OPINION
MANNHEIMER, Judge.
Following a jury trial, Kevin W. Pease and Marvin L. Roberts were convicted of robbing and murdering a teenage boy, and also assaulting an adult, Franklin Dayton, in a separate incident that same night. A more detailed description of these events is contained in Pease v. State, 54 P.3d 316, 319-321 (Alaska App.2002)-the opinion in which we affirmed Pease's and Roberts's convictions.1
About six months after this Court affirmed Pease's and Roberts's convictions, two journalism students from the University of Alaska Fairbanks reported that the jurors had engaged in a group experiment during their deliberations. Further investigation revealed the details of this experiment: During the jury deliberations, the jurors asked the bailiff to take them outside to the northwest corner of the Anchorage courthouse (the corner of Third Avenue and I Street). Two of the jurors then crossed to the north side of Third Avenue and walked west to the location formerly occupied by the Elevation 92 restaurant-that is, to a point a few yards west of the intersection of Third Avenue and K Street, approximately 480 feet from the other jurors. So positioned, the two groups of jurors looked at each other across this distance to see if they could recognize each other. (The significance of this experiment will be explained shortly.)
Having learned all this, Pease filed a petition for post-conviction relief, arguing that he was entitled to a new trial because the jurors [987]*987conducted this unauthorized visual experiment. The superior court ultimately agreed. Relying on this Court's decision in Gorz v. State, 749 P.2d 1349, 1355 (Alaska App.1988), Superior Court Judge Ben J. Esch concluded that the jurors committed misconduct when they performed this experiment, and that the result of this experiment had probably influenced the jury's verdict. Judge Esch therefore granted Pease's petition for post-conviction relief and ordered a new trial.
We conclude that the superior court erred in ordering a new trial. Our conclusion is based on a three-part analysis.
We conclude that the jurors could properly conduct this type of experiment, and that there would have been no error if the experiment had been conducted within the confines of the jury room.
We further conclude that the jurors committed misconduct when they failed to seek court permission to leave the jury room and go outside to perform the experiment. But not all acts of jury misconduct require a new trial.
Here, the jurors' act of going outside to conduct the experiment did not make the results of their experiment less reliable or valid. Indeed, as we explain here, conducting the experiment outside helped make the results more reliable and valid.
Thus, because the jurors were entitled to conduct this type of experiment in the first place, we conclude that the fact that this experiment took place outside did not prejudice the fairness of Pease's trial.
Underlying facts
One of the important government witnesses at trial was Arlo Olson, who testified that he saw Pease, Roberts, and two other accomplices (Eugene Vent and George Frese) commit an assault upon Franklin Dayton down the street from the Eagles Hall in Fairbanks.
Olson was attending a late-evening wedding reception at the Eagles Hall. According to his testimony, he was standing on the front steps of the hall between 12:80 and 1:00 in the morning when Pease, Roberts, and the two other men drove up in Roberts's car.
Olson was not well-acquainted with Pease and Roberts, but Olson had known one of the other men, Eugene Vent, since high school, and Olson was friends with the fourth man, George Frese. Frese asked Olson if he wanted to get high with them, but Olson declined, so the four men drove off.
Olson testified that, about half an hour later, while he was still on the front steps of the Eagles Hall, he saw the four men attack Franklin Dayton as Dayton walked down the street. Olson testified that he saw Pease push Dayton to the ground, and then all four men began kicking Dayton. When one of the four men yelled, "Give me your fucking money, bitch," Dayton handed something to one of the four. The four men then ran to Roberts's car and drove away.
Pease's defense was alibi; he claimed to have been elsewhere that night. According ly, Pease's attorney attacked Olson's identification of Pease as one of the assailants.
Part of this attack took the form of cross-examination. Olson admitted that he had drunk a considerable quantity of alcoholic beverages on the evening in question, and he further admitted that he had smoked marijuana earlier in the day. Moreover, Olson admitted that the attack on Dayton had lasted only about thirty seconds, and that the four assailants had had their backs to Olson for much of this time. Olson conceded that he did not get a good look at the four men until they were running back to their car.
