Justice BENDER
delivered the Opinion of the Court.
In this consolidated opinion, we review the propriety of jury questioning in Colorado. The court of appeals held that juror questioning in these two cases did not violate the defendant’s right to a fair trial and impartial jury and, that under the terms of the Colorado Jury Reform Pilot Project, the defendants were not prejudiced by the questions posed by the jurors. People v. Medina, No. 02CA0202, slip op. at 6-7, 2004 WL 1119296 (Colo.App. Feb.5, 2004) (not selected for official publication); People v. Moses, No. 02CA0925, slip op. at 7-8, 2004 WL 817399 (Colo.App. Apr.15, 2004) (not selected for official publication).
We hold that juror questioning of witnesses does not create a per se violation of a defendant’s right to a fair trial. When the applicable rules of law and evidence are applied and after consulting with counsel, the decision of whether to ask a juror’s question is committed to the sound discretion of the trial court. Like other instances where a trial court errs in admitting otherwise inadmissible evidence, improper juror questions which are asked by the court will be reviewed for harmless error. We also hold that while it was error to ask one of the juror’s questions in Medina’s case, the impact of the evidence adduced from the question was harmless. And, we hold that in Moses’s case, no error occurred despite the possibility that the jury overheard a bench conference to discuss the admissibility of a juror’s question and where prejudicial statements were not made. Hence, we affirm the court of appeals and remand these cases to that court to return them to the trial court for action consistent with this opinion.
I. Facts and Proceedings Below
These two cases raise questions about the validity of allowing jurors to ask witnesses questions during a criminal trial. In September of 2000, Chief Justice Mary J. Mul-larkey authorized a pilot project to study the effects of permitting jurors to submit written questions to witnesses during certain criminal trials.1 The Colorado Jury Reform Pilot Project Subcommittee provided a list of policies and procedures that the district court was to follow when jurors asked questions. Pursuant to these policies, jurors were allowed to submit written questions to the court before a particular witness was excused from the witness stand. Trial courts were not required to ask all questions submitted by jurors. Rather, courts were instructed that the purpose behind the project was to clarify testimony and to help jurors understand the evidence. Thus, before asking a question, the court first reviewed the questions and heard all objections from counsel, on the record, outside the jury’s hearing. Keeping in mind the rights of all parties to due process and the right to a fair trial, the trial court then was directed to apply the applicable rules of law and evidence, and if the question was proper in light of these considerations, the court asked the witness the question. Once the question was answered, the attorneys were given an opportunity to ask follow-up questions of the witness. At the conclusion of the trial, the jurors, judge, and attorneys completed survey forms [848]*848about their experience with jurors asking questions in that particular case. These two cases in this consolidated opinion were randomly selected at the trial level to be part of the pilot project.
In this opinion, we first set forth the facts that led to the petitioners’ Oases before us. We then give an overview of juror questioning in this country and establish the standard of review for determining if a defendant is prejudiced by a juror’s question. We then apply that standard to the cases of the respective petitioner.
A. Yvonne Medina
Yvonne Medina was convicted by a jury of second degree assault, crime of violence, criminal mischief, first degree criminal trespass, and menacing. The court sentenced her to five and one-half years in the Department of Corrections. The evidence presented at trial showed that the victim, her ex-boyfriend, was home in his apartment when Medina knocked at his door. The victim refused to let her in because his new girlfriend was in the apartment. Medina then broke a window next to the door, reached in, and unlocked the front door. Once inside, she picked up a piece of glass, approached her ex-boyfriend, and stabbed him several times while the other woman hid in a closet. The victim then left his apartment to seek help and when he returned, he found his apartment vandalized.
At the beginning of the trial, the court advised counsel and the jury that jurors would be allowed to submit questions to witnesses through the court pursuant to the pilot project guidelines authorized by the Chief Justice. Defense counsel objected to the procedure. Only one question was asked by the jury throughout the trial.
Medina’s theory of defense was that the victim and the other woman concocted the story about what had occurred and that Medina was not at the ex-boyfriend’s apartment the night of the incident. In support of this theory, Medina called the district attorney’s investigator to testify about pre-trial interviews he had with the witnesses. She stated that the victim and the woman made a number of inconsistent statements during these interviews with the investigator. After being examined by the attorneys for the defense and prosecution, a juror submitted a written question to the court asking how frequently witnesses change their stories or make inconsistent statements. The question asked:
Roughly what percentage of reports that you have taken in your career reflect some inconsistencies from witnesses, i.e., how common is it for witnesses to add or subtract information from their original statements?
Pursuant to the procedures set forth by the pilot project guidelines, a bench conference was held where defense counsel objected to the juror’s question. First the defense reasserted its general objection to jurors asking questions and second the defense argued that the question was not relevant. The prosecution did not comment on the objection. The court overruled defense counsel’s objection and asked the question. The investigator responded that people do change their stories and that individuals involved in domestic violence cases are more likely to change their story than in other situations. She said:
It probably depends on the type of case. Domestic violence, it probably happens more so than your vehicular homicide or your vehicular assault case or a DUI or something. I hadn’t really thought about that....
