found at the scene came from a .22 caliber handgun, and none of the bullets or shell casings came from a .380 caliber handgun. Moreover, all the shell casings originated from the same .22 caliber handgun, corroborating Maldonado’s testimony that only defendant fired a gun in the apartment.

*373Furthermore, there is no significant probability that the trial court’s failure to instruct the spectators to remove or cover the T-shirts upon defense counsel’s request contributed to the verdict. The record reflects that only a few members of Walton’s family were wearing the shirts. The shirt was not particularly inflammatory, and its inscription — “Remembering Leo Walton” — did not ask the jury to convict defendant or otherwise convey a message to the jury regarding the spectators’ beliefs about defendant’s guilt. The spectators did not call attention to themselves or their shirts in any way. Further, the trial court found that the spectators were seated in the second row of the courtroom and that most of them were wearing an outer garment over the shirt, such that the jurors would not have been able to see the shirts in their entirety.

Accordingly, the order of the Appellate Division should be affirmed.

(concurring). This case calls upon the Court to determine the standard applicable to conduct by courtroom spectators that may pose a risk to the defendant’s right to a fair trial. Supreme Court decisions in this area have left us with a “clean slate” upon which to write such a rule (see United States v Farmer, 583 F3d 131, 149 [2009], cert denied 559 US 1058 [2010]).

I concur with the majority that defendant was not deprived of a fair trial. I write separately because I would adopt the standard the Supreme Court applies to state-sponsored courtroom practices that raise similar issues instead of the abuse of discretion standard applied by the majority (see majority op at 369-370). Under this standard, appellate courts would examine the totality of the circumstances to determine whether the spectator conduct at issue presents “an unacceptable risk ... of impermissible factors coming into play” (Holbrook v Flynn, 475 US 560, 570 [1986] [internal quotation marks and citation omitted]). The majority’s approach is thoughtful and, as it applies to these facts, correctly focuses on the trial court’s failure to take remedial measures. An abuse of discretion standard may lead, however, to inconsistent rulings by trial courts and permits harmless error analysis by appellate courts even though spectator conduct may implicate a defendant’s “self-standing” right to a fair trial (People v Crimmins, 36 NY2d 230, 238 [1975]).

The law with respect to courtroom conduct by state actors is clear. In several habeas corpus proceedings, the Supreme Court *374considered whether such conduct in the courtroom led to a defendant being deprived of a fair trial. In Estelle v Williams (425 US 501 [1976]), the Court confronted a defendant’s claim that he was “compell[ed] ... to stand trial in jail garb” (id. at 505). Holbrook v Flynn involved a jury trial where four uniformed law enforcement officers sat “in the first row of the spectators’ section,” ostensibly for security purposes (475 US at 562). The Court’s analysis in both cases was grounded in the presumption of innocence:

“The presumption of innocence, although not articulated in the Constitution, is a basic component of a fair trial under our system of criminal justice. . . .

“To implement the presumption, courts must be alert to factors that may undermine the fairness of the fact-finding process. In the administration of criminal justice, courts must carefully guard against dilution of the principle that guilt is to be established by probative evidence and beyond a reasonable doubt” (Williams, 425 US at 503 [citation omitted]).

Certain conduct, the Supreme Court found, was so inherently prejudicial that it denied the defendant a fair trial. In analyzing whether that constitutional violation had taken place, the Court explained “the question must be not whether jurors actually articulated a consciousness of some prejudicial effect, but rather whether ‘an unacceptable risk is presented of impermissible factors coming into play’ ” (Flynn, 475 US at 570, quoting Williams, 425 US at 505).

Also common in Flynn and Williams was the fact that conduct complained of was perpetrated by state actors, namely, prison or law enforcement officials. A subsequent habeas corpus proceeding, however, involving spectator conduct — trial attendees wearing buttons with the victim’s photograph — did reach the Supreme Court, but the issue of what standard should apply to evaluate that conduct was not addressed because of the procedural posture of the case (see Carey v Musladin, 549 US 70, 76 [2006]). The Court explained it “ha[d] never addressed a claim that such private-actor courtroom conduct was so inherently prejudicial that it deprived a defendant of a fair trial” (id. [footnote omitted]). As a result, the Court determined that the conclusion of the state appellate court in *375the defendant’s underlying criminal action was not “contrary to or an unreasonable application of clearly established federal law as determined by this Court” as required for federal habeas relief (id. at 77; see Farmer, 583 F3d at 149-150 [“Carey v. Musladin . . . left it to lower courts to address claims” based upon “courtroom displays by private actors”]).