The defense also introduced expert testimony to attack Olson's identification of Pease. Olson testified that he observed the attack on Dayton from about 400 feet away. To rebut Olson's testimony that he could identify the four assailants from a distance of 400 feet, the defense presented Dr. Geoffrey Loftus, an expert on human visual perception and memory. Dr. Loftus testified that it was virtually impossible for a human being to distinguish one person from another at that distance-indeed, at even half that distance.
According to Dr. Loftus, "[the] ability to distinguish [any one person] from somebody else-in other words, to recognize them later on-is essentially nil" beyond 200 feet. Lof-[988]*988tus claimed that this was true even when under the best of cireumstances-that is, even when the lighting is good, and the observer has "unlimited time" to make the observation, and when the observer is "in an optimal physiological state" to make the observation ("awake, alert, [and] not ill, drunk, [or] under the influence of any drugs").
This dispute as to whether Arlo Olson's identification of Pease was humanly possible figured prominently in the summations delivered by the prosecutor and the defense attorneys at the close of the trial. As Judge Esch noted in his decision, "all counsel discussed the testimony of these two witnesses and attempted to [either] bolster or denigrate its value."
During their deliberations, the jurors attempted to test this issue by looking out of the jury room window to see if they could recognize people at a distance. (The jury room windows in the Anchorage courthouse face north; these windows overlook city streets, a parking lot, and (farther in the distance) portions of the Alaska Railroad yard and the Port of Anchorage.) However, this effort to discern people's features at a distance through the jury room window proved unfruitful because the jurors had differing estimates concerning how far away a particular observed person was. To better investigate this issue, the jurors decided to go outside and view each other across a distance that could be paced off.
It was in this context that the jurors left the courthouse and conducted their experiment. Apparently, the experiment had mixed results: at least six of the jurors indicated that they were able to recognize each other across this distance, but at least one juror acknowledged that he could not.
In Pease's petition for post-conviction relief, he characterized the jurors' experiment as an attempt to assess the credibility of a key government witness, Arlo Olson. But as Judge Esch recognized, the cireumstances of the jurors' experiment show that the experiment was only peripherally relevant to assessing the credibility of Olson's testimony.
As explained above, Olson had been drinking, and he made his observations in the middle of the night (in October, when it was dark). The jurors made no attempt to replicate these conditions, even though these factors were obviously important to assessing the trustworthiness of Olson's identification. Rather, based on the circumstances of the experiment and the jurors' accounts of their actions, Judge Esch found that the primary purpose of their experiment "was ... to test the validity of Loftus'[s] testimony". That is, the main purpose of the experiment was to test Loftus's assertion that it was not humanly possible, even under the best conditions, to recognize another person at a distance of greater than 200 feet.
Cases upholding the authority of a jury to engage in experimentation to test the factual assertions made by witnesses or the theories suggested by the attorneys
Centuries ago, the role of the jury was to investigate allegations of eriminal behavior and then render a decision based on their investigation.2 In modern times, however, the parties-that is, the government and the defendant-are in charge of investigating and presenting the facts of the case (with the trial judge having a residual authority to call additional witnesses and to cross-examine the witnesses presented by the parties).3
Because the jury no longer has a formal investigative role, courts often declare that juries are required to base their decisions solely on the evidence presented in court, [989]*989and that juries are barred from independently investigating the case or from otherwise considering extraneous information (.e, information that was not elicited during the trial).4 Here, for example, is what this Court said on this subject in Gorz v. State:
The law [on this point] is well settled. Jurors have a duty to consider only the evidence presented in open court.... Evidence that has not been subjected to the procedural safeguards of trial impinges on the constitutional rights to confrontation, cross-examination, and counsel.
Gorz, 749 P.2d 1349, 1355 (Alaska App.1988).
But this is an over-simplification. The law does, in fact, sometimes allow jurors to rely on information that was obtained outside of the court proceedings.