I’m trying to think percentage-wise about how many percentages of people would change their story. It’s a high percentage. I don’t think it’s up to 50 percent, but I hadn’t thought about how many people change. But it’s probably more than 10 percent. Somewhere maybe even more than 20 percent, but you know, that’s probably the best I can do. Sorry.
Medina appealed her conviction and in an unpublished opinion the court of appeals affirmed. Medina, slip op. Medina argued that allowing jurors to question witnesses undermined her right to a fair trial. The court of appeals, however, followed other decisions of that court which held that questions from the jury do not violate a defendant’s right to a fair trial. It also held that this particular question did not prejudice the [849]*849defendant and, if anything, it helped the defendant’s ease because it showed that the testimony of the main witnesses in this case, the victim and his new girlfriend, may be unreliable. Medina, slip op. at 6-7.
We granted certiorari to review the decision of the court of appeals.2
B. Philip Moses
Philip Moses was at his place of work, a car dealership in Aurora, Colorado, when two police officers arrived to arrest him for a municipal violation. Moses was seated in his car and when the officers approached, he put his ear in reverse and backed out of his parking space. One of the officers approached the car and Moses accelerated forward, hit the officer with the vehicle, and knocked him to the ground. Moses drove off and was stopped by a police roadblock where he got out of his car and fled on foot. Law enforcement then found Moses hiding in some bushes a short time later and arrested him.
Moses was charged with first degree assault on a peace officer, felony menacing, and mandatory sentencing for a crime of violence. Early in the trial, the court overruled defense counsel’s general objection to the practice of allowing jurors to ask questions and allowed the jury to ask questions through the court of the various witnesses according to the terms of the pilot project. The record in this case contains seventeen written questions from the jury with some containing multiple sub-questions. Four of the written questions were denied by the trial court. Not all of the bench conferences were held on the record.3
After the second witness was called, jurors submitted several written questions to the court. Counsel approached the bench to discuss the admissibility of the jurors’ questions. Before discussing the questions, Moses’s attorney stated to the court that the defendant was able to hear the attorneys and the court at the first bench conference where the parties discussed the admissibility of four questions from the jury.4 The court acknowledged to the parties that the volume to the headset was set too loud and suggested that the attorneys try and keep their voices down. There is no other indication in the record that the jury had in fact heard the previous bench conference or heard any subsequent bench conference.
The jury acquitted Moses of first degree assault on a peace officer but convicted him of felony menacing, resisting arrest, and reckless driving. The court sentenced him to three years in the Department of Corrections and he appealed. The court of appeals affirmed these convictions. Moses, slip op. It held that allowing juror questions does not violate a defendant’s constitutional right to due process and equal protection and that there was no evidence to indicate that the jury overheard the bench conversation or that Moses was prejudiced by any of the objections or comments of the attorneys. Id., at 7-8.
We granted Moses’s petition for certiorari to review the decision of the court of appeals.5
[850]*850II. Analysis
We now address the issue raised by the defendants in these two cases of whether allowing jurors to ask witnesses questions during trial violates a defendant’s constitutional rights to an impartial jury, due process, and proof beyond a reasonable doubt. We begin our analysis with a short history of jurors asking questions in the Anglo-American criminal justice system before turning to how other jurisdictions have handled this matter. We then address whether the practice violates Colorado’s Constitution and establish the standard for reviewing questions on appeal and apply that standard to the facts of these two cases.
A. Juror Questions — Past and Present
To set the stage for our decision, we give a brief history of how the manner in which the jury conducts its charge has evolved over time. The current criminal justice system has progressed from one where bad luck and poor social status were factors that were more likely to result in a conviction than actual proof of a criminal act. In Sir William Blackstone’s Commentaries on the Laws of England, the noted jurist gives an account of the many ways in which a criminal trial proceeded prior to our understanding of the adversarial criminal trial. 4 William Blackstone, Commentaries on the Laws of England, *342-351 (Cooley ed. 1899). These early tests of guilt incorporated society’s preoccupation with the supernatural and included trial by ordeal, test of the morsel, and trial by battel.6 The presumption underlying these tests was that God would intervene and, through a miraculous sign, indicate which party was in the right. Landsman, Development of the Adversary System, 44 Ohio St. L. J. at 718. Such methods of determining guilt by way of divine intervention fell out of use as the age of reason made clear the “folly and impiety of pronouncing a man guilty, unless he was cleared by a miracle.” Blackstone, Commentaries, at *343.
The early Anglo-American jury system originated sometime in the twelfth century. The trial by jury in the English system of justice grew out of a general apprehension of state power denigrating the rights of the common man. “Our law has therefore wisely placed this strong and two-fold barrier, of a presentment and a trial by jury, between the liberties of the people and the prerogative of the crown.” Blackstone Commentaries, at *349.
