Justice LOHR
delivered the Opinion of the Court.
In People v. Harper, 796 P.2d 4 (Colo. App.1989), the Colorado Court of Appeals [79]*79affirmed the conviction of the defendant, David Joe Harper, for sexual assault on a child. The court of appeals held that the trial court did not abuse its discretion when it denied Harper’s request to poll the jury after the mid-trial publication of a local newspaper article. We granted certiorari to decide whether the reasonable possibility of prejudice test announced in Wiser v. People, 732 P.2d 1139 (Colo.1987), required the trial court to poll the jury to determine whether any jurors had read the article. Although we conclude that the Wiser test is not the appropriate standard, we hold that the court abused its discretion in declining to poll the jury. We therefore reverse the judgment of the court of appeals.
I.
In February 1988, Harper was tried in the Mesa County District Court, in Grand Junction, on charges of sexual assault on a child, in violation of section 18-3-405, 8B C.R.S. (1983 Supp.). Harper had allegedly molested a neighbor’s eight-year-old daughter during November 1983. Harper had previously been convicted, in May 1987, of sexual assault on a child by one in a position of trust, § 18-3-405(2)(b) (current version at § 18-3-405.3, 8B C.R.S. (1990 Supp.)), stemming from an unrelated incident involving a five-year-old boy. The defense moved to exclude introduction of evidence of the May 1987 conviction in the February 1988 trial; the prosecution confessed that motion.1 After the jury had been empaneled and before presentation of opening statements, the court instructed the jurors not to engage in outside reading on the case. The court specifically admonished them, “[D]o not read about the case in the newspapers or listen to radio or television broadcasts about the trial. You must base your verdict solely on the evidence presented at the trial.”
On the second day of trial, The Daily Sentinel, a Grand Junction newspaper, printed a short article about the case. That article, located on page 4B, was in-eluded in a column entitled “News Shorts,” containing three brief articles under the subheading “Grand Junction.” The article in question read:
Trial begins in sexual assault case
Opening statements were to be made today in the Mesa County District Court trial of David Joe Harper on sexual assault charges involving a young girl.
Jury selection began Tuesday in the trial before Judge Charles Buss.
Harper, a 31-year-old New Mexico man, is seeking a new trial on his May 1987 conviction on sexual assault charges involving a 4-year-old boy [2] between Oct. 1 and Nov. 30, 1983.
In this trial, Harper is accused of assaulting the young girl during the same period of time.
The two offenses are not related, except that they occurred within the same time period.
Harper was convicted to six years in prison by Judge Jose D.L. Marquez after his May conviction. He filed an appeal in October alleging that actions by the Mesa County District Attorney’s Office had denied him a chance to defend himself against the charge.
According to Harper’s appeal, he was denied an effective defense because of the length of time in which the offense allegedly occurred. He couldn’t claim he wasn’t present at the time the child said she was assaulted because of the time span involved.
No action has been taken on Harper’s appeal of the conviction in the first case.
Daily Sentinel (Grand Junction), February 10, 1988, at 4B.
Harper’s counsel brought the article to the court’s attention the next morning outside the presence of the jury. After describing generally the contents of the article, Harper’s attorney asked the court to question the jury to determine whether any juror had read articles or learned extrane[80]*80ous information about Harper.3 The court received the article in evidence4 but declined to question the jury because the defense had not established that any juror had read the article. The court explained,
[t]he issue is if anybody’s read it, and the Defendant has to establish, I think, independent of asking the jurors if they’ve read it, whether or not they’ve read it. The jurors are not subject to willy-nilly being examined during the course of a trial, whether or not they are currently obeying the Court's instructions regarding their behavior_ [A]ll the defendant has presented at this point is an article in the newspaper, The Daily Sentinel, which is not sufficient to show the jurors have been either exposed to it or disobeyed the Court’s order not to pay attention to these matters.