The Alaska Supreme Court has held that jurors are allowed to discuss, and to rely on, any "pre-existing ... knowledge of a general nature" that individual jurors bring to the deliberative process. Titus v. State, 963 P.2d 258, 262 (Alaska 1998).
For example, in Titus, the supreme court approvingly cited a case where, during deliberations, an individual juror shared his expertise in X-ray technology. Id. at 262.5 And, employing the Titus rule, this Court held that it was not misconduct for individual jurors to share their personal knowledge of how loud a shot from a .22 caliber rifle would be, or their knowledge of the breakage characteristics of the type of glass used in construction vehicles. See Larson v. State, 79 P.3d 650, 654 (Alaska App.2003).
See also Brooks v. Zahn, 170 Ariz. 545, 826 P.2d 1171, 1176-78 (Ariz.App.1991), where one of the jurors was a registered nurse who relied upon her medical knowledge and experience during the jury's deliberations. The Arizona court held that the juror's prior knowledge was not "extraneous". Similarly, in State v. Aguilar, 169 Ariz. 180, 818 P.2d 165 (Ariz.App.1991), one of the jurors was a medical doctor. He "shared with the other jurors [his] fund of knowledge ... regarding alcohol and cocaine ... blackouts", and he also told the other jurors that he disagreed with the defendant's expert witness who had testified that the defendant lacked criminal intent because of his alcohol and cocaine consumption. Id. at 166-67. The Arizona court held that the doctor's knowledge was not "extraneous" information.
As can be seen from these examples, a juror's personal knowledge of such topics may be quite important, even crucial, in evaluating the credibility or plausibility of the testimony presented during the trial. Nevertheless, this juror knowledge is not considered "extraneous" information, even though it is obtained extrajudicially. It is not misconduct for jurors to share this information with each other, nor is it misconduct for the jurors to rely on this information during their deliberations. Titus, 963 P.2d at 262.
There is another line of authority, more pertinent to Pease's case, that addresses the issue of jury experimentation during deliberations. Courts have repeatedly upheld jurors' efforts to test the credibility or plausibility of trial testimony by experimenting with items of physical evidence admitted during the trial, or by re-enacting the events or conditions described by witnesses.
An early example of this principle is found in Rex v. Smith (1915), a case in which the defendant was charged with murdering his wife by drowning her in their bathtub. The bathtub was produced as an exhibit, and during the trial the jury foreman informed the judge that "[one of the jury hald] expressed a wish that someone should be put in the bath for ocular demonstration." The judge replied, "I can only suggest to you that when you examine these baths in your private room, you should put one of yourselves in." 6
[990]*990Twenty years earlier, on this side of the Atlantic, the Virginia Supreme Court upheld an even more intrusive jury experiment in Taylor v. Commonwealth, 90 Va. 109, 17 S.E. 812 (1893). Taylor was a murder prosecution in which the government introduced expended cartridges that bore particular marks left by the murder weapon's "plunger"-what we now call the firing pin. The defendant responded by introducing evidence that the firing pin of his rifle did not make such marks on expended cartridges. During their deliberations, the jurors called for the defendant's rifle to be brought to the jury room, and then they dismantled the weapon to see if someone had tampered with the firing pin. (The jurors discovered that, indeed, the firing pin "had been recently tampered with and fixed for the occasion of the trial".) Rather than condemning the jurors' action, the Virginia Supreme Court commended the "intelligent and serutinizing jury" for making this discovery. Id., 17 S.E. at 816.
Several more recent cases are listed and described in this Court's decision in Bowlin v. State, 823 P.2d 676, 679 (Alaska App.1991).
In People v. Kurena, 87 Ill.App.3d 771, 43 Ill.Dec. 277, 410 N.E.2d 277 (1980), jurors constructed a cardboard replica of a knife that had been admitted into evidence, and then they used this replica to re-enact the assault, to see if the victim's wounds could have been inflicted by a left-handed or right-handed person, and to see if the weapon could have been concealed in a sleeve.