These juries were composed of qualified individuals, chosen by the king, and who reported to the court on facts or issues in dispute. See generally, Landsman, A Brief Survey of the Development of the Adversary System, 44 Ohio St. L.J. 713 (describing the evolution of the jury). Jurors did not hear witnesses, but rather conducted investigations of those who had knowledge of the facts.7 Then, in the fifteenth century, juries began hearing the testimony of witnesses in court, a practice that became commonplace in the sixteenth century. Id. at 723; 2 McCormick on Evidence § 244, at 90-91 (J. Strong et al. 5th ed.1999); 5 John H. Wigmore, Evidence § 1364 (Chadbourn rev.1974). Thus as part of the search for truth, jurors, judges, and counsel in the English courts were allowed to ask questions and call witnesses. See John Langbein, The Origins of Adversary Criminal Trial 319 (A.W. Brian Simpson, Oxford University Press, 2003).
Concern over state power in the administration of justice carried over to the framers [851]*851of the United States Constitution who consistently agreed on the right of an individual to be tried by a jury. See Albert W. Alshuler & Andrew G. Deiss, A Brief History of the Criminal Jury in the United States, 61 U. Chi. L.Rev. 867, 870-71 (1994). The jury provided protection against failures inherent in our system, and served to “guard against the exercise of arbitrary power — to make available the commonsense judgment of the community as a hedge against the overzealous or mistaken prosecutor and in preference to the professional or perhaps over-conditioned or biased response of a judge.” Fields v. People, 782 P.2d 1145, 1151 (Colo.1987) (quoting Taylor v. Louisiana, 419 U.S. 522, 530-31, 95 S.Ct. 692, 697-98, 42 L.Ed.2d 690 (1975)). The jury’s presence thus provides a formidable restraint on government power in our criminal courts. Therefore, it is only logical that jurors who are the arbiters of truth and hedge against government power also have the tools available to think critically and to seek clarification of the issues presented to them.
Allowing jurors to ask witnesses questions is “neither radical nor a recent innovation.” State v. Doleszny, 176 Vt. 203, 844 A.2d 773, 778 (2004). It is a practice with “deeply entrenched” roots in the common law. United States v. Bush, 47 F.3d 511, 515 (2nd Cir.1995). In 1907, the Supreme Court of North Carolina recognized the long history of allowing jurors to ask questions and the potential benefits that the practice provides in the search for truth. The court stated:
This course [of allowing jurors to ask questions] has always been followed without objection so far as the writer has observed, in the conduct of trials in our superior courts, and there is not only nothing improper in it when done in a seemly manner and with the evident purpose of discovering the truth, but a juror may, and often does ask a very pertinent and helpful question in furtherance of the investigation.
State v. Kendall, 143 N.C. 659, 57 S.E. 340, 341 (1907); see also Schaefer v. St. Louis & S. Ry. Co., 128 Mo. 64, 30 S.W. 331, 333 (1895) (stating: “We ... do not see why [asking questions] was not a commendable thing in both the court and the jury ... so that they could properly determine the case before them.”); Chicago ’Hansom Cab C. v. Havelick, 131 Ill. 179, 22 N.E. 797 (1889).
The purpose of the jury “in criminal cases [is] to prevent government oppression” and “to assure a fair and equitable resolution of factual issues.” Colgrove v. Battin, 413 U.S. 149, 157, 93 S.Ct. 2448, 2453, 37 L.Ed.2d 522 (1973). While the manner in which the jury conducts its -charge of finding facts in a criminal case has changed over time — from being inquisitors directly involved in the gathering of evidence and testifying to,that evidence, to a more passive role where it considers the evidence presented by the government and the defendant — jurors have consistently been empowered to ask questions of witnesses so as to satisfy their weighty responsibility and purpose.
This historical trend continues to the modern day where the vast majority of courts from other jurisdictions allow jurors to ask questions in criminal cases with only a handful of states disallowing the practice. WTiile the United States Supreme Court has not specifically decided, the constitutionality of juror questioning, none of the federal circuits prohibit the practice.8 In addition, the majority of state courts also allow jurors to ask witnesses questions9 and several states spe[852]*852cifically provide for juror questions in court rules or state statute.10
The reasons for adopting the practice vary, however, a general theme appears from these jurisdictions. These courts do not believe that allowing jurors to ask questions prejudices a criminal defendant nor does it deprive a defendant of his right to an impartial jury. See e.g., State v. Fisher, 789 N.E.2d at 229 (stating: “the mere possibility that a juror may submit a biased question or engage in premature deliberation does not violate the Ohio or United States Constitution” (emphasis added)). Instead, these jurisdictions believe that if proper procedures are put in place by the trial court to screen objectionable questions, a number of benefits from jury questioning may be realized without prejudice to the defendant. See e.g., State v. Doleszny, 844 A.2d at 779 (allowing trial courts in their discretion to allow juror questions and finding “a significant recent trend towards endorsement of the practice and emphasis on its benefits”).