The jury found Harper guilty of the sexual assault charge.5 On appeal, the court of appeals noted that the trial court had admonished the jury not to read about the case in the newspapers. Relying on People v. Holmes, 191 Colo. 477, 553 P.2d 786 (1976), the court held that absent a showing to the contrary, the trial court’s admonition created a presumption that the jurors followed the court’s instructions to ignore newspaper articles concerning the trial. The court of appeals held that in absence of a showing that any of the jurors read the newspaper article or that Harper was prejudiced by the article, the trial court did not abuse its discretion by denying Harper’s request to poll the jury. The court therefore affirmed his conviction.
II.
Harper argues that the trial court abused its discretion by not polling the jury after Harper brought the newspaper article to the court’s attention. In support of this argument, Harper asserts that the article was highly prejudicial as it contained information about the excluded conviction; counsel had brought the article to the court's attention promptly; and the ethical prohibition against counsel contacting jurors during trial, see DR 7-108, created a serious obstacle to obtaining additional evidence about the jurors’ exposure to the article. Harper further asserts that polling would have indicated whether the jurors had been exposed to the article and if they had, the court then could have taken timely corrective measures to ensure him a fair trial.
Our earlier cases required that a defendant demonstrate that actual prejudice resulted from juror exposure to extraneous information or influences in order to be awarded a new trial. See, e.g., Niemand v. District Court, 684 P.2d 931, 934 (Colo.1984) (juror consulted Black’s Law Dictionary for definitions of words used in court’s instructions); People v. Hickman, 684 P.2d 228, 233 (Colo.1984) (reporter questioned jurors during noon recess); People v. Dunoyair, 660 P.2d 890, 895 (Colo.1983) (juror inadvertently failed to disclose acquaintance with prosecution witness); People v. Thatcher, 638 P.2d 760, 770 (Colo.1981) (jurors casually observed scene of the crime while returning home from court); People v. Mackey, 185 Colo. 24, 31, 521 P.2d 910, 914 (1974) (juror conversed with a police officer during sequestration concerning burglarizing of a room of one of jurors); Thistle v. People, 119 Colo. 1, 5, 199 P.2d 642, 643 (1948) (jurors read article including information that defendant was facing charges of absence without leave from army); McPhee v. People, 108 Colo. 530, 120 P.2d 814 (1941) (jurors read article referring to prior con[81]*81victions; case remanded for evidentiary hearing on actual prejudicial effect).6 Particular factual circumstances, however, can create a rebuttable presumption of prejudice. People v. Boulies, 690 P.2d 1253, 1256 (Colo.1984) (presumption of prejudice arises if alternate juror was present during jury deliberations). Cf. Walker v. People, 169 Colo. 467, 471, 458 P.2d 238, 240 (1969) (pre-trial publicity so pervasive that actual prejudicial effect need not be shown). The principle to be derived from these cases is that an evidentiary hearing would generally be appropriate to determine the actual effect of the potentially prejudicial material on the jury’s deliberations. See McPhee, 108 Colo, at 533, 120 P.2d at 815. But see Quintana v. People, 158 Colo. 189, 194, 405 P.2d 740, 743 (1965) (jurors’ actual exposure to article describing defendant as ex-convict and describing sentence received by his co-defendant required reversal without the need for an evidentiary hearing on prejudicial effect); Perry v. People, 63 Colo. 60, 65-68, 163 P. 844, 846-47 (1917) (the possibility of jury prejudice resulting from false article about defendant’s escape plans, including concealment of weapons in his cell, warranted new trial despite trial court finding of no prejudicial effect).