Concurring in the judgment in Musladin, Justice Souter asserted that Williams and Flynn evinced an “intent to adopt a standard at [a] general and comprehensive level . . . that . . . reaches the behavior of spectators” (Musladin, 549 US at 82 [Souter, J., concurring]; see also id. at 78-79 [Stevens, J., concurring]). I agree, and would apply the Williams/Flynn standard here.

As the Musladin majority noted, the inquiry in Williams and Flynn asked “whether the practices furthered an essential state interest,” suggesting the standard “applied] only to state-sponsored practices” (id. at 76). Certainly, the question of state interest is a factor with no relevance to the facts and circumstances here. Nevertheless, as Justice Souter concluded, the trial court “has an affirmative obligation to control the courtroom and keep it free of improper influence” whether the improper conduct is by. a state actor “or an individual” (id. at 82 [Souter, J., concurring]).

Accordingly, with respect to conduct of private actors in the courtroom, the same standard should apply to answer the critical question of whether the spectator conduct presented “ ‘an unacceptable risk ... of impermissible factors coming into play’ ” (id. at 75 [citation omitted]).

Answering that question, as the Supreme Court observed in Williams, is challenging:

“The actual impact of a particular practice on the judgment of jurors cannot always be fully determined. But this Court has left no doubt that the probability of deleterious effects on fundamental rights calls for close judicial scrutiny. Courts must do the best they can to evaluate the likely effects of a particular procedure, based on reason, principle, and common human experience” (425 US at 504 [citations omitted]).

Three factors appear most relevant when assessing the prejudicial effect of the conduct at issue. First, courts should examine the nature of the conduct or display and its potential *376to influence the jury verdict (see e.g. Woods v Dugger, 923 F2d 1454, 1458-1460 [11th Cir 1991] [considering prejudice from presence of uniformed off-duty prison guards attending, as spectators, the trial of the defendant for murder of a guard]; State v Allen, 182 Wash 2d 364, 385-386, 341 P3d 268, 278-279 [2015] [“Silent showings of sympathy or support do not pose an unacceptable threat to the defendant’s right to a fair trial so long as the display does not advocate for guilt or innocence”]; People v King, 215 Mich App 301, 305, 544 NW2d 765, 768 [1996] [“We are not persuaded . . . that the wearing of the buttons, which were less than three inches in diameter, . . . could have influenced the panel”]). Second, appellate courts must consider whether the record of the courtroom situation is adequate to facilitate review (see e.g. State v Iromuanya, 282 Neb 798, 823, 806 NW2d 404, 429 [2011]; State v Speed, 265 Kan 26, 48, 961 P2d 13, 29-30 [1998]; Nguyen u State, 977 SW2d 450, 457 [Tex Ct App 1998], affd on other grounds 1 SW3d 694 [Tex Crim App 1999]). Third, and not necessarily determinative, appellate courts should consider the response, if any, by the trial court to the conduct (see e.g. Farmer, 583 F3d at 150 [“Moreover, once defense counsel called the T-shirts to the district court’s attention, the court instructed the government ‘to urge (the spectators) not to come into this courtroom with shirts with the picture’ ”]; People v Houston, 130 Cal App 4th 279, 316, 29 Cal Rptr 3d 818, 848 [2005]; State v Franklin, 174 W Va 469, 475, 327 SE2d 449, 455 [1985]).

