(concurring in part and dissenting in part). Malice Green died en route to the hospital, as a result of blunt force trauma to the head1 sustained while the defendant police officers attempted to arrest him. Following a joint trial before separate juries, defendants were found guilty of second-degree murder, MCL 750.317; MSA 28.549, and were sentenced to terms of eight to eighteen years (Budzyn) and twelve to twenty-five years (Nevers). The defendants appealed, first in the Court of Appeals and then in this Court, raising numerous claims of error. I *112would affirm the convictions, finding that none of the issues raised constitute error requiring reversal.
The difficulty of this Court’s decision in this highly publicized and emotionally charged case was not caused by the emotional nature of the case, or by the publicity surrounding it. Nor was it caused by the infusion of race into some of the legal issues presented.2 This case demanded careful examination because the legal questions presented are difficult and close.
The issues of greatest concern to both the majority and to myself relate to whether defendants were denied their right to a fair trial as a result of exposure to certain extraneous influences. In particular, I am most concerned with the jurors’ exposure to the movie Malcolm X and to media accounts of contingency plans of suburban law enforcement in case of a riot. I will examine the issue of jury taint in detail. While I have also closely examined the other issues presented by the defendants, I will not discuss them in depth, finding them either lacking in merit or properly disposed of by the Court of Appeals.
i
FACTS
The Court of Appeals unpublished opinion per curiam sets forth the following factual account of the case:
*113Defendants’ convictions arise from events that occurred on the evening of November 5, 1992, in the city of Detroit, on the street outside of the storefront building apartment of Ralph Fletcher. Defendants were on duty as Detroit City Police Officers, in plain clothes, patrolling the area in an unmarked police car. They were keeping “loose surveillance” on a stolen Ford Tempo automobile in the area. Malice Green was driving Fletcher to his apartment in a similar model automobile, a red Topaz.
Shortly after Green parked his vehicle in front of Fletcher’s home, defendants drove by, stopped, backed up, and parked their police car behind Green’s vehicle. At that time, Theresa Pace, Robert Hollins, and Robert Knox were leaving Fletcher’s apartment from the back door. Defendants initially focused on Green’s vehicle to determine whether it was the stolen Tempo. They then spotted Knox, who appeared to run toward the back of Fletcher’s house. Budzyn then pursued Knox and escorted him back to the area in front of the house. Fletcher and Green were standing beside Green’s vehicle while defendant Nevers was standing beside the parked police car.
When Green was asked for his driver’s license, he reentered his vehicle through the passenger side door and reached for the glove box. As he was sitting in the car sideways with his legs extending outside, Budzyn saw Green drop something, which appeared to be rock cocaine, and hold something in his right hand. When Green failed to obey Budzyn’s orders to open his hand, Budzyn reached in and grabbed Green’s right forearm. A struggle ensued.
According to Fletcher, Pace, Hollins, Knox, and Manuel Brown (who had also been stopped by defendants while walking in the area), Budzyn climbed on top of Green, straddled him as he laid back across the front seat, and hit him repeatedly with his flashlight. These witnesses further testified that Green did not appear to fight back, but continued to clench his right hand, despite Budzyn’s repeated orders and the urging of the bystanders. Budzyn testified before his jury that he never hit or straddled Green. Nevers also testified to his separate jury that he never saw Budzyn hit Green, and although he heard some sounds of impact, *114he denied knowing or believing that those sounds were caused by Budzyn striking Green.
Fletcher, Pace, Hollins, Knox, and Brown observed Budzyn strike Green with his flashlight on the hand or hands; however, not all of these witnesses could see whether Budzyn ever struck Green in the head or elsewhere. Fletcher, who was standing directly beside Green’s car approximately three to five feet away at the time, testified that he only saw Budzyn strike Green on the hand. Fletcher explained, however, that he had stepped away when Budzyn climbed further into the car and therefore was unable to see whether Budzyn ever hit Green in the head or not. Pace and Hollins, who stood a few yards away from Fletcher beside the front of the parked police car, only saw Budzyn hit Green in the hand as well. However, once Budzyn was on top of Green inside the car, they saw Budzyn swing his flashlight up over his shoulder and back down toward Green several times. They also testified that it appeared as though the blows were landing on top of Green’s head, possibly his forehead, from the sounds of impact and the positions of Green and Budzyn. Hollins estimated that Budzyn landed at least six or seven such blows. Brown, who was standing at a somewhat elevated position on the curb behind Green’s car and approximately fifteen feet from the others, testified that he actually observed Budzyn hitting Green in the head with his flashlight approximately ten times.
At some point, Nevers walked over to the passenger side of Green’s car and struck Green’s clenched hand and knees with his flashlight. Nevers retrieved some rock cocaine from the car and told Fletcher and the other citizens nearby to leave, or that they were free to leave. As Nevers walked around to the driver’s side door, Pace and Hollins walked to a parked truck and drove off. Fletcher testified that as he was walking toward his home, he turned and saw Nevers strike Green twice on the head with his flashlight from the open driver’s side door, using a sweeping “golf swing.” Brown saw Nevers strike Green several times from a position where it appeared to Brown as though the blows would have been landing on Green’s head, although he was *115unable to see whether there was actual contact. He estimated that Nevers hit Green approximately 15 times, including the initial hits on Green’s hand and knuckles. Nevers admitted hitting Green in the head with his flashlight a total of five or six times. He explained that he hit Green two or three times when Green grabbed the handle of his sidearm and then struck Green another two or three times when it appeared as though Green was holding a razor blade or knife in his hand.