In Wiser v. People, 732 P.2d 1139 (Colo.1987), we re-examined the problem of the exposure of jurors to information not presented at trial. Wiser had been convicted of first degree burglary and felony menacing, following a jury trial. During Wiser’s trial, one juror asked a legal secretary about the source of jury instructions and another juror consulted a dictionary for a definition of burglary, one of the crimes with which the defendant was charged. This information was presented to the trial court by affidavit in support of a motion for new trial and was based on information gleaned from the jurors at an informal meeting after return of the verdict. The trial court held that even if the information was true, the evidence was insufficient to establish that the jury was prejudiced by the misconduct. The trial court therefore denied the new trial motion. The court of appeals affirmed. It recognized that the consultation of outside sources was misconduct but held that the defendant had the burden to establish that the misconduct was prejudicial. The court of appeals held that because the affidavits contained no information that the jurors’ exposure to the information actually influenced their votes, the defendant had failed to carry his burden of establishing prejudice. People v. Wiser, No. 82CA0990 (Colo.App. Sept. 6, 1984) (not selected for official publication).
Upon certiorari review in Wiser we explained the problems inherent in a standard that requires the defendant to show that exposure to information outside the record caused actual prejudice. A longstanding common law rule proscribes jurors’ testimony about their mental processes during their deliberations. See CRE 606(b), incorporating that rule.7 We noted the problem of proof created because this proscription prevents inquiry into the deliberative process of jurors even though such an inquiry is necessary to ascertain actual prejudicial effect. Wiser, 732 P.2d at 1141. CRE 606(b)’s prohibition of any juror testimony about the effect of the extraneous information precludes introduction of the only relevant evidence about actual prejudice. Id.
In recognition of these evidentiary constraints, Wiser substituted an objective [82]*82test in place of the actual prejudice standard by which to assess prejudicial effect when the issue arises after completion of the trial. The trial court must determine the effect of the extraneous information or influences on a typical jury. Id. at 1142. A new trial is necessary when there is a reasonable possibility that the verdict was tainted by the introduction of outside information or influences into the jury deliberations. Id. at 1143.8
III.
We must determine the appropriate manner of evaluating the prejudicial effect of media reports when the issue is presented during trial. The trial court refused to poll the jury because Harper had not presented independent evidence of the jurors’ exposure to the article. The court of appeals affirmed the trial court, ruling in reliance on People v. Holmes, 191 Colo. 477, 481, 553 P.2d 786, 789 (1976). It did not cite Wiser.
A.
In Holmes, decided before Wiser, we upheld the trial court’s refusal to poll the jury after the mid-trial publication of a newspaper article about the trial. Immediately before recess on the night preceding the publication of the article, the trial court instructed the jury to ignore newspaper articles concerning the trial and to refrain from listening to newscasts on radio and television. The defendant presented no independent evidence showing that any juror had read the newspaper article.9 Moreover, the trial court commented that the article in question could be interpreted as favorable to the defendant. Under these circumstances we adhered to the presumption adopted in Hammons v. People, 153 Colo. 193, 197-98, 385 P.2d 592, 595 (1963), that the jury had followed the court’s instruction. See also Tomsak v. People, 166 Colo. 226, 229-30, 442 P.2d 825, 827 (1968). Therefore, we held that polling the jury was unnecessary.
We conclude that the rationale of Wiser, recognizing the difficulty of obtaining proof of jury prejudice, undercuts the analysis in Holmes. The Holmes decision is based principally on the presumption that jurors followed the trial court’s instructions to ignore newspaper articles and the absence of a showing by the defendant that any of the jurors read the article in question. We characterized the defendant's motion to poll the jury as “based merely upon unsupported speculation that some member of the jury might have read the newspaper article and might have been prejudiced thereby.” Holmes, 191 Colo, at 481, 553 P.2d at 789. This reasoning is flawed in two respects.
First, the newspaper article could have come to the attention of a juror inadvertently. State v. Bey, 112 N.J. 45, 548 A.2d 846, 865 (1988). If an article appeared in an inconspicuous location in a newspaper, as was the case here, a juror could have read it in whole or in part before becoming aware that it related to the case being tried, or perhaps a family member, a coworker, or a friend could have mentioned the article. In such a situation the juror would have acquired the information without violating the court’s instructions.