Applying the first factor, the offending shirts in this case bore the victim’s photograph and the phrase “Remembering Leo Walton.” Such images could “raise a risk of improper considerations” inasmuch as the photograph and written message could be construed as “an appeal for sympathy . . . and a call for some response” that a juror might interpret to mean “a verdict of guilty” (Musladin, 549 US at 82-83 [Souter, J., concurring]; see generally People v Stevens, 76 NY2d 833, 835 [1990] [“photographs of the victim taken while he or she was alive . . . may . . . arouse the jury’s emotions”]). As other courts confronted with similar displays have noted, however, jurors were just as “likely to have viewed the buttons as signs of grief instead of a collective call for . . . conviction” (Iromuanya, 282 Neb at 828, 806 NW2d at 432). The spectators were silent, evidently few in number, and their T-shirts were partially covered. Moreover, unlike the wearing of law enforcement uniforms by persons in the gallery, these T-shirts gave no suggestion of state approval of the spectator’s message (cf. *377Woods, 923 F2d at 1458-1460 [noting “(a)bout half of the spectators appear to be wearing prison guard uniforms” and “(t)he officers in this case were there ... to communicate a message to the jury” even though “no state interest c(ould) justify the uniformed presence of these off-duty correctional officers” (footnotes omitted)]).

Next, the limited record makes the potential impact of the conduct more difficult to assess. Defense counsel apparently did not notice the shirts at first, and we are left with the trial court’s brief description of the T-shirts and somewhat inconsistent recounting of the spectators’ conduct in court. At summation, upon consideration of defense counsel’s and the prosecutor’s arguments, the trial court determined the conduct was not prejudicial, explaining his reasons on the record. Thereafter, at the CPL 330.30 hearing, the court clarified the record before us:

“the jury was not inflamed by the simple wearing of the [T]-shirts by members of the decedent’s family. They sat in the second row of the audience. I noticed one of the grieving members of the family wearing the shirt . . . several times.

“I guess now it would be appropriate [for me] to make a better record of what the shirt was. It was [a] white [T-shirt] with a silk screen with a picture of the deceased with some written language on it.

“I had noticed that shirt, [but] couldn’t read what was written on it. It was not flauntily displayed in front of the jury, nor in any way did any members of the family bring undue attention to it.

“In fact, most of the members of the family had an outer garment on top of the [T]-shirt. So it wasn’t even capable of seeing the entire thing.”

We know nothing about how well, if at all, the jury could see the T-shirts. Moreover, only four people wore the shirts, which does not on this record amount to “a formidable, albeit passive, influence on the jury” (Franklin, 174 W Va at 474-475, 327 SE2d at 454-455 [noting “from ten to thirty MADD demonstrators remained in court throughout the trial” for an alcohol-related vehicular homicide “and sat directly in front of the jury. Some cradled sleeping infants in their laps and all prominently displayed their MADD buttons”]).

*378Lastly, I agree with my colleagues that “[t]rial courts have the inherent authority and the affirmative obligation to control conduct and decorum in the courtroom, in order to promote the fair administration of justice for all” (majority op at 367). Unquestionably, the better practice here would have been for Supreme Court to have responded to the spectator conduct by taking steps to end the display, but I cannot agree that all spectator displays of a deceased victim’s photograph should be banned outright (see majority op at 370-371). This is not to say that any particular memorial or other display by spectators is permissible, or that certain conduct necessitates a specific response. The risk of prejudice presented by spectator conduct should always be evaluated on a case-by-case basis (see Musladin, 549 US at 83 [Souter, J., concurring]; Iromuanya, 282 Neb at 827, 806 NW2d at 432 [noting that Justice Souter “declined to embrace a per se rule” regarding memorial buttons and instead concluded that the issue in each case is whether the risk is unacceptable]).

In sum, after considering the nature of the conduct and the record presented, and factoring in the trial court’s response, or rather lack thereof, to the conduct, there is no reasonable probability that the conduct by the spectators created an unacceptable risk to defendant’s right to a fair trial. The spectators’ silent display was not overwhelming and seemed to be in the nature of an expression of sympathy, not a play to the passion of the jury. Expressions of grief by a decedent’s family members and loved ones are to be expected — though not necessarily tolerated — during a homicide trial. Trial courts should continue to take measures to address the risks of such conduct and avoid even the suggestion that improper factors may have influenced the jury (see e.g. People v Pennisi, 149 Misc 2d 36, 37, 40 [Sup Ct, Queens County 1990] [trial court determined that ribbon corsages worn by family members of victim and other spectators could not be worn in the courtroom]). Nevertheless, the court’s failure to respond here, while not optimal, is not sufficient grounds for a new trial.

Order affirmed.