Two ems ambulance units from the Detroit Fire Department, designated as units “Impact 5” and “Medic 7,” arrived on the scene. Each unit had two medical technicians who testified that Nevers repeatedly struck Green in the head with his flashlight as he lay across the front seat of the car with his head outside the driver’s side door while ordering Green to open his hands and/or to be still. Technicians Albino Martinez and Mithyim Lewis of the Impact 5 unit saw Nevers strike Green in the head a total of four times. Technician Scott Walsh of the Medic 7 unit saw Nevers strike Green with the flashlight once on the head, and later on the chest and abdomen. Technician Lee Hardy of the Medic 7 unit saw Nevers strike Green on the head about ten times and on the left hand four or five times. According to each technician, Green was bleeding heavily and covered with blood. He also was dazed or in a stupor, neither resisting nor responding to Nevers’ commands, but only moaning or muttering incomplete words. Ultimately, Green was pulled from the car by Robert Lessnau, a uniformed police officer who had arrived on the scene. Green was handcuffed while he lay face down in the street in a pool of his own blood. Later, Green suffered a seizure while the medical technicians were rendering treatment and died during the transport to a hospital.
An autopsy was performed by Assistant Wayne County Medical Examiner Kalil Jiraki. There were nine cuts and a bruise on Green’s scalp, and other bruises and cuts on the face and behind the ears. Jiraki opined that these wounds were caused by at least fourteen separate blows to the head. Most of the wounds were consistent with having been made with a cylindrical instrument such as a flashlight. *116Jiraki found no fracturing of the skull, and only minimal swelling of the brain. Although Jiraki found levels of cocaine in Green’s blood consistent with being under the influence of cocaine at the time of death, he opined that those levels were too small to have any significance and had no bearing on Green’s death at all.
Additional expert medical testimony was rendered from defense experts Dr. L. J. Dragovic, Chief Medical Examiner for Oakland County, Dr. Haresh Mirchandani, Chief Medical Examiner of Philadelphia, Dr. Lucy Rorke, as well as the prosecution’s rebuttal witness, Dr. Michael Baden. All of the defense experts disagreed with Dr. Jiraki’s opinion that the cause of death was blunt force trauma alone, without any effect from cocaine, and disagreed with his description of the mechanism or mechanisms of death as well. For example, Dr. Dragovic opined that the mechanism of death was seizure activity triggered as a result of the trauma upon the surface of Green’s head while his brain was “wired up” on cocaine. Dr. Mirchandani opined that Green died of a cardiac arrhythmia caused by a combination of cocaine use, physical exhaustion and trauma from the beating. Although there was some disagreement whether the head trauma alone might have caused death, all of the defense experts agreed that the beating at least played a role in Green’s death. All of the experts, except Dr. Rorke, opined that Green would not have died at that time without the beating. Only Dr. Rorke suggested that Green might have died even without the beating![3]
Defendants were tried together but before separate juries. Robert Lessnau, who was also charged in this case, was granted a bench trial. Following selection of defendant Nevers jury, Nevers’ motion for change of venue based upon pretrial publicity was denied. Ultimately, defendants’ respective jurors found them guilty of second-degree mur*117der, and the trial court sentenced both defendants within the low range of their respective guideline sentence ranges — defendant Budzyn receiving an 8- to 18-year sentence and defendant Nevers receiving a 12- to 25-year sentence. . . . [Issued March 22, 1995 (Docket Nos. 170477, 170478), slip op at 1-3.]
From the outset, the case engendered a flurry of media reports. This was due in part to the similarities between the case, involving white police officers and a detained black motorist, and the California acquittals of white Los Angeles police officers charged in the beating of black motorist Rodney King. Extensive media coverage continued through the trial, including reports speculating on contingency plans in place in the event of postverdict rioting. Jurors on both Budzyn’s and Nevers’ panels were exposed to portions of these reports.
Near the end of the trial, during a two-day recess in the proceedings, the court provided videotape movies for the jurors’ entertainment. One of the movies provided was Malcolm X, which depicts, in its opening credits, footage of the Rodney King beating with accompanying dialogue from a speech by Malcolm X. Both juries viewed at least part of the film. Defendants moved for a mistrial, arguing that this opening footage caused special prejudice and that the film was inappropriate for viewing in the context of this case. This motion was denied.
After the juries rendered their verdicts, the defendants filed motions for new trials, in part on the basis of the potential prejudicial effect on the jury of exposure to the movie Malcolm X and to media reports. This motion was also denied.
*118The defendants appealed their convictions, raising numerous issues including those relating to improper jury instructions, juror taint and bias, various improprieties on the part of the trial judge and the prosecution, and insufficiency of evidence. Finding no errors requiring reversal, the Court of Appeals affirmed the defendants’ convictions. Unpublished opinion per curiam, issued March 22, 1995 (Docket Nos. 170477, 170478). This Court granted leave to appeal.4
n
EXTRANEOUS INFLUENCES; JURY TAINT
Defendants who choose a jury trial have an absolute right to a fair and impartial jury. Duncan v Louisiana, 391 US 145; 88 S Ct 1444; 20 L Ed 2d 491 (1968); People v Tyburski, 445 Mich 606, 618; 518 NW2d 441 (1994). The defendants argue that they were denied this right by a combination of prejudicial publicity surrounding the case and instances of extraneous influences on the jury during the trial. I will focus on what I believe to be the two most troubling of these claims: jury exposure to the movie Malcolm X and to media reports during the trial of existing contingency plans in case of rioting after the verdict.
Before examining these claims of jury taint, I shall review the general principles and analysis applicable to determining claims of extraneous influences on jury verdicts.
*119A
GENERAL PRINCIPLES
Implicit in our jury system is an initial presumption of jury impartiality. See United States v Ruggiero, 56 F3d 647, 652 (CA 5, 1995). Evidence that extrinsic factual matters tainted the jury’s deliberations can defeat this initial presumption. Sheppard v Maxwell, 384 US 333, 351; 86 S Ct 1507; 16 L Ed 2d 600 (1966); Ruggiero, supra at 652, quoting United States v O’Keefe, 722 F2d 1175, 1179 (CA 5, 1983). While the use of postverdict juror affidavits or testimony to impeach a jury verdict is generally prohibited, an exception applies where the jury is exposed to “extraneous influences” that may have improperly influenced its deliberations. Tanner v United States, 483 US 107, 117; 107 S Ct 2739; 97 L Ed 2d 90 (1987), and People v Larry Smith, 106 Mich App 203, 211-212; 307 NW2d 441 (1981).