Second, requiring independent evidence of the jury’s exposure to outside information as a prerequisite to polling the jury fails to acknowledge the significant obstacles to obtaining such evidence. Although CRE 606(b) permits juror testimony about exposure to extraneous prejudicial information, ethical rules prohibit counsel from communicating with jurors or their family members during the trial. See DR 7-108(B), (F);10 cf. People v. Radinsky, 176 [83]*83Colo. 357, 363, 490 P.2d 951, 953-54 (1971) (pretrial telephone inquiries by attorney’s investigator to jury panel members about their attitudes towards insurance companies, large verdicts, and whiplash injuries constituted jury tampering). This ban on direct contact with jurors significantly hampers a defendant’s ability to procure independent evidence of jurors’ actual exposure to prejudicial information. Realistically, such evidence would be unavailable unless volunteered. See Bey, 548 P.2d at 866. This leaves a defendant with little recourse when extremely prejudicial extraneous material has been published.
A presumption that jurors follow court instructions not to permit themselves to be exposed to media reports, therefore, does not adequately take into account either the likelihood that a juror could acquire information without violating the court’s instructions or the difficulty of discovering whether jurors were exposed to such a report. We conclude that to resolve the issue of prejudice on the basis of such a faulty presumption fails to provide adequate protection for the defendant’s constitutionally based right to a fair trial.
B.
The reasonable possibility of prejudice test articulated in Wiser, however, is not suitable for resolving the effect of media reports occurring during the trial when the issue of the prejudicial effect of such reports is presented before completion of the trial. The Wiser test requires a determination of whether there is a reasonable possibility that extraneous information or influence affected the verdict and requires for this purpose that the court consider the effect on a “typical jury.” Wiser, 732 P.2d at 1132. This test envisions a situation where, as in Wiser, the issue does not arise until the trial has been completed and the jury discharged. In Wiser we held that resolving the prejudice issue by the objective test enunciated in that case “is consistent with CRE 606(b)’s purpose of protecting the privacy of jurors,” and “is also more likely to enhance the stability of jury verdicts than an approach that required jurors to reconstruct conditions of jury deliberations and retrace their mental processes sometimes long after a verdict has been rendered.” Id., 732 P.2d at 1142-43. These considerations have no applicability when the possibility of prejudice is broached during the pendency of the trial.
C.
Other courts have considered how to evaluate the possibility of prejudice resulting from mid-trial media reports when the defendant raises that issue during the trial. We derive from these cases an approach that will be consistent with the rationale of Wiser but will create an opportunity for the trial court to determine whether the reports in fact prejudiced the deliberations of the particular jury trying the case. The United States Court of Appeals for the Second Circuit describes a three-step process well adapted to accomplish this objective:
The simple three-step process is, first, to determine whether the coverage has a potential for unfair prejudice, second, to canvass the jury to find out if they have learned of the potentially prejudicial publicity and, third, to examine individually exposed jurors — outside the presence of the other jurors — to ascertain how much they know of the distracting publicity and what effect, if any, it has had on the juror’s ability to decide the case fairly.
