*401Francisco Javier Solorio appeals his conviction for first degree murder ( Pen. Code, § 187, subd. (a) ), arguing that the trial court erred in denying his motion for a new trial on grounds of jury misconduct. Although it denied his motion, the court made a factual finding that the jury discussed Solorio's decision not to testify "several times" despite repeated admonitions not to consider that topic. Prejudice from this type of misconduct is presumed, and on this record we cannot conclude the presumption of prejudice was rebutted. ( People v. Lavender (2014) 60 Cal.4th 679, 692, 181 Cal.Rptr.3d 28, 339 P.3d 318 ( Lavender ).) Accordingly, we reverse and remand the matter for a new trial.
FACTUAL AND PROCEDURAL BACKGROUND
A jury convicted Francisco Javier Solorio of the first degree premeditated murder of his neighbor, Albert Ramos. ( Pen. Code, § 187, subd. (a).) The prosecution presented evidence that Solorio, motivated by revenge, killed Ramos four months after Ramos stabbed Solorio's brother, Rudy.
Solorio and Ramos lived on same side of the same block in Brawley, California. Their residences were separated by two houses. Tamara R.1 and her daughter Sara J. lived across the street and were friends with both Solorio and Ramos. After Rudy was stabbed, Solorio told them he would handle matters on his own rather than approach law enforcement. Solorio told Sara he planned to "cap" Ramos, "blast *583him," and "kill him" for stabbing Rudy. When someone burned Ramos's car afterwards, Solorio told Tamara that things "would only get worse." *402In March 2013 Tamara and Sara had just returned from a one-hour shopping trip with Ramos. Both had a clear line of sight and saw the incident unfold as Ramos walked from their parked car toward his house across the street. As Ramos was walking, Solorio stepped outside his property, walked toward Ramos, pointed a gun wrapped in a bandana at him, and said, "remember what you did to my brother?" Ramos said, "Fuck you" and threw a plastic cup he was holding at Solorio. Solorio grabbed Ramos around the neck, and shot him three times in the arm, head, and chest, killing him.
At trial, Tamara and Sara testified that Ramos was not armed, did not threaten Solorio, and did not approach, confront, or hit Solorio at any point during the incident. Sara recalled saying, "Don't do it" loud enough for Solorio to hear when she saw him pull out the gun.
Tamara immediately called 911. Solorio could be heard in the background exclaiming, "He had a knife!" Tamara could be heard saying, "That's bullshit Javier. That's bullshit. I saw it." A knife was recovered at the scene, and Solorio had a superficial knife wound on his arm. The prints on the knife did not match Ramos or Solorio and instead were a possible match for Solorio's brother, Steven. Steven told the responding police officer Brian Smith that upon arriving home, he saw someone on top of his brother and heard three shots fired.
Solorio did not testify at trial, but his counsel argued the killing was in self-defense. In a taped police interview played for the jury, Solorio told officers about Ramos's drug use, assaultive behavior, and stabbing of Rudy four months earlier. Solorio stated that Ramos had repeatedly punched him and made him fear for his life. At trial, police officers testified that there were no signs of trauma or swelling on Solorio's face or bruises or scrapes on Ramos's hands to support the theory that Ramos attacked Solorio.
Solorio's brothers Rudy and Victor testified about Ramos's unprovoked stabbing of Rudy and threatening statements to Victor. A neighbor named Ofelia testified that Ramos was scaring her children, engaging in unusual behavior, and had attacked two of her uncles. One of those uncles testified about the attack, and a father and son in the neighborhood described Ramos's threatening behavior in the four months before his death. Ramos's toxicology results indicated that at the time of his death, he had methamphetamine and amphetamine in his bloodstream.
After the verdict, Solorio filed a motion for a new trial on grounds of jury misconduct, arguing that the jurors repeatedly discussed Solorio's decision not to testify during deliberations. The court held an evidentiary hearing and determined that although misconduct occurred, it was not prejudicial. On that basis it denied the motion for a new trial.
*403Solorio was sentenced to serve 50 years to life in prison, consisting of 25 years to life for first-degree murder, plus 25 years to life for the firearm enhancement. ( Pen. Code, § 12022.53, subd. (d).)
DISCUSSION
Additional Background
Solorio filed a new trial motion on the ground of juror misconduct. ( Pen. Code, § 1181, subd. (3) [allowing a new trial for jury "misconduct by which a fair and due consideration of the case has been prevented" ].) He argued that jurors violated the *584court's instruction not to consider Solorio's decision not to testify.2 In support of his motion Solorio attached a declaration by Juror 11, the foreperson, that stated:
"During the deliberations, the entire jury brought up and discussed the fact that the defendant Francisco Solorio did not testify in his own behalf. Some jurors were asking why the defendant did not testify. Some jurors thought that the defendant Solorio felt guilty and he knew he 'did it' and that is why he did not testify."
Solorio also submitted declarations by defense investigators, Juror 6, and (an unsigned declaration by) Juror 9 suggesting the jury had discussed "at length" that Solorio did not take the stand because he had something to hide.
The People opposed Solorio's motion on grounds the proffered declarations were inadmissible under Evidence Code section 1150, subdivision (a).3 In the alternative, the People urged the court to hold an evidentiary hearing to resolve disputed factual issues. ( People v. Hedgecock (1990) 51 Cal.3d 395, 415, 272 Cal.Rptr. 803, 795 P.2d 1260 [trial court may conduct an evidentiary hearing (a Hedgecock hearing) to resolve conflicting allegations of juror misconduct].)