Thus, postverdict inquiries into extraneous influences on the jury must balance the defendant’s right to a fair trial by an impartial juiy with the long-established common-law rule prohibiting the admission of juror testimony to impeach a jury verdict. See Tanner, supra at 117-127. The purposes underlying the general rule against impeachment of jury verdicts include promoting finality of jury verdicts, discouraging harassment of jurors, reducing incentives for jury tampering, and encouraging free and open discussion among jurors. See Tanner at 119-120. The common-law rule and the exception for extraneous influences has been codified in federal law.5 Michigan follows *120the federal rule. People v Larry Smith, supra at 211-212.
Extraneous influences include publicity received and discussed in the jury room, consideration of evidence not admitted in court, and communications or other contact between jurors and third persons. In contrast, evidence of discussions among jurors, intimidation or harassment of one juror by another, and other intrajury influences do not come within the exception and may not be used to impeach a verdict. United States v Williams-Davis, 821 F Supp 727, 733 (D DC, 1993). In other words, while the exception allows the use of juror testimony or affidavits, the general rule still prohibits probing into the actual deliberative process. Thus, the jurors may testify regarding their exposure to these influences, but they may not relate how such exposure affected the deliberative process.
B
BURDEN-SHIFTING ANALYSIS
Several state and federal cases are instructive regarding how prejudice resulting from extraneous influence may be proved and when such influences require reversal. A review of these cases reveals that *121most courts employ a burden-shifting analysis. Once the defendant comes forward with evidence establishing the existence of an extraneous prejudicial influence, the burden shifts to the government to rebut the presumption of prejudice by showing that the extraneous influence was harmless error. I agree with and would adopt this two-part analysis. Thus, as will be more fully set forth in this section, to determine whether the extraneous influences merit reversal, I would hold, as does the majority, that reviewing courts must first determine whether the defendant has made the required threshold showing that the extraneous materials were prejudicial and must then examine whether the prosecution has shown that the jury’s exposure to the materials was harmless.
l
THRESHOLD SHOWING OF PREJUDICE
Regarding the initial threshold showing required of the defendant, the United States Court of Appeals for the Second Circuit, in a seminal opinion by Judge Henry Friendly, stated that
[t]he touchstone of decision ... is thus not the mere fact of infiltration of some molecules of extra-record matter, with the supposed consequences that the infiltrator becomes a “witness” and the confrontation clause automatically applies, but the nature of what has been infiltrated and the probability of prejudice. [United States ex rel Owen v McMann, 435 F2d 813, 818 (CA 2, 1970).]
Thus, generally, to impeach a verdict, the defendant must first establish the existence of the extraneous influence. Even with this burden met, however, the defendant will not be entitled to a new trial unless *122there is some further showing of prejudice. Williams-Davis, supra at 733; United States v Boylan, 898 F2d 230, 258 (CA 1, 1990).6
The extraneous matters will not meet the initial threshold unless “there is a ‘reasonable possibility’ that such material could have affected the verdict.” Williams-Davis, supra at 734, quoting United States v Ortiz, 942 F2d 903, 913 (CA 5, 1991). Courts have articulated guidance, but have not created “bright line” rules for determining when the real and substantial possibility of prejudice standard is met.
There is no magic formula for deciding whether extrinsic material disclosed in the jury room was reasonably prejudicial. The determination in each case in which such an issue is presented must be evaluated in light of its own unique circumstances. [United States v Gaffney, 676 F Supp 1544, 1553 (MD Fla, 1987).]
The Ninth Circuit and other courts have set forth several factors relevant to determining whether extrinsic materials could reasonably have been prejudicial. For example, in State v Watkins, 526 NW2d 638, 641 (Minn App, 1995), the court considered the following factors, among others, in estimating the probable prejudicial effect of racial slurs on the jury: “(1) the nature and source of the prejudicial matter, (2) the number of jurors exposed to its influence . . . and ([3]) the likelihood that curative measures were effective in reducing the prejudice.”
*123The Ninth Circuit has identified similar factors, including: (1) whether the extrinsic material was actually received and if so how, (2) the length of time it was available to the jury, (3) the extent that the jurors discussed and considered it, (4) whether it was introduced before a verdict was reached and if so at what point during the deliberations, and (5) the length of time the jury deliberates after exposure to the extrinsic material.7 Regarding this last factor, if the jury was deliberating for a significant time before exposure to the extrinsic evidence and then quickly rendered a verdict after the exposure, the information likely affected the verdict. The Ninth Circuit does not consider any one factor as being determinative in any given case. Dickson v Sullivan, 849 F2d 403, 406 (CA 9, 1988).8
2
HARMLESSNESS OF THE EXTRINSIC MATERIALS
Once the defendant has met the burden of coming forward with competent evidence regarding prejudicial extraneous matters, the burden shifts to the state to prove that exposure to the extraneous influences was harmless error. Cases provide guidance regarding how the government may establish harmless error in this context. Exposure to extraneous prejudicial mat*124ters may be deemed harmless if the extraneous material is merely duplicative of evidence introduced in open court, or if the other evidence amassed at trial was so overwhelming that the jury would have reached the same result even without the extraneous material. United States v Caro-Quintero, 769 F Supp 1564, 1574 (CD Cal, 1991); Hughes v Borg, 898 F2d 695, 700 (CA 9, 1990). The Court of Appeals for the Ninth Circuit has stated that such a determination of overwhelming evidence will rebut the presumption of prejudice in this context when “the other evidence adduced at trial [is] ‘so overwhelming that “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.” ’ ” Caro-Quintero, supra at 1575, quoting Hughes, supra at 701. We believe this standard is generally appropriate and essentially comports with the standard for harmless error applicable to errors of constitutional magnitude as adopted by this Court in People v Anderson (After Remand), 446 Mich 392; 521 NW2d 538 (1994).9 Consequently, exposure to prejudicial extrinsic materials is harmless error if the reviewing court may conclude, beyond a reasonable doubt, that the other evidence adduced at trial is so overwhelming that there is no real and substantial possibility that the challenged evidence contributed to the conviction. Anderson, supra at 406.