United States v. Gaggi, 811 F.2d 47, 51 (2d Cir.), cert. denied, 482 U.S. 929, 107 S.Ct. 3214, 96 L.Ed.2d 701 (1987). These inquiries must be conducted in the framework of the facts of each case, and the trial court has broad discretion in deciding the [84]*84ultimate issue of whether the media reports prejudiced the defendant’s right to a fair trial. Id. Many other courts have adopted essentially the same procedure and standards for evaluating prejudice in these circumstances. See, e.g., United States v. Thompson, 908 F.2d 648 (10th Cir.1990); Government of Virgin Islands v. Dowling, 814 F.2d 134, 139-41 (3d Cir.1987); United States v. Halbert, 712 F.2d 388, 389 (9th Cir.1983), cert. denied, 465 U.S. 1005, 104 S.Ct. 997, 79 L.Ed.2d 230 (1984); United States v. Trapnell, 638 F.2d 1016, 1022 (7th Cir.1980); United States v. Hood, 593 F.2d 293, 296 (8th Cir.1979); United States v. Crowell, 586 F.2d 1020, 1024 (4th Cir. 1978), cert. denied, 440 U.S. 959, 99 S.Ct. 1500, 59 L.Ed.2d 772 (1979); Brown v. State, 601 P.2d 221, 232 (Alaska 1979); Morris v. United States, 564 A.2d 746, 748 (D.C.App.1989); People v. Barrow, 133 Ill.2d 226, 139 Ill.Dec. 728, 549 N.E.2d 240, 256-67 (1989), cert. denied, — U.S. -, 110 S.Ct. 3257, 111 L.Ed.2d 767 (1990); State v. Bey, 112 N.J. 45, 548 A.2d 846, 866-67 (1988); cf. United States v. Herring, 568 F.2d 1099, 1104-05 (5th Cir.1978) (failure to conduct required inquiry of ju rors raises presumption of prejudice);11 People v. Adcox, 47 Cal.3d 207, 253 Cal.Rptr. 55, 763 P.2d 906, 931 (1988) (court must make whatever inquiry is reasonably necessary), cert. denied, — U.S.-, 110 S.Ct. 1500, 108 L.Ed.2d 634 (1990); Lynch v. State, 588 A.2d 1138, 1139 (Del.1991) (presumption of prejudice if there is a rea sonable probability of juror taint); Commonwealth v. Duddie Ford, Inc., 28 Mass.App.Ct. 426, 551 N.E.2d 1211, 1215 (1990) (voir dire required if a serious question of possible prejudice arises) (citing Commonwealth v. Jackson, 376 Mass. 790, 383 N.E.2d 835 (1978)). But see United States v. Metzger, 778 F.2d 1195, 1209 (6th Cir. 1985) (jury presumed to follow court’s instruction; defendant must show that jury violated the admonition), cert. denied, 477 U.S. 906, 106 S.Ct. 3279, 91 L.Ed.2d 568 (1986).
In implementing the first step, the trial court should focus principally upon whether the content of the media report is inherently prejudicial. Other relevant factors include whether the report contained information that would not be admissible at trial or that was not in fact adduced before the jury, Crowell, 586 F.2d at 1020, and how closely related the publicity is to the matters at issue in the trial, the timing of the publication, and the likelihood the jury was exposed to it, Brown, 601 P.2d at 232. Cf. People v. McCrary, 190 Colo. 538, 545, 549 P.2d 1320, 1326 (1976) (listing factors a court should evaluate to decide whether pre-trial publicity warrants venue change). The court also should consider the likely effectiveness of any instruction not to read, watch or listen to reports concerning the defendant or the trial in light of the nature and manner of dissemination of the news reports. The existence of admonitions, alone, does not sufficiently neutralize news reports in the community where the trial is being held that may reasonably be believed to have come to the attention of the jurors. Trapnell, 638 F.2d at 1022. See also Thompson, 908 F.2d at 651. Doubt about the existence of prejudice should be resolved by proceeding to step two and polling the jurors as a group. In many cases such a poll may disclose that the jurors did not learn of the report, making it unnecessary to question individual jurors. See United States v. Baker, 855 F.2d 1353, 1361 (8th Cir.1988), cert. denied, 490 U.S. 1069, 109 S.Ct. 2072, 104 L.Ed.2d 636 (1989). Cf. Gallegos v. People, 146 Colo. 546, 549, 362 P.2d 178, 179 (1961) (no merit to contention defendant was prejudiced by newspaper article admittedly not seen by any juror). When such individualized questioning is warranted, it should be conducted outside the presence of the other jurors. Ultimately, the trial judge, exercising informed discretion, must determine the prejudicial effect, if any, of the publicity. Gaggi, 811 F.2d at 51.
[85]*85IV.