In State v. Thompson, 164 Mont. 415, 524 P.2d 1115 (1974), jurors used a handgun that had been admitted into evidence to re-enact the struggle described in the testimony, to see if the gun could have been fired if it was held in the manner described in the testimony. -_
In Allen v. State, 141 Tex.Crim. 94, 146 S.W.2d 384, 386 (1940),7 the jury experimented with a pistol admitted into evidence to see if, as claimed by the defendant, the cylinder would "hang" or stick at a particular place in its revolution and cause the handgun to discharge accidentally.
In State v. Best, 89 S.D. 227, 232 N.W.2d 447, 457 (1975), a defendant on trial for child abuse claimed that her infant's injuries had been inflicted when the baby's two-year-old brother struck the baby with a telephone. The jurors experimented with the telephone (which had been admitted into evidence), testing its weight to see if it was conceivable that a two-year-old could have used it to inflict serious injury.
In People v. Engler, 150 A.D.2d 827, 540 N.Y.S.2d 591, 594 (1989), jurors experimented with a vaporizer to test the defendant's claim that a child's injuries had been sustained when the child carelessly played near the vaporizer.
And in Taylor v. Reo Motors, Inc., 275 F.2d 699, 705 (10th Cir.1960), the jurors dismantled and reassembled a heat exchanger to test a witness's testimony about the way it functioned.
In addition to these cases listed in Bowlin, several other appellate decisions have approved experiments conducted by the jury during deliberations.
In State v. Balisok, 123 Wash.2d 114, 866 P.2d 631, 663-34 (1994), the court held that the jurors' re-enactments of the alleged struggle between the defendant and the vice-tim were not "extrinsic evidence" and thus did not constitute jury misconduct.
In United States v. Hephner, 410 F.2d 930, 936 (7th Cir.1969), the court upheld a jury experiment in which one juror covered his head and donned sunglasses so that the jurors could evaluate whether it was possible to identify a person who was disguised in this manner.
In People v. Agado, 964 P.2d 565, 567-68 (Colo.App.1998), the court upheld the jurors' experimentation with a gun admitted into evidence to determine its trigger pull, and thus to evaluate defendant's testimony.
In State v. Chamberlain, 112 N.M. 723, 819 P.2d 673, 676, 683 (1991), a case in which the defendant claimed self-defense, and in which the government and the defendant offered [991]*991competing versions of when a police officer had pulled his gun from its holster, the jurors experimented with the gun and holster to ascertain the sound made when the gun was removed from the holster, and then they listened for that sound when they reviewed the audio recording of the encounter between the officer and the defendant.
In United States v. Abeyta, 27 F.3d 470, 477 (10th Cir.1994), the defendant was charged with stabbing another man during a confrontation outside a bar. Using a pocket knife owned by one of the jurors, the jurors engaged in a re-enactment of the events de-seribed by the witnesses. The cireuit court of appeals held that this jury experiment did not entitle the defendant to a new trial. The court stated, "There is simply no constitutional command preventing a jury from using common sense and ordinary and uninflammsa-tory props to reenact a crime in the privacy of the jury room."
In United States v. Silks, unpublished, 1988 WL 141090 (9th Cir.1988), the defendant was charged with conspiracy to unlawfully import controlled substances. The government introduced a composite drawing of the pilot of the airplane that was used to transport the controlled substances. However, in conformity with witness accounts, the pilot depicted in this composite drawing was clean shaven, while Silks was wearing a beard during the trial. During their deliberations, the jurors requested additional copies of a composite drawing of the pilot. At least one member of the jury drew a beard on the drawing-presumably to see if the drawing, so modified, resembled the bearded Silks.
The Ninth Cireuit upheld this jury experiment: "It is not misconduct when the jury uses its own senses in weighing the evidence presented to it. The jurors must be given enough latitude to permit them to use common experiences and illustrations in reaching their verdict.... The jurors had observed Silks, and the type of beard he was presently wearing, in the courtroom during trial. The jurors' own attempts to evaluate the evidence to reach a verdict do not rise to the level of juror misconduct." 1988 WL 141090 at *2.