*125With these guiding principles in mind, I shall consider whether the extraneous influences asserted require reversal of the convictions.
m
MALCOLM X
I will first examine the movie Malcolm X. The first inquiry is whether the defendants have met their threshold burden by bringing forward competent evidence of a prejudicial extraneous influence. Defendants rely, in part, on sealed affidavits of four jurors10 to establish that the jurors had watched the movie Malcolm X during a recess. I do not question the limited use of the signed affidavits to show exposure to this extraneous influence. I also note that existence of this influence was established on the record in the motion for mistrial.
As noted earlier, beyond showing exposure to an extraneous influence, the defendants, absent an unauthorized ex parte communication,11 must make a showing of prejudice reverberating from the extraneous information. The test for determining whether the threshold level of prejudice is met is whether defendants have established a real and substantial possibility that the extraneous matter could have affected the verdict.
*126The first factor in this threshold inquiry is the nature and source of the extraneous material. The defendants focus their claim of prejudice resulting from the showing of Malcolm X on the opening footage of the movie. In an attempt to make his movie more provocative, producer-director Spike Lee presents opening footage of the beating of motorist Rodney King by Los Angeles police officers against the backdrop of a burning American flag. Along with these images, the viewer hears words from a speech by Malcolm X.
Virtually every person in America had been exposed to the footage detailing the brutal beating given Mr. King by members of the Los Angeles Police Department. While it was unfortunate that the jurors were perhaps once again confronted with this footage, is it the defendants’ position that the jurors became inflamed and were engulfed by emotion and thus rendered incapable of rendering a fair verdict? The possibility that a few police officers will absolutely violate department rules and procedures is never far from the minds of African-Americans, no matter their station in life. The presence of racism in any community can, and sometimes does, cause American citizens to react emotionally and render less than clear judgment. But there is no reasonable probability that the Rodney King footage affected the jurors in this case.
Racism has not and will not quell the desire of African-Americans to fully exercise all the rights and responsibilities associated with democracy. If this were not true, then the four-hundred-year-old struggle for freedom and economic opportunity would have been abandoned long ago. The defendants have not *127established that the picture of one black man being beaten by police officers could reasonably have prevented the jurors from rendering a fair verdict.
The words of Malcolm X in the opening sequence are indeed provocative.12 But when these words are taken in the context of the effect of the overall message of the movie, their effect is greatly lessened.13 Please be clear, I do not dismiss the power of words. I do dismiss the conclusion that there is a real and substantial possibility that conscientious jurors watching this movie, shown during a recess purely for entertainment purposes, would have allowed it to affect their verdict.14
Regarding the nature and source of this extraneous influence, the defendants emphasize that the movie was supplied by the court; this factor alone is not determinative. While this fact does tend to bolster the *128movie’s potential prejudicial effect, the nature of this particular extraneous influence weighs heavily against a finding of prejudice. The movie was chosen by the jurors from three others made available to them purely for purposes of entertainment.15 The jurors should be credited with enough sophistication to separate entertainment from evidence.
Another important factor relevant to the threshold inquiry is the timing of the jurors’ exposure to the movie. The jurors were not exposed to this extraneous influence during deliberations, but during a two-day recess near the end of the trial. Extraneous influences during the deliberative process are deemed more prejudicial than those finding their way to the jury earlier in the trial. This fact distinguishes this case from others finding that juror exposure to a movie of a potentially prejudicial nature during deliberations required reversal. See Seekers v State, 35 Ala App 40; 44 So 2d 628 (1949); Gonzales v State, 593 SW2d 288 (Tenn, 1980).16
*129Other factors relevant to the initial threshold inquiry have not been argued or sufficiently developed by the defendants in order to support their burden. Regarding the number of jurors exposed to the movie, while the affidavits seem to indicate that every member of both juries was present in the room during the playing of at least part of the movie, they do not clearly indicate that every member actually viewed it. Regarding discussion of the extraneous material among the jurors, the defendants have not established that the movie was extensively discussed. Finally, the timing of exposure and the length of deliberation do not support a conclusion that the movie affected the verdict. As already noted, those jurors who watched the movie did so well before deliberations began, and before being charged by the court to consider only the evidence before them, free *130from prejudice or sympathy.17 Further, both juries deliberated for a substantial period, more than a week, before rendering a guilty verdict. Consequently, rather than supporting their threshold burden, the timing and length of deliberation here supports the conclusion that the movie did not likely play a role in the verdicts.
Given the unique facts of this case, defendants have failed to meet the threshold burden of establishing a real and substantial possibility that the viewing of the movie Malcolm X affected the verdicts. Virtually every factor courts have identified as relevant weighs against a finding that the movie could reasonably have affected the verdicts.