The newspaper article published during Harper’s trial had great potential for unfair prejudice. The article discussed Harper’s prior conviction for sexual assault on a child, the same type of offense for which he was prosecuted in the present case. It also related that the incident upon which the prior conviction was based took place sometime between October 1 and November 30, 1983, the same period within which he allegedly committed the crime for which he was on trial. Evidence of prior criminal convictions is generally inadmissible because of its highly prejudicial effect. E.g., Martin v. People, 738 P.2d 789, 794 (Colo.1987); People v. Abbott, 690 P.2d 1263, 1269 (Colo.1984). “[EJvidence of pri- or criminality casts ‘damning innuendo likely to beget prejudice in the minds of juries.’ ” People v. Lucero, 200 Colo. 335, 343, 615 P.2d 660, 665 (1980) (quoting Stull v. People, 140 Colo. 278, 284, 344 P.2d 455, 458 (1959)) (disclosure of defendant’s two prior felony convictions to jury panel at inception of trial among the prejudicial errors requiring a new trial).
Even in the limited circumstances permitting admission of prior convictions, see, e.g., § 13-90-101 (prior convictions admissible to impeach a witness’ credibility), CRE 404(b) (past crimes admissible for purposes including proof of motive, plan, absence of mistake), stringent standards apply and cautionary instructions must accompany the introduction of such prior convictions. See, e.g., CJI-Crim. 4:07, 4:02, 2:02; Hollis v. People, 630 P.2d 68, 70 (Colo.1981); Stull v. People, 140 Colo. 278, 284-85, 344 P.2d 455, 458-59 (1959). The highly prejudicial information concerning Harper’s pri- or conviction for sexual assault on a child certainly would taint the jury verdict if the jurors had been exposed to the newspaper article. See Murphy v. Florida, 421 U.S. 794, 798, 95 S.Ct. 2031, 2035, 44 L.Ed.2d 589 (1975); Marshall v. United States, 360 U.S. 310, 312-13, 79 S.Ct. 1171, 1172-73, 3 L.Ed.2d 1250 (1959); Quintana v. People, 158 Colo. 189, 194, 405 P.2d 740, 743 (1965) (newspaper article described defendant in aggravated robbery and conspiracy to commit aggravated robbery trial as an ex-convict and said a co-defendant was serving a sentence for his part in the robbery).
The taint created by juror exposure to the newspaper article would likely have had an especially serious prejudicial effect since the evidence against Harper was equivocal. The eight-year-old victim and her younger sister shared a bedroom and a bed in Harper's home where the children were staying while their mother was away. The incident occurred in the middle of the night. Although the children were acquainted with Harper, neither child could identify the perpetrator. The victim described the perpetrator generally as a male Caucasian with a medium build and a full beard. Harper fit this description, but so did his brother. Evidence varied over whether the brother stayed at Harper’s house during the night in question. There was also conflicting evidence whether one of the doors to the house was locked. Although additional circumstantial evidence suggested that Harper committed the crime, the evidence of Harper’s guilt was not overwhelming. Evidence of his prior conviction for sexual assault on a child could have caused jurors to believe that Harper had a propensity to commit such a crime. The prior conviction would tend to substantiate the current accusation and cause the jury to overlook the inconclusiveness of the evidence against Harper. See Trapnell, 638 F.2d at 1023 (information about defendants’ prior air piracy convictions might have prevented the jury from weighing fairly the instant air piracy and related kidnapping charges). The record adequately establishes the prejudicial effect of the newspaper article.