“Juror inspired questions may serve to advance the search for truth by alleviating uncertainties in the jurors’ minds, clearing up confusion or alerting the attorneys to points that bear further elaboration.” Richardson, 233 F.3d at 1290 (quoting United States v. Sutton, 970 F.2d 1001, 1005 n. 3 (1st Cir.1992)). Commentators also agree that juror questioning facilitates the search for truth and justice, clarifies the facts in complex cases, provides the jury with an essential tool to fulfill its role as the finder of fact, and increases juror attentiveness during trial and satisfaction with the judicial process.11
Despite this national trend toward allowing jurors to ask questions through the court, a handful of state courts have decided to prohibit the practice of allowing jurors to ask questions altogether.12 These courts have held that juror questioning has the potential of disrupting the neutral role that jurors play in the adversarial system of justice. Costello, 646 N.W.2d .at 213 (stating: “we conclude that maintaining the neutral role of jurors in an adversarial system outweighs whatever enhancement to the truthfinding function that juror questioning allows”). Thus, these [853]*853jurisdictions chose not to allow the practice due to potential problems that may arise.
B. Reasons to Prohibit Juror Questioning
The defendants in these cases claim that allowing jurors to ask questions undermines the traditional role of the jury. First, they argue that the practice encourages the jury to decide facts and form opinions about the ease before all of the evidence is presented. Second, it allows the prosecution to restructure its case according to the questions asked. Juror questioning also can put counsel in the awkward position of choosing between objecting to a juror’s question, and risk antagonizing a juror, or choosing not to object because he wants to keep the juror “on his side” despite the effect that a failure to object would have on appeal.
These arguments were adopted in large part by the Minnesota Supreme Court in State v. Costello. That court reasoned that juror questioning encourages jurors to create a tentative opinion of the case before all the evidence is presented. This creates the risk that jurors will “draw conclusions or settle on a given legal theory before the parties have completed their presentations, and before the court has instructed the jury on the law of the case.” Costello, 646 N.W.2d at 211 (quoting Morrison, 845 S.W.2d at 887). Second, the court concluded that the practice can upset the burden of production and persuasion in a criminal case. Id. at 212.
These concerns, while theoretically plausible, were not verified in the scholarly work which has examined juror questioning. Professors Steven Penrod and Larry Heuer conducted two separate empirical studies where they solicited data from judges, lawyers, and jurors about the impact that juror questioning had on the trial. Steven D. Penrod & Larry Heuer, Tweaking Commonsense Assessing Aids to Jury Decision Making, 3 Psychol. Pub. Pol’y & L. 259 (1997) (hereinafter Tweaking Commonsense); Larry Heuer & Steven Penrod, Increasing Juror Participation in Trials Through Note Taking and Question Asking, 79 Judicature 256 (1996) (hereinafter Increasing Participation)-, see also Nicole L. Mott, The Current Debate on Juror Questions: “To Ask Or Not To Ask, That Is The Question”, 78 Chi.-Kent L.Rev. 1099 (2003) (discussing, and confirming, results of Heuer and Penrod studies).
One of the Heuer and Penrod studies examined the effects of juror questioning in 29 courtrooms in Wisconsin and the other study solicited input from 103 courtrooms in 33 different states around the United States. These studies, based on empirical evidence, concluded that the purported harmful consequences of juror questioning are wholly unsupported by the data and that the “effects of [juror questioning are] really rather innocuous.” Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 280; see also Increasing Participation, 79 Judicature at 261.
In addition to these two studies, this court also commissioned a study which followed 239 randomly selected trials in Colorado which participated in the Juror Reform Pilot Project. This study gathered empirical evidence from the selected trials and culminated in a report which found that juror questioning had little negative effect on the impact of trial proceedings and may improve courtroom dynamics. Mary Dodge, Should Jurors Ask Questions in Criminal Cases? A Report Submitted to the Colorado Supreme Court’s Jury System Committee (Fall 2002), available at http://www.courts. state.co.u s/supct/ committees/jury reformdocs/dodge report.pdf (hereinafter Dodge Report).
With respect to the assertion that juror questioning encouraged jurors to decide facts and form opinions about the case, the data collected by Heuer and Penrod demonstrated that juror questioning did not impact the verdicts given and that jurors’ questions were only modestly helpful come deliberation time. These findings, they concluded] contradicted the general assertion that jurors become advocates rather than remain neutral and that juror questions did not have a prejudicial effect on the trial. Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 278-79. This same result was realized in the Dodge Report which also concluded that juror questioning does not influence jury verdicts and that factors other than juror questions ultimately influenced the outcome. Dodge Report at 13.