I would further find that, even if the defendants had met their initial burden, a reversal is not required here because it can be said, beyond a reasonable doubt, that the other evidence adduced at trial was so overwhelming that there is no real and substantial possibility that viewing the movie might have contributed to the convictions. Anderson, supra. In this regard, I agree with the Court of Appeals conclusion that reversal, on the basis of the showing of Malcolm X, was not required. The Court of Appeals explained that reversal was not required
given the overwhelming evidence against defendants in favor of the convictions. By all eyewitnesses’ accounts, both defendants struck Malice Green repeatedly with their flashlights while he offered little, if no, resistance. Fletcher, Pace, Hollins, Knox and Brown testified that Budzyn initiated the beating by repeatedly striking Green with his flash*131light on the hands. Despite their inability to directly observe Budzyn striking Green in the head, these witnesses testified that Budzyn was on top of Green inside the car and was swinging his flashlight over his shoulder down toward Green. Based on this position and the sounds, it appeared to them that Budzyn’s blows impacted Green’s head. Furthermore, Brown actually saw Budzyn strike Green in the head several times. In addition to some of these witnesses, the testimony from the four medical technicians established that Nevers struck several blows while Green was covered in blood.
Moreover, the number of cuts and bruises that Green sustained exceeded the number of blows to which defendants admitted rendering. Budzyn denied hitting Green at all, whereas Nevers admitted hitting Green five or six times. However, the medical examiner found nine cuts on Green’s scalp and other cuts and bruises on his face and behind the ears. In all, the medical examiner opined that Green sustained at least fourteen separate blows to his head. Along with the testimony from civilian and medical eyewitnesses, the obvious implication is that either or both defendants were minimizing the amount of force utilized in arresting Green in order to justify the murder. Thus, in light of this evidence, we cannot say that the jury convicted defendants on the basis of emotions that could have arisen from watching the videotape. The evidence clearly supports the verdicts of second-degree murder. [Slip op at 6.]
I respectfully disagree with the majority’s conclusion that the evidence supporting second-degree murder against defendant Budzyn was not overwhelming. As noted by the Court of Appeals, and supported in the record, all five of the eyewitnesses, Fletcher, Pace, Hollins, Brown, and Knox, testified that Budzyn initiated the beating by repeatedly striking Green on the hands. Witnesses Pace and Hollins, who observed the scene from essentially the same vantage point, saw Budzyn swinging his flashlight over his shoulder down toward Green in the direction of Green’s head. *132They both testified that on the basis of the positioning of Budzyn and Green, and the sound of impact, it appeared that Budzyn was repeatedly hitting Green on the head with his flashlight.18 Further, eyewitness *133Brown, who observed the scene from a different vantage point, standing on the curb at a slight elevation and to the rear of the car, testified that he actually *134saw Budzyn strike Green on the head with the flashlight several times.19 Witness Fletcher’s testimony, which did not indicate that Budzyn’s blows appeared to land on Green’s head, was not necessarily inconsistent with the conclusion of the other witnesses. Fletcher explained that he did not see everything that transpired after observing the blows to Green’s hand and after Budzyn positioned himself further into the car on top of Green because Officer Budzyn was positioned in a way that obstructed his view for a period.20
*135I cannot help but conclude that the evidence concerning Budzyn’s actions was overwhelming to support the verdict of second-degree murder. This conclusion would not be different even if, as the majority suggests, the injuries actually causing Green’s death could have been inflicted solely by Nevers. While Budzyn’s actions were perhaps less brutal than Nevers’, he initiated the encounter with Green that *136ultimately resulted in Green’s death. His actions went far beyond the level of force necessary in the situation. Rather than calling his partner to assist him in pulling Mr. Green out of the car and handcuffing him, Officer Budzyn exerted excessive force in attempting to get Green to open his clenched fist. Further, Budzyn did not step in to stop Nevers’ beating of Green when Nevers continued the beating with actions that were clearly an unjustified and brutal show of force.
The majority attempts to discount the evidence against Budzyn by questioning the credibility of the eyewitnesses. It may or may not be true that because each of these witnesses had been implicated in criminal activity, or because they may have smoked cocaine or consumed alcohol before witnessing the events leading up to Green’s death, they were less credible than the ems witnesses who testified against defendant Nevers. This Court, however, is not in a position to judge the relative credibility of the witnesses against the two officers. The jurors heard the testimony from eyewitnesses Pace, Hollins, Brown, and Fletcher, all of whom were subjected to thorough cross-examination, and evidently found it to be credible.21
w
MEDIA REPORTS
I will next turn my attention to the jurors’ exposure to media reports concerning contingency plans in case of postverdict rioting. Defendants argue that exposure to these reports tainted their juries. Only *137two of the four juror affidavits offered by the defendants mention exposure to this information; one from the Nevers panel and one from the Budzyn panel. The Nevers juror recounted exposure to a news report regarding possible rioting and freeway closings while home overnight during the deliberation period. The Budzyn juror stated that during deliberations someone had mentioned news reports that the city was preparing for possible rioting and that businesses and freeways might be closed if defendants were acquitted.
Initially, we note that, unlike the viewing of Malcolm X, exposure to these reports occurred during the deliberation period. Consequently, the timing of this exposure is more troubling because the sanctity of jury deliberations is more directly implicated. However, virtually every other aspect of this particular extraneous influence is less troublesome than the viewing of Malcolm X.
In contrast to the affidavit statements regarding the movie, which arguably support a finding of exposure by virtually every juror on both panels, affidavit statements regarding media reports do not indicate this degree of exposure. The Budzyn affidavit is equivocal regarding how many on that panel were exposed. It relates that someone mentioned media reports, but does not indicate how many on the panel heard the comment, or whether only the writer of the affidavit heard it. The affidavit from the Nevers juror attests to that juror’s own exposure, but says nothing about whether the juror discussed the reports with other jurors or whether other members of the panel had also heard the reports. From the affidavits, one can only be certain of exposure by one Budzyn juror and *138at least two Nevers jurors. Consequently, defendants have not produced sufficient evidence that either panel, in its entirety, was tainted by the reports. While it is not necessary to a finding of prejudice that the entire panel be exposed, the number of jurors exposed to the prejudicial information is relevant to determining the possibility of prejudice. Watkins, supra. The lack of proof of extensive juror exposure to these reports weighs against a new trial. See United States v Ruggiero, supra (holding that one juror’s exposure to extraneous information concerning prior criminal acts of the defendant, which she did not disclose to the other jurors before they rendered a verdict, did not merit a new trial).