In United States v. Avery, 717 F.2d 1020, 1026 (6th Cir.1983), the defendant was charged with attempting to blow up a building. The police received a late-evening burglar alarm signal from a medical office building; within minutes, an officer arrived to investigate and found the defendant crouched down outside a storage area at the rear of the building. A subsequent inspection of the building revealed that the rear storage door had been pried open. Inside a crawl space underneath the building, the police found a substantial quantity of flammable materials, including four canisters of propane, two gallon jugs and fifteen milk cartons containing gasoline, newspapers (some of which had address labels listing the defendant's name and address), and a roll of wax paper.
During the summations to the jury, the defense attorney argued that it was impossible that the defendant could have hidden all of these materials in the crawl space within the approximately. three minutes between the time the burglar alarm went off and the time the police arrived to investigate. The prosecutor responded that he thought it would be possible, and he described how he thought it could be done. Id. at 1026.
On appeal, the defendant claimed that the prosecutor's remark was, essentially, an invitation to the jurors to conduct an unauthorized experiment. The cireuit court of appeals concluded that, even if the jurors had experimented with the evidence in the manner suggested by the prosecutor, this would have been proper:
The record reveals that the defense, through expert testimony, put into issue the question of how long it would take to place the materials into the crawl space. Indeed, the defense theory of the case was that the defendant could not possibly have moved and assembled all of the materials in a three[-]minute period. This argument was repeated by defense counsel on at least two occasions during his closing argument. Under these circumstances, we believe the prosecutor could respond to [the] defendant's argument by asking the jury to handle the milk cartons and recreate portions of the defendant's actions. Even assuming the jury did recreate the defendant's actions, we find no error in such conduct.... [J]urors must be given [992]*992enough latitude in their deliberations to permit them to use common experiences and illustrations in reaching their verdict.
Avery, 717 F.2d at 1026.
See also People v. Smith, 59 N.Y.2d 988, 466 N.Y.S.2d 662, 453 N.E.2d 1079 (1983), where the New York Court of Appeals applied this same principle to an experiment conducted by a single juror. One of the disputed issues at Smith's trial was a police officer's purported ability to observe the interior of a car through its rear window. During deliberations, one of the jurors tested the officer's testimony by looking through the rear windows of various cars, both while the juror was walking to dinner with the other jurors and while the juror was riding in a bus with the other jurors to their hotel. The New York high court upheld the juror's experiments.
In each of the cases discussed above, juries performed experiments that could theoretically be viewed as infringing the defendant's rights to confrontation and cross-examination. Moreover, in many instances, the circumstances or conditions of the jury's experiment arguably-or even clearly-deviated from the cireumstances or conditions of the events being litigated. This leads to an obvious difficulty: because the jury experiment takes place in private, neither the judge nor the defense attorney can point out the potential weaknesses in the jury's method of procedure, or point out the cirenmstances or conditions that the experiment omits or fails to take adequate account of, or point out the potential ambiguities in the experiment's result.
Given these problems, one might argue that juries should always be barred from engaging in unsupervised experimentation. And yet no rule of prohibition emerges from these cases. Instead, these courts have condoned jury experiments-sometimes fairly elaborate jury experiments-under the principle that "[JJurors must be given enough latitude in their deliberations to permit them to use common experiences and illustrations in reaching their verdict." United States v. Hephner, 410 F.2d at 936.
In Judge Stewart's dissent, he notes that many of these jury experiment cases were decided before this Court issued its decision in Gorz. Judge Stewart then concludes that "it is [not] beneficial to rely on ... authority [from other jurisdictions] when the rule adopted in Gorz is clear".
It may be true that the rule stated in Gorz is "clear" and uncomplicated, but this is only because Gorz does not accurately describe the complexity of the law on this point.
In Gorz, the discussion of the issue of jury experimentation is both short and concluso-ry:
The law governing this issue is well settled. Jurors have a duty to consider only the evidence presented in open court. Thus, a juror commits misconduct by conducting an unauthorized experiment and either personally relying on the results or communicating [those results] to other members of the jury. Evidence that has not been subjected to the procedural safeguards of trial impinges on the constitutional rights to confrontation, cross-examination, and counsel. Turner v. Louisiana, 379 U.S. 466, 472-73, 85 S.Ct. 546, 549-550, 13 L.Ed.2d 424 (1965).