[854]*854Heuer and Penrod also found that juror questions were not very helpful in getting jurors to the truth and that judges and attorneys both agreed that juror questions did not provide useful information of the jury’s thinking. Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 275. Nor did questions alert counsel to issues requiring further development. Increasing Participation, 79 Judicature at 260. In Colorado, while the Dodge Report noted that juror questions provided attorneys with information about what evidence may have confused the jurors, the findings also indicated that juror questions did not help either side achieve a particular result. Dodge Report at 26-29. Thus, according to the data accumulated from these empirical studies, jury questions did not cause either party to restructure its case according to the questions asked.
These studies also discredit the defendants’ arguments that a juror may become antagonistic to one side should counsel object to his question. The majority of the jurors in the Heuer and Penrod studies who asked questions stated that they were not embarrassed or angry when counsel objected and that they typically understood the basis for the objection. Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 277; Increasing Participation, 79 Judicature at 260. Nor does the data suggest that attorneys shied away from objecting to questions. Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 276-77; Increasing Participation, 79 Judicature at 260. The findings showed that the attorneys and judges did not believe that juror questioning caused the jury to become prejudiced against a particular side. Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 278; Increasing Participation, 79 Judicature at 260. Again, the Dodge Report corroborates these findings. Judges and Attorneys did not report unfavorable reactions from jurors when a question was declined. Dodge Report at 24. Furthermore, the majority of jurors reported that they did not have an adverse reaction when their questions were not asked. Id. at 38.
These studies all agreed that juror questioning encourages jurors to become more engaged in the trial and attentive to witnesses. It also promotes juror understanding of the facts and issues of the case. Tweaking Commonsense, 3 Psychol. Pub. Pol’y & L. at 274; Dodge Report at 2. In addition, the data shows that the act of asking a question does not necessarily transform an otherwise passive juror into an advocate.13 As with any empirical analysis, these studies cannot provide a conclusive answer to the issue we confront. However, what we can learn from these studies is that the empirical data does not support the argument that juror questions are per se unconstitutional. We should therefore not wholly dismiss the value of juror questioning. Thus, we agree with the Supreme Court of Vermont which dismissed the assertion that a juror may become biased if allowed to ask a question because this argument “trades a speculative increase in neutrality for a likely reduction in juror comprehension of the evidence.” Doleszny, 844 A.2d at 785.
The rationales which the Minnesota and Texas Courts have used as reasons to disallow jury questions are not supported by the empirical studies. More importantly, those courts which prohibit the practice have based their decisions primarily upon policy reasons and not upon constitutional grounds. See Costello, 646 N.W.2d at 214 n. 14 (“Because juror questioning of witnesses is proscribed under our supervisory power, we need not reach Costello’s constitutional claims.”) These jurisdictions do not provide empirical support for their policy conclusions, but rather base their decisions to prohibit jury questions on possibilities and speculation regarding factual scenarios not present in the cases before them.14
[855]*855With this understanding of the history and broad support recognizing that juror questioning of witnesses does not violate a defendant’s right to a fair trial and may provide actual benefits to the trial process, we now address whether the Colorado Constitution prohibits juror questioning in all criminal eases.
C. Colorado
Medina and Moses both argue that Colorado should break from the majority of states and presume a prejudice inherent in juror questioning. They suggest that a defendant’s constitutional rights are violated when a juror is allowed to ask a question irrespective of the question asked and the effect of that question on a juror, the jury, or the trial as a whole.
The Colorado Constitution states that the criminally accused “shall have the right to ... [a] trial by an impartial jury.” Colo. Const, art. II § 16; see also Colo. Const, art. II § 23 (“The right to a trial by jury shall remain inviolate in criminal cases”); Colo. Const, art. II § 25 (“No person shall be deprived of life, liberty or property, without due process of law.”). We have held that this right to a jury trial entitles a defendant to a jury drawn from a fair cross section of the members of his or her community. Fields v. People, 732 P.2d 1145, 1151 (Colo.1987). And, counsel may not discriminate based on race, gender, national origin, religion, age, economic status, or occupation when selecting jurors from the jury pool. Id. at 1153 n. 15.
But, we have also held that where a juror can put aside his personal opinion or preconceived notion as to guilt or innocence and render a verdict based upon the evidence presented and the law at hand, the defendant’s right to a fair trial is not violated. [856]*856Carrillo v. People, 974 P.2d 478, 487 (Colo.1999); People v. McCrary, 190 Colo. 538, 547, 549 P.2d 1320, 1328 (1976); People v. Buckner, 180 Colo. 65, 70, 504 P.2d 669, 672 (1972). “[A] prospective juror’s indication of a preconceived belief as to some facet of the case does not warrant exclusion for cause.” People v. Vecchiarelli-McLaughlin, 984 P.2d 72, 75 (Colo.1999).