Even if defendants had shown extensive juror exposure, I would not conclude that there was a real and substantial possibility that it affected the verdict. Unlike other cases finding taint from exposure to media reports,22 the information conveyed in these reports did not relate directly to a material aspect of the case. Caro-Quintero at 1574. “Reviewing courts will not disturb jury verdicts on appeal when extraneous information relates only to issues not material to the guilt or innocence of the defendant.” United States v Bagnariol, 665 F2d 877, 887 (CA 9, 1981). *139Information in the media reports was not material to the defendants’ guilt or innocence.
Further, I cannot reasonably conclude that the jurors would allow fear of potential rioting to sway them to convict. While this is, of course, a possibility, it is not a necessary, or even a reasonable, conclusion. It must be assumed that most, if not all, of the jurors were aware of the Los Angeles rioting occurring after the Rodney King acquittals. Given this reality, exposure to reports that officials were prepared and had contingency plans in place may have had the opposite effect and lessened any pressure to convict. Again, on the basis of the juror affidavits, I am not convinced that fear of rioting had a real and substantial possibility of shaking the jurors’ resolve to abide by the judge’s instructions that they should decide the case only on the evidence and not let sympathy or prejudice influence their decisions.
Finally, even if the defendants had met their burden of showing that exposure to the media reports reasonably affected the verdict, a new trial would not be warranted because of the overwhelming evidence against the defendants, as explained earlier in this opinion.
v
JUROR DELIBERATIONS INVOLVING STRESS
I disagree with the majority’s conclusions regarding the jurors’ discussion of the Stop the Robberies Enjoy Safe Streets unit, and of their apparent belief that defendants had been members of that unit. Clearly, as indicated in voir dire questioning of potential jurors, the defendants were aware that STRESS was a potentially inflammatory topic and that the nature and exis*140tence of the unit was common knowledge among many of the potential jurors. In spite of this, defense counsel made no effort to deal with the issue at trial and did not seek any cautionary instructions beyond the normal instruction given by the trial court that the jurors should consider only the evidence before them and that they should view this evidence without prejudice or sympathy.
The jurors’ personal knowledge and life experiences as urban-dwelling Americans probably made inevitable a discussion of the wider problem of police brutality and specifically of the notorious stress unit. That their deliberations would include a discussion of stress was, perhaps, to be expected. That they might speculate and discuss their own knowledge or beliefs that one or both of the defendants had once been members of stress is also not surprising.
Can it be seriously argued that defendants are compromised because the acts of which they stand convicted invoked a discussion of the official lawlessness epitomized by the jurors’ apparent understanding of stress? Police brutality was what this trial was about. Juror discussions about the history of police brutality, in their own home town, was a natural part of their deliberative process and should not be relied on by this Court in support of defendants’ claims of error.23
vi
other issues
While the defendants raise other extraneous influences that they believe, in combination with the *141issues discussed, tainted the jury, I find them to be substantially less troubling than those I have discussed. Further, I am unpersuaded that a combination of factors denied the defendants a fair trial by an impartial jury. A careful review of the record assures that the trial judge generally did an exemplary job of ensuring a fair trial for the defendants. A review of the voir dire transcripts, particularly those portions concerning seated jurors, also assures that the jurors on both panels were unbiased.24 The defendants have failed to convince me that any extraneous influences rose to a level of prejudice sufficient to taint the jury panels.
*142Finally, I have carefully reviewed the defendants’ numerous other claims, and find that some are lacking in merit and that others were properly dealt with by the Court of Appeals. Consequently, I will not discuss them further.
vn
CONCLUSION
All citizens must fiercely guard the processes and institutions that ensure our liberty. The chosen occupation of these defendants is one that plays a crucial role in protecting the life and liberty of all citizens. I appreciate the difficulty faced by those who seek to keep the peace on our city streets. I recognize that the great majority of police officers conscientiously perform their duties with the utmost regard for the rights of those they seek to protect. However, when police officers cross over the line from keepers of the peace and protectors of lives to aggressors, exerting force far beyond that required in the given situation, the law must treat them no differently than other defendants.
The right to a fair trial by an impartial jury is of paramount importance in securing our liberty. Consequently, this Court has properly taken very seriously the defendants’ claims that their juries’ exposure to extraneous influences denied them this right. After careful review however, I am left with the distinct and firm impression that the defendants received a fair trial.25 The potential prejudice from the asserted influences is real. However, it does not rise to a level *143sufficient to persuade me of a real and substantial possibility that the verdicts were affected. Additionally, given the overwhelming evidence supporting the verdicts, it can be said, beyond a reasonable doubt, that these influences did not contribute to the verdicts.
For all the above reasons, I would affirm the Court of Appeals decision.
Wayne County Medical Examiner autopsy report.
The facts of this case bear some similarities to the Los Angeles case in which the beating of black motorist Rodney King by white police officers was heavily broadcast throughout the country. Severe rioting followed the acquittals of the officers in that case. In the present case, the victim was black. The defendant police officers are white. Ten of the twelve Nevers jurors were black; eleven of the twelve Budzyn jurors were black.
3 I note at this juncture that defendants initially argued that the jury had not been properly instructed regarding causation. Since submitting their initial briefs, this Court decided People v Tims, 449 Mich 83; 534 NW2d 675 (1995), which held that the causation element is satisfied if the defendant’s actions are merely acause, not necessarily the cause of death. Consequently, defendants have conceded this claim of instructional error.