[[Image here]]
In the present case, ... a member of [the] jury committed misconduct by personally determining the amount of time it would take to walk from the scene of the explosion to the Alaska Motor Inn. This point is not seriously disputed by the state.
Gorz, 749 P.2d at 1855.
The Gor opinion does not acknowledge any of the court decisions discussed above (or any others like them). In these decisions-many of which were issued decades before Gorz-courts upheld jury experiments that were similar in nature to the experiment conducted by the jurors in Pease's case.
Judging from this Court's comment in Gorg that the claim of juror misconduct "[was] not seriously disputed by the state", it appears that this Court may not have received adversarial briefing on the issue of what types of jury experimentation are lawful-and, thus, this issue may not have received focused attention. But in any event, a review of the cases in this area demonstrates [993]*993that the bright-line rule stated in Gorz does not accurately reflect the law on this subject.
Application of this law to Pease's case
Under the cases discussed above, if a question had arisen in Pease's case as to whether a particular placard was legible from ten feet away, or whether the color of a particular object could be discerned in dim light, there would be no inmipropriety in the jury's experimenting with these objects in the privacy of the jury room.
The experiment conducted in Pease's case presents even less cause for concern in terms of a defendant's rights of confrontation and cross-examination. The focus of the jury's inquiry was not a case-specific assertion about the quality or condition of a particular object or event. Rather, as Judge Esch found, the jury's attention was primarily focused on testing the generalized assertion made by Pease's expert witness, Dr. Lof-tus-the assertion that it is impossible, even under the best conditions, for one human being to recognize another at distances of more than 200 feet.
It appears that the jurors first attempted to test Dr. Loftus's assertion in the jury room-by looking out of the window to see if they could recognize people at a distance. Under the cases discussed above, this type of experiment was proper. The jurors took advantage of an experience afforded by everyday life-the experience of observing people at a distance from the window of a multistory building-to directly test the accuracy or inaccuracy of Dr. Loftus's assertion.
(Indeed, this Court has little doubt that, by this point in our discussion, many of our readers have been tempted-or have already succumbed to the temptation-to peer out of their own windows and try the experiment for themselves.)
When this question concerning the limits of human perception became a disputed issue at Pease's trial, the jurors were undoubtedly entitled to rely on their own experience when they discussed and ultimately assessed the credibility of Dr. Loftus's assertion. And, based on the cases discussed above, the jurors were also authorized to make test observations through the jury room window to help them resolve this question.
The real problem in this case arises because the jurors disagreed in their estimates as to how far away the observed people were. The jurors therefore decided to move the test outside, where they could get a better measurement of the distance.
By leaving the jury room, the jurors engaged in conduct that courts are much less likely to condone. We acknowledge that, in numerous instances, courts have held that jurors committed misconduct when they ventured out of the jury room to view the scene of the crime, or to assess how long it would take to travel from one place to another, or to conduct other case-specific experiments. Our own decision in Gorz v. State is a prime example.
We nevertheless conclude that the jurors' conduct in Pease's case presents a materially different situation from the ones described in the preceding paragraph. As Judge Esch found, the cireumstances surrounding the experiment show that the jurors were not attempting to re-create Arlo Olson's nighttime observation of the assault on Franklin Dayton. Rather, the jurors were attempting to test Dr. Loftus's general assertion about the limits of human perception.
Conceivably, when diéagreement arose among the jurors concerning the distance between the jury room window and the people whom the jurors were observing through that window, the jurors might have resolved this matter by deferring to the judgement of individual jurors who had more experience viewing people and objects at a distance-for example, a pilot or a soccer coach,. If the jurors had done so (thus obviating the need for them to leave the jury room), this would be an easier case.
But the jurors (for whatever reason) were unable to resolve their disagreement by discussion and debate. So, instead, they decided to renew their observations outside, where they could use a more objective method for measuring the distance.