It is the trial court’s duty to determine the competency and credibility of each juror. People v. Young, 16 P.3d 821, 824 (Colo.2001); McCrary 190 Colo, at 547, 549 P.2d at 1328. The trial court’s decision to deny a challenge for cause in the jury selection process or decision to excuse a juror will not be disturbed absent an abuse of discretion. Carrillo, 974 P.2d at 485; People v. Abbott, 690 P.2d 1263, 1266 (Colo.1984). The fact that the jury may have been exposed to pretrial publicity or is familiar with the general nature of the case does not by itself cause a jury to be constitutionally defective. People v. Loscutoff, 661 P.2d 274, 276 (Colo.1983).
Our precedent recognizes that juries are formed out of the various members of the community, with each juror bringing his or her own life experiences and understanding of human behavior. In a criminal case, jurors are instructed that they should bring these life experiences and observations when considering the evidence presented at trial. CJI-Crim. 3:01 (“[Y]ou should consider all the evidence in the light of your observations and experience in life”); see also People v. Piarían, 109 P.3d 616, 632 (Colo.2005) (stating: “We expect jurors to bring their backgrounds and beliefs to bear on their deliberations but to give ultimate consideration only to the facts admitted and the law as instructed.”). Jurors are also admonished to not perform any research on the facts of the case, discuss the case, or form opinions about the evidence presented until after all the evidence is presented. CJI-Crim. 2:07. We presume that a jury follows these instructions given by the trial court. People v. Trujillo, 83 P.3d 642, 648 (Colo.2004).
In addition to a trial by jury, due process entitles a defendant to a fair trial but not a perfect trial. People v. Dunaway, 88 P.3d 619, 631 (Colo.2004); People v. Roy, 723 P.2d 1345, 1349 (Colo.1986). Due process also requires that the prosecution prove every element of a charged offense beyond a reasonable doubt. Victor v. Nebraska, 511 U.S. 1, 5, 114 S.Ct. 1239, 1242, 127 L.Ed.2d 583 (1994); Vega v. People, 893 P.2d 107, 111 (Colo.1995). This does not mean that the government’s burden is lessened simply because a juror asks a question which solicits additional relevant evidence in a criminal case. To hold otherwise would undermine the rule that an appellate court is required to consider the evidence as a whole — not simply the evidence presented by the prosecution15 —before determining if there was sufficient evidence to support a finding of guilt beyond a reasonable doubt by a rational finder of fact. People v. Dunaway, 88 P.3d 619, 625 (Colo.2004); Romero v. People, 170 Colo. 234, 256-57, 460 P.2d 784, 795 (1969) (court is to review entire course of proceedings to determine if a defendant’s right to due process and a fair trial are violated).
Thus, while a defendant does have a right to an unbiased jury, he is not entitled to have his case presented to a jury that sits as a [857]*857passive receptacle of information. Nor does it mean that once a juror forms a bias about a piece of evidence introduced at trial or about the credibility of a witness, that the defendant’s rights have been violated.
Juror questioning undoubtedly changes courtroom dynamics but not to the extent that it renders a criminal trial constitutionally infirm. Minor delays for bench conferences to discuss whether a piece of evidence is admissible or a party’s decision to reorganize its theory of the case based upon new evidence adduced at trial are not foreign concepts to a criminal trial. Our system also provides a mechanism to weed out potentially biased jurors during voir dire — a practice which is closely controlled by the trial judge and subject to appellate review. Jurors are instructed that they are not to discuss the case or come to conclusions until the end of all evidence. We choose not to adopt the defendants’ position and speculate about a juror’s state of mind or that juries wholeheartedly disregard the trial judge’s instructions and their serious and weighty responsibility in a criminal trial simply because a question is or is not asked. Therefore, we hold that the fact that a juror asks a question during a criminal trial, in and of itself, does not violate a defendant’s constitutional rights to a fair trial and an impartial jury.
We are confident that, with adequate safeguards in place, a defendant’s constitutional rights will be protected should a juror propound an inadmissible question or should a juror’s bias manifest itself through the questioning of a witness. Under the terms of the pilot project, the court was not required to ask all questions submitted by jurors. Rather, the trial judge, with the advice of counsel, screened the jurors’ written questions before asking the witness the question. Then, keeping in mind the rights of all parties, the court was instructed to apply the applicable law and rules of evidence before determining whether a question was proper before asking the witness. With these safeguards in place, the defendant still has the ability for appellate review should a trial court allow an otherwise impermissible question from the jury; a course of action which is not different from instances where trial counsel asks an impermissible question.16 Trial courts will also continue to have the power to dismiss and replace jurors who are unable to continue with their duties as jurors. Thus, where errors do arise, our trial and appellate courts will continue to have the responsibility to review the impact of the error to assure that a defendant’s constitutional rights are protected.
D. Standard of Review
Having determined that the practice of allowing jurors to ask questions of witnesses through the court is not per se unconstitutional, we now turn to the question of the appropriate standard for appellate courts to review the impact of an improper question from a juror.