451 Mich 884 (1996).
FRE 606(b) provides:
*120[A] juror may not testify as to any matter or statement occurring during the course of the jury’s deliberations or to the effect of anything upon that or any other juror’s mind or emotions as influencing the juror to assent to or dissent from the verdict . . . , except that a juror may testify on the question whether extraneous prejudicial information was improperly brought to the jury’s attention .... Nor may a juror’s affidavit or evidence of any statement by the juror concerning a matter about which the juror would be precluded from testifying be received for these puiposes. [Emphasis supplied.]
An exception applies to certain ex parte communications. For example, ex parte communications from judge to jury during deliberations are presumed prejudicial. People v France, 436 Mich 138; 461 NW2d 621 (1990).
See United States v Caro-Quintero, 769 F Supp 1564, 1575 (CD Cal, 1991); Marino v Vasquez, 812 F2d 499, 506 (CA 9, 1987), citing Bayramoglu v Estelle, 806 F2d 880, 887 (CA 9, 1986).
Some of the cases from the Court of Appeals for the Ninth Circuit discuss the factors relevant to determining prejudice as part of the harmless error determination and some simply combine the threshold prejudice prong of the test with the harmless error prong. I would apply the relevant factors as part of the threshold inquiry because they are more applicable to determining the level of potential prejudice than to the harmless error inquiry.
The asserted error is of constitutional magnitude. Outside influences are not subject to cross-examination and cannot be confronted by the defendant. Consequently, this type of claimed error directly affects the right to a fair trial and must be evaluated under the stringent beyond a reasonable doubt standard.
One Budzyn juror and two Nevers jurors submitted signed affidavits. An additional Nevers juror prepared, but did not sign, a statement.
I would reject the defendants’ argument that because the court provided the videotapes from which the jurors made their choice, the movie should be viewed as an ex parte communication from the judge to the jury, thus warranting a presumption of prejudice. People v France, n 6 supra. Defendants’ conclusion in this regard goes beyond the holding in France.
For example:
Brothers and Sisters, I am here to tell you that I charge the white man, I charge the white man with being the greatest murderer on earth. ... I charge him with being the greatest kidnapper on earth, I charge him with being the greatest murderer on earth .... We didn’t see any democracy on the streets of Harlem ... on the streets of Detroit .... We’ve experienced only the American nightmare.
I note that the Nevers jurors did not watch the entire movie, but only watched it for approximately forty-five minutes. Another Nevers juror indicated that they watched the entire first cassette, but did not watch the second cassette. Although the Nevers jurors did not view the entire movie, they viewed enough of it to have a flavor of its message and to place the opening sequence in context.
In evaluating the nature of this extraneous influence and its likely effect on these jurors, I cannot ignore the fact that most of the jurors were African-Americans. These jurors were undoubtedly intimately and personally aware of the problems in this country brought to the forefront by the Rodney King incident. If they could set aside their personal knowledge of police brutality, they could surely set aside any potential emotional effect of the movie.
Other choices included Body Guard, Leap of Faith, and Scent of a Woman.
Other cases requiring reversal on grounds of juror exposure to a movie before deliberations are also distinguishable. In State v RW, 200 NJ Super 560; 491 A2d 1304 (1985), the court refused to instruct jurors near the beginning of the trial not to watch a movie airing on television that depicted child abuse. The court reversed the child sexual abuse convictions, finding that the movie at least combined with other errors to require a new trial. Significant to the court’s analysis was the fact that the only evidence of defendant’s guilt was the testimony of the three-year-old child and that the trial judge could have avoided any potential prejudice by instructing the jurors to avoid watching the movie.
In State v Blanks, 479 NW2d 601 (Iowa App, 1991), the court focused on parallels between the plot of the movie Gorillas in the Mist with the facts of the case before it. The prosecutor referred to the movie in closing argument. Because of these parallels, and the potential that reference to the movie could be interpreted as having racial overtones by the jury, the court held that a new trial was required.
*129A review of these, and other, “movie” cases reveals certain trends in the decisions. See anno: Permitting jurors to attend theater or the like during course of criminal trial as ground for mistrial, new trial, or reversal, 33 ALR2d 847. The early cases are characterized by a reluctance by courts to “speculate” regarding prejudice. If there were any chance that the movie might have influenced the verdict at all, reversal was deemed required. Later cases, however, reveal a more searching analysis regarding whether the movie was likely to have a prejudicial effect in light of all of the circumstances. Factors relevant to the inquiry appear to be: (1) whether the jury viewed the movie apart from others, thus lessening the chance of interaction and potential influence from outside persons, (2) whether the movie was viewed during deliberations, (3) the similarity of the film to the case and the potential for prejudice resulting from any similarities, and (4) the weight of the evidence against the defendant.
These factors weigh against ordering a new trial in this case. The jurors viewed the movie alone, separated from outside persons. Further, they did not view it during deliberations. Apart from the brief opening sequence, and some isolated historical depictions of police brutality, the overall plot and message of the movie bore no similarities to the facts in the case. Finally, the weight of evidence against the defendants was substantial.
It is also noteworthy that the trial court, throughout the trial, repeatedly admonished the jury to view only the evidence admitted at trial and took scrupulous care to remind them to avoid outside influences.
For example, Theresa Pace testified as follows:
Q. All right. When you say that Defendant Budzyn goes over, grabs his hand, starts beating it, he then, you said, straddled him, or words to that effect; is that right?
A. Yes.
Q. And when he actually straddles him, is he in a face to face position with Malice Green?
A. Yes, he is.
Q. Where do you see him striking him at this point with the flashlight?
A. I couldn’t see that much into the car. There wasn’t much space left on his body he could have been hitting him but his head because of the way he was straddled on him.
Q. When he began to hit him, hit Malice Green in the knuckles, can you show us what motion he used to do that as best you can?