It is true that the jurors conducted this experiment without the authorization of the court. But we believe that, given the cireum-[994]*994stances, reasonable judges would have allowed the experiment if the jurors had asked. Because the main objective was to test Dr. Loftus's assertion about human perception under optimal conditions (rather than to recreate Arlo Olson's personal experience of viewing other people at a particular distance on the night of the crimes), it did not matter that there were obvious differences between the jurors' experiment and the actual events being litigated.
On this point, see Bierria v. Dickinson Manufacturing Co., Ltd., 36 P.3d 654, 658-59 (Alaska 2001), a case that addresses the issue of what courts mean when they insist that the conditions of a proposed experiment be "substantially similar to the conditions at the time of the event in issue." 8 In Bierria, our supreme court emphasized that the concept of "substantial similarity" hinges on what the experiment is offered to prove. Often, there are identifiable dissimilarities between the cireumstances of the experiment and the real-life event being litigated; nevertheless, the experiment is still "substantially similar" if these dissimilarities are irrelevant to what the proponent of the experimental evidence is trying to prove.
Here, the jurors wished to test Dr. Lof-tus's assertion that it was impossible, even under the best of conditions, to recognize another person at a distance of greater than 200 feet. Given this purpose, it is irrelevant that the jurors' observations of each other, in daylight on Third Avenue in Anchorage, bore obvious dissimilarities to the nighttime observation described by Arlo Olson in his testimony.
For this same reason, the fact that neither Pease nor his defense attorney were present to witness and critique the experiment is not particularly important. In Pease's pleadings (both in the superior court and on appeal), he argues that the jurors' experiment was irremediably flawed because the jurors may have failed to perceive, or may have misper-ceived, crucial differences between the circumstances and conditions of their experiment and the cireumstances and conditions of Arlo Olson's observation on the night in question. But, as explained above, we conclude that these differences are immaterial.
We note, moreover, that Pease and his attorney would not have been present if the jurors had followed the lawful course of remaining inside the jury room and resolving their disagreement about the distances involved through discussion and debate, or by deferring to the judgement of individual jurors who were deemed to have better skill at estimating distances.
In sum, we conclude that the jurors' experiment was a proper type of inquiry for the jury to make, even though the jurors committed misconduct when they left the jury room for this purpose without the court's permission.
The remaining question is whether the jurors' act of leaving the jury room without authorization requires reversal of Pease's convictions.
As this Court noted in Gorz, 749 P.2d at 1855, not all acts of jury misconduct require reversal of the jury's verdict. In fact, our supreme court has declared that "generally the verdict should stand unless the evidence clearly establishes a serious violation of the juror's duty and [this violation of duty] deprives a party of a fair trial." Fickes v. Petrolane-Alaska Gas Service, Inc., 628 P.2d 908, 910 (Alaska 1981), quoting West v. State, 409 P.2d 847, 852 (Alaska 1966).9
We conclude that the act of jury misconduct presented in Pease's case-ie., the act of leaving the jury room without authorization to conduct an experiment that was otherwise proper-is not a "serious violation of the [jurors'] duty" for purposes of the Fickes/ West test.
Moreover, even assuming that the act of leaving the jury room was a serious violation of the jurors' duty, we further conclude that the jury's experiment did not deprive Pease of a fair trial. As we have explained, the jurors' primary purpose was to test Dr. Lof-[995]*995tus's general assertion about the limits of human perception under optimal conditions. The experiment, in itself, was proper. And, given the purpose of the experiment, there is no reason to believe that the experiment yielded a false or misleading result (4.e., a result that the jurors could not understand or meaningfully evaluate).
Thus, even though the result of the experiment may have influenced the jury's ultimate decision, the fact that the jurors conducted this experiment does not require reversal of Pease's convictions.
Conclusion
For the reasons explained here, the superi- or court's decision to grant Pease's petition for post-conviction relief is REVERSED, and Pease's convictions are reinstated.
COATS, Chief Judge, concurring.
STEWART, Judge, dissenting.