When reviewing for an alleged error in the trial process, we apply different standards of review depending upon whether an objection is made and the type of error that transpired. The first step for an appellate court is to determine if an error occurred. If the court determines that an error occurred, and that error is preserved for appeal, reversal is required unless the appellate court determines that the error was harmless. That is, viewing the evidence as a whole, the error did not substantially influence the verdict or impair the fairness of the trial. People v. Stewart, 55 P.3d 107, 124 (Colo.2002) (citing People v. Gaffney, 769 P.2d 1081, 1088 (Colo.1989)); People v. Bowers, 801 P.2d 511, 518-19 (Colo.1990). Only where the error impacts the framework of the trial itself do we apply structural error which warrants automatic reversal. Griego v. People, 19 P.3d 1, 7 (Colo.2001).
As stated above, neither the United States Supreme Court nor the majority of state courts have held that allowing a juror to ask a question impacts the framework in which a criminal trial proceeds. The defendant comes to trial knowing that witnesses will be called to testify and be required to respond [858]*858to questions. It is the testimony and evidence that counsel chooses to extract from these witnesses and not a potential question from a juror which dictates how one side wishes to present its case. We do not believe that the mere possibility that a juror might ask a question about the evidence and testimony presented, changes the charges brought by the prosecution or the theory of defense. Although a juror may ask a particularly poignant question, bringing to light a fact that otherwise may have been omitted, the addition of this fact is not any different from other facts which may be revealed at trial through the voluntary statements of a witness.
Thus, we reject the defendants’ assertion that we should apply structural error when an improper question from the jury is asked of a witness. Instead, we find guidance in our prior cases which have held that when a party objects to a trial court’s erroneous evidentiary ruling, that ruling will be reviewed for harmless error. Bowers, 801 P.2d at 518-19. Similarly, questions from the trial court rise to reversible error only when the trial judge’s conduct through his questioning “departed from the required impartiality to such an extent as to deny the defendant a fair trial.” Coria, 937 P.2d at 391; People v. Adler, 629 P.2d 569, 573 (Colo.1981) (pointing out deficiency in the prosecution’s case by the trial court and suggesting that it call another witness did not breach standard of impartiality).
Juror questions which are submitted through the court, like those of counsel or the trial judge, either adduce additional evidence, or challenge the strength of existing testimony. A juror’s question which is wrongfully introduced into the trial process can have its impact and that of the answer assessed on appellate review. Thus, in instances where an otherwise impermissible question from a juror is asked by the trial court, the harmful result to the defendant from that question would be the introduction of otherwise inadmissible evidence.
Therefore, like other erroneous evi-dentiary rulings made by the trial court, we hold that where the court errs by asking an improper question from the jury, the impact of that question shall be reviewed for harmless error. Under this standard of review, reversal is required unless the appellate court determines that the trial court’s evi-dentiary ruling was harmless — that is, viewing the evidence as a whole, the error did not substantially influence the verdict or impair the fairness of the trial.17 Stewart, 55 P.3d at 124; Bowers, 801 P.2d at 519.
With these principles in mind we now turn to the alleged errors which occurred in the cases of Yvonne Medina and Phillip Moses to determine whether error occurred and, if so, whether the error substantially influenced the verdict or impaired the fairness of the trial. Stewart, 55 P.3d at 124 (citing CRE 103(a) ([“Error may not be predicated upon a ruling which admits or excludes evidence unless a substantial right of the party is affected”]); C.A.R. 35(e) ([“The appellate court shall disregard any error or defect not affecting the substantial rights of the parties.”] ); and Crim. P. 52 ([“Any error ... which does not affect substantial rights shall be disregarded.”])).
[859]*859III. Application
During the presentation of Medina’s case, defense counsel called an investigator from the district attorney’s office to testify about inconsistent statements made by the victim and his girlfriend. After examination by the attorneys, a juror submitted a -written question to the court which asked how often a witness’s story changed. The juror inquired:
Roughly what percentage of reports that you have taken in your career reflect some inconsistencies from witnesses, i.e., how common is it for witnesses to add or subtract information from their original statements?
A bench conference was held where defense counsel objected to the question based on relevance. The prosecution said nothing in response and the court overruled the objection and asked the question. The witness responded that “a high percentage ... more than ten percent ... maybe even more than twenty percent” of witnesses do change their statements. Defense counsel then followed-up on this interchange to which the witness admitted that she had not done a statistical analysis and that in some instances, a witness may change his story because he is lying.
Only relevant evidence is admissible. CRE 402 (2004). To be relevant, the evidence must have the tendency to make the existence of a fact more or less probable than without the evidence. CRE 401 (2004). Trial courts have substantial discretion in determining whether evidence has logical relevance, and an evidentiary ruling of the trial court will not be overturned unless the ruling was manifestly arbitrary, unreasonable, or unfair. People v. Melillo, 25 P.3d 769, 773 (Colo.2001) (citing People v. Dunlap, 975 P.2d 723, 741 (Colo.1999)).