A. Like this.
Q. And you are indicating an up and down motion?
A. Yes.
Q. Over the hand; is that right?
A. Yes.
Q. He straddles him; is that right?
A. Yes, he does.
Q. Describe what you see then.
A. The flashlight is still in his hand, he’s like this (indicating) hitting Malice.
Q. And you are indicating an overcoming up and down motion?
A. Yes, sir.
Q. Over your shoulder; is that right?
A. Yes.
Q. Can you tell us what part, based on where you were at and what you saw, what part of Malice Green’s person would have been underneath this light . . .
*133A. His forehead.
Q. ... as it came down? Can you actually see the contact being made?
A. I couldn’t actually see it but there was nothing left of him elsewhere to hit.
Q. Could you hear the blows being struck?
A. Yes, you could hear ’em.
Q. Did you see him, when Defendant Budzyn came over and grabbed his hand and started to straddle him, did you see him fighting Defendant Budzyn in the sense of punching him or kicking him or biting him or scratching him or whatever?
A. No.
Similarly, Mr. Hollins testified:
Q. All right. Now, you initially saw the flashlight hitting him in the hand; is that correct?
A. Yes.
Q. Could you see where it was hitting him at this point?
A. I couldn’t see actually where contact was made, no. I just could see that it was — had to be above the shoulders.
Q. . . . given the position that Officer Budzyn was at over Malice Green, what part of him would he have been striking?
Mr. Goldpaugh-. The question has been asked and answered, your Honor.
The Court-. I’ll take it again.
By Mr. Baker:
Q. What part of him would he have been striking given the way they were positioned from what you saw?
A. From what I saw it would have to be the head area.
Q. Could you hear it?
A. Could I hear it?
Q. Yes.
A. Yes, I could hear it making contact.
Mr. Brown testified as follows:
Q. Now, what did you see Officer Budzyn do with this flashlight?
A. Began to strike Malice Green in the head.
Q. Now, explain, please, were you on a curb?
A. Yes, I was.
Q. And were you elevated looking down into the car?
A. Yes, I was.
Q. And from where you were standing, how was it that you could see into the car? What part of the car could you see into?
A. Through the rear window.
Q. And could you also see into the side passenger door?
A. Yes, I could.
Q. Was it open at that time?
A. Yes, it was.
Q. How was it that you saw Officer Budzyn striking Malice Green in the head?
A. Because the interior dome light was on, plus the lighting on the car, on the unmarked car behind it.
Q. Where was Malice Green’s head when you saw him being struck?
A. In between the seats.
Q. When you say, in between the seat, you mean between the front passenger’s seat and the front driver seat?
A. From where I was looking, they had like split, the seats, with a gap in between ’em, and that’s where his head was.
Relevant portions of Mr. Fletcher’s testimony are as follows:
Q. ... He only hits Mr. Green in the hand?
*135A. That’s all.
Q. He never hit him anywhere else, did he?
A. I didn’t see him hit him anywhere else.
Q. Now, you also, in response and I don’t remember which attorney it was, to one of the — I think it was Mr. Batchelor, I think, and he was talking about when you could no longer see the hands of Malice Green and see Officer Budzyn was because he had his back to you in the car?
A. Yes, ma’am.
Q. And when this happened you could not see what was going on inside of the car; is that right?
A. Yes, ma’am.
Q. So you didn’t actually see the strikes hitting his hands at that point, did you?
A. No, ma’am.
Q. Did you see the flashlight going up at that time?
A. No, ma’am. I had backed away then.
Q. When your view — I understand that you testified that most of the time you could see his hands, I mean his flashlight actually striking the hands of Malice Green?
A. Yeah.
Q. When your view was obstructed when you backed up, could you see the exact portion of the body that the flashlight was landing?
A. No, ma’am.
Q. Could you hear that it was landing on some part of his body?
A. Yes, I could.
Q. And, again, just so we are all clear that is during the time when your view was obstructed because he was directly in front of you?
A. Yes, ma’am.
I further note that whether the evidence may also have supported a lesser verdict is not the question before us.
An example of such a case is United States v Bruscino, 662 F2d 450, 461 (CA 7, 1981), in which a juror read, and cut out, an article reporting a codefendant’s guilty plea to a charge of conspiracy. The juror stated that she cut the article out of the paper and took it to court so the other jurors could read it. Because the information contained in the article supported the government’s theory concerning the defendant’s guilt, exposure to the article mandated a new trial.
Typical news accounts deemed prejudicial are those with “massively inflammatory revelations of prior felony convictions, past criminal conduct!, or] admitted and alleged improprieties . . . .” United States v Solomon, 422 F2d 1110, 1118 (CA 7, 1970).
In contrast, the movie Malcolm X and the media reports of riot contingency plans were clearly extraneous to the issues in this case.
For example, we note the following statements during voir dire made by some of the seated jurors, detailing their exposure and reactions to media reports:
No, I don’t read the newspaper, very seldom, and like I say, a person is Innocent until proven guilty. That’s my strong belief and I [would] fight for his right to have it so.
* ** *
I cannot believe any of it right now because I have not heard— it’s like trying to find out the truth to a mathematical equation, okay? I have to actually work through the process in order to get to a conclusion. I cannot just draw an opinion based on someone else’s, whatever they, their opinions are. Follow me?
Well, I read that police brutality killed this man. I don’t know if it’s true or not because the papers write anything, the news say anything, so it’s hard to believe. ... I didn’t pay . . . any attention.
* * *
I was upset, I mean, by a death, I mean sure, but the opinion was more or less neutral, because I didn’t know all the elements of the case, so — and if I was in any of those people’s position, all I know, I would want a fair trial.
While I recognize that the defendants’ trial was not perfect, “[w]e require a fair trial, not a perfect trial.” People v Beach, 429 Mich 450, 491; 418 NW2d 861 (1988).