The circumstances of the article’s publication present a reasonable possibility of jury exposure. The defendant did not assert mere vague speculation of possible prejudice. He identified a particular newspaper account having a specific and certain prejudicial content. The article appeared during the second day of trial in the local newspaper of a relatively small city. “[TJhe court must investigate the asserted impropriety upon merely a colorable show[86]*86ing of extrinsic influence.” United States v. Rowe, 906 F.2d 654, 656 (11th Cir.1990). On the other hand, the court had instructed the jury not to read media accounts of the trial. The article was short and located on an interior page of the paper, although the obscurity of the article may have led a juror to read it inadvertently. The trial court needed to evaluate these factual circumstances to determine the likelihood that the jury had been exposed to this significantly prejudicial material. See United States v. Williams, 809 F.2d 1072, 1092 (5th Cir.1987) (citing United States v. Herring, 568 F.2d 1099, 1105 (5th Cir.1978)), cert. denied, 484 U.S. 896, 108 S.Ct. 228, 98 L.Ed.2d 187 (1987).
In light of the article’s highly prejudicial contents, some direct inquiry of the jury was necessary. See People v. Swan, 130 A.D.2d 6, 519 N.Y.S.2d 581, 583 (1987). “The more serious the potential jury contamination, especially where alleged extrinsic evidence is involved, the heavier the burden to investigate.” United States v. Harris, 908 F.2d 728, 734 (11th Cir.1990) (quoting United States v. Caldwell, 776 F.2d 989, 998 (11th Cir.1985)), cert. denied, — U.S.-, 111 S.Ct. 979, 112 L.Ed.2d 1063 (1991). Polling the jury would provide an effective means of ascertaining that exposure. See ABA Standards for Criminal Justice: Fair Trial and Free Press 8-3.6(f) (2d ed.1986).12 See also, e.g., State v. Durr, 58 Ohio St.3d 86, 568 N.E.2d 674, 680 (1991). The trial court should have proceeded to step two of the procedure described by Gaggi and questioned the jurors collectively about their exposure to the article. Cf. United States v. Thompson, 908 F.2d 648, 650 (10th Cir. 1990) (“At a minimum the court had a duty to ask whether the jurors had read the article concerning this case.”). Cf. People v. Adcox, 47 Cal.3d 207, 253 Cal.Rptr. 55, 763 P.2d 906, 931 (1988) (failure to conduct reasonably necessary inquiry into juror prejudice would require reversal). Framing this inquiry in general terms would minimize any implication that an article containing information adverse to the defendant had been published.13 Courts in other jurisdictions have recognized the utility of inquiry of the entire jury collectively after the publication of prejudicial articles. See, e.g., United States v. Gaggi, 811 F.2d 47, 51 (2nd Cir.), cert. denied, 482 U.S. 929, 107 S.Ct. 3214, 96 L.Ed.2d 701 (1987); United States v. Trapnell, 638 F.2d 1016, 1022 (7th Cir.1980); United States v. Crowell, 586 F.2d 1020, 1024 (4th Cir.1978), cert. denied, 440 U.S. 959, 99 S.Ct. 1500, 59 L.Ed.2d 772 (1979); Brown v. State, 601 P.2d 221, 232 (Alaska 1979). Under the circumstances of this case, such collective questioning, in combination with the court’s admonition, could have sufficed to establish absence of prior or prospective juror exposure to the article. See United States v. Baker, 855 F.2d 1353, 1361 (8th Cir.1988). The trial court’s complete refusal to inquire about possible contamination constituted reversible error. See, e.g., Government of Virgin Islands v. Dowling, 814 F.2d 134, 141 (3d Cir.1987); United States v. Williams, 809 F.2d 1072, 1093 (5th Cir.1987), cert. denied, 484 U.S. 896, 108 S.Ct. 228, 98 L.Ed.2d 187 (1987); State v. Bey, 112 N.J. 45, 548 A.2d 846, 870 (1988).
Y.
We hold that the three-step procedure outlined in Gaggi applies to possible juror exposure to mid-trial prejudicial publicity. The article at issue here was prejudicial to the defendant. The trial court erred in requiring independent evidence of juror exposure before questioning the jury [87]*87about exposure to the article. We therefore reverse the judgment of the court of appeals and direct that court to reverse the judgment of conviction and remand the case for a new trial.
ROVIRA, C.J., dissents, and VOLLACK, J., joins in the dissent.