The evidence admitted in this case concerned the likelihood that a witness would change his or her story and make inconsistent statements. The juror’s question asked “how common is it for witnesses to add or subtract information from their original statements?” This question sought the police investigator’s opinion with respect to the frequency that a witness would change his or her story. Generally, a witness is precluded from providing an opinion on issues that are based on scientific, technical, or other specialized knowledge unless that witness is qualified as an expert. CRE 702 (2004). The answer to the question in this ease is not based upon the witness’s first-hand impressions, such as the time of day or weather conditions when the event occurred, but based on specialized training or education of how witnesses behave when giving statements to police investigators. See Stewart, 55 P.3d at 124 (holding that police officer must be qualified as an expert before being allowed to testify about accident reconstruction and inferences he drew therefrom).
Turning to the record, the witness who responded to this juror’s question was not qualified by the defense as an expert. Furthermore, the question was irrelevant to the facts of the case. Whether ten percent or even fifty percent of people have inconsistencies in their recollection of events has no bearing on whether the witnesses in this case made inconsistent statements or whether the statements from these witnesses can be believed. Therefore, we hold that the trial court erred in allowing the question because it called for irrelevant expert testimony.
Even though the trial court erred and should not have allowed the question, we do not see how the answer to the question substantially influenced the verdict or impaired the fairness of Medina’s trial. The evidence is not probative of the witnesses’ credibility because it could be interpreted two different ways. One interpretation is that it is common for witnesses to have inconsistent statements and that despite the inconsistencies in these witnesses’ statements they are still truthful. The second understanding is that the witnesses cannot be relied upon and that they were lying. It is thus not reasonably possible that the solicited testimony substantially influenced the verdict in this ease.
Hence, we hold that although the trial court erred in allowing the juror’s question, the answer to the question in this case was harmless.
[860]*860B. Phillip Moses
Apart from the general objection to jury questioning addressed above, Moses does not challenge any of the trial court’s rulings with respect to particular juror questions which were asked by the trial court. Rather, he argues that the jury may have overheard the first bench conference where defense counsel objected to one of the juror’s questions causing prejudice to the defendant. At the second bench conference defense counsel notified the court that Moses heard the first bench conference from his seat in the courtroom. There is no indication that the jury had in fact heard this first bench conference or any other bench conference during the trial.
The trial court is in the best position to judge the effect of any improper influence on the jury. People v. Raehal, 971 P.2d 256, 260 (Colo.App.1998) (giving deference to trial court’s finding that comments made by prosecutor about defendant’s prior criminal record during bench conference were not heard by the jury) (cert, denied Feb. 16, 1999). And, the mere possibility of prejudice to the defendant is not sufficient to warrant reversal. People v. Gladney, 194 Colo. 68, 74, 570 P.2d 231, 235 (1977).
Moses concedes that the jury did not hear any other bench conference after the judge advised counsel to keep their voices down and he turned down the volume to his headphones at this second conference. At the first bench conference, which may have been overheard by the jury, the court and counsel discussed the admissibility of the juror’s questions. Four questions were submitted of which defense counsel objected to one based on relevance and lack of foundation and the prosecutor agreed. The court sustained the objection. The following conversation appears in the record.
THE COURT: For our record, we have four questions proposed to [the witness]. And by agreement, questions 1, 3, and 4 are being given without objection. Though, I want to give [defense counsel] an opportunity to make a general objection, and then — specifically, he’s objecting to question number 2. So let’s start with the general objection.
[DEFENSE COUNSEL]: My objection is generally to this procedure. I understand the Supreme Court authorized this procedure, but I’m objecting to the jury being allowed to ask questions whatsoever. My client’s constitutional — affective [sic] assistance of — and the Colorado Constitution?
THE COURT: Thank you, sir.
Then the question, number 2, for [the witness] is: The dents were made as a result of the knife stick — I would change that to night stick — would have — they been smaller or larger?
You object, sir?
[DEFENSE COUNSEL]: Object, calls for speculation on the part of the officer. Also lacking foundation that he’s a qualified expert in this area. I think it calls for speculation — foundation.
THE COURT: Do you wish to make any record?
[PROSECUTION]: Briefly. Your Honor, the evidence has been that the officer tapped on the hood which he obviously observed. I think he can testify to that— but it calls for something that he, at the time, he may or may not know the answer.
THE COURT: Thank you. After reflection, I agree it calls for speculation. Therefore deny the question.
During this conference neither party criticizes the juror asking the question or makes statements about the question other than it calls for speculation.18 No statements about testimony not already admitted into evidence were made. Furthermore, the defense and prosecution both agreed that the question called for speculation on behalf of the officer. [861]*861Thus, we decline to find an error simply because the jury may have overheard a bench conference where no prejudicial statements were made.19
IV. Conclusion
For the reasons stated, the decisions of the court of appeals are affirmed and these cases are remanded to that court to return them to the trial court for actions consistent with this opinion.
Justice COATS concurs in the judgment only.
Justice KOURLIS does not participate.