The court determined that the quoted portion of Juror 11's declaration above was admissible, whereas declarations of Juror 9 and defense investigators were inadmissible. ( § 1150, subd. (a).) It found a material *404factual dispute between Juror 11's declaration and Juror 6's declaration and decided to hold a Hedgecock hearing.4 The court subpoenaed all 12 jurors to testify and followed a script, asking jurors if they recalled whether the topic came up, if they participated and how many participated in those discussions, how long the discussions were, and whether jurors were admonished.
The court found all of the jurors to be credible and rejected Solorio's attempt to impeach the credibility of Juror 9 with the defense investigator's testimony.
• Jurors 1, 2, 3, 4, 6, 8, and 9 had no recollection of the topic coming up.
• Juror 12 did not recall whether or not the topic came up.
• Juror 5 recalled the topic coming up briefly and that any discussion took "maybe a minute." He did not recall the foreperson or anybody else giving a reminder that jurors were not to discuss or consider that issue.
*585• Juror 7 recalled a "short discussion" lasting "several minutes" among "half" of the jurors but thought the discussion stopped once the foreperson stated it could not be considered.
• Juror 10 recalled a "really short" discussion after one juror brought it up, which stopped once she and Juror 11 reminded the others they could not consider it.
• Juror 11, the foreperson, recalled a group of six or seven jurors repeatedly discussing Solorio's decision not to testify. The discussion happened several times over the course of an hour. Each time the topic came up, Juror 11 and one or two other jurors would admonish the group that they could not consider that issue. The group of jurors would stop, talk about other things, and circle back to the topic. At some point after about an hour, all conversation on the subject stopped, and the matter never came up again during deliberations.
The court credited Juror 11's version of events that the matter came up more than twice during deliberations and was promptly admonished each *405time it came up. Although Juror 11's declaration was not received into evidence, the court considered his testimony in conjunction with his declaration.
Specifically, the court found:
"We do not have before us, although an inquiry was made of the jurors, when these discussions took place. Regardless of when they took place, they ceased upon admonishment and went on and continued their deliberations." [...]
"The discussions of this issue were raised several times over the course of about an hour within the 12 to 13 hours of deliberations. They were promptly admonished, promptly ceased discussion."
The court found that there was misconduct, giving rise to a presumption of bias :
"Was there misconduct in this case? Yes, it's obvious. Just the mere raising of the issue of the failure to testify is misconduct, even if it's stopped and only mentioned one time.... So there is a presumption of bias."
Thereafter, the court applied a two-part test from People v. Tafoya (2007) 42 Cal.4th 147, 64 Cal.Rptr.3d 163, 164 P.3d 5905 to assess whether the presumption of prejudice had been rebutted:
"Test one talks about a bias being found if the extraneous material judged objectively is inherently and substantially likely to have influenced a juror. [¶] And test two, if that test is found to be negative, you look at the totality of the circumstances."
The court concluded there was no prejudice under the first test because each time the topic came up, jurors "were promptly admonished" and "ceased discussion." It found support in People v. Avila (2009) 46 Cal.4th 680, 94 Cal.Rptr.3d 699, 208 P.3d 634 ( Avila ), which it read as indicating that the topic came up more than once and admonishments rebutted the presumption of prejudice.6 It also gave "great consideration"
*586to the fact that eight jurors *406had no recollection that the discussions occurred, stating that the discussion "was so insignificant at the time that they have no recollection of it now."
Turning to the second test, the court found "no substantial likelihood of bias" based on the totality of the circumstances:
"There were 12 or 13 hours of deliberations, yes. Was there misconduct? Yes. Was the issue discussed or raised by the jury that should not have been? Yes. Is the jury perfect? No. Are they human? Yes. Do they make mistakes? Yes. Was it inherently and substantially prejudicial? Based upon the testimony of the 12 jurors, and twice by Juror Number 9, this Court is not convinced, weighing it all, that there was any inherent and substantial likelihood of bias against the defendant in this particular case."
"Furthermore, this court, as noted in People v. Tafoya , considers the issue in light of the evidence in this particular trial. This was a case in which it was readily admitted from the beginning that the defendant shot the victim. No doubt that was admitted. The issue was the issue of self-defense. There was weeks and weeks of testimony and numerous witnesses we heard on the issue and several eyewitnesses were brought from out of state, a mother and daughter that testified. So the Court has as well in its determination of this issue considered the evidence in this particular case against the defendant."
"Recalling as well that the defendant did speak during the trial, in effect, because his interview was played and his statements to the officer were played or stated to the jury as to his profession of self-defense. So the Court did weigh all of the testimony and evidence received as well in its determination to deny the defendant's motion for new trial."
The parties agree that jury misconduct occurred but disagree whether it was prejudicial. Solorio argues that the presumption of prejudice was not rebutted because the jury repeatedly discussed his silence at trial despite admonishments. The People argue that even accepting Juror 11's testimony that the topic was discussed several times, prompt admonitions were sufficient to rebut the presumption of prejudice.
Analysis
The jury was instructed with CALCRIM No. 355 not to "consider, for any reason at all, the fact that the defendant did not testify" and not to "discuss that fact during your deliberations or let it influence your decision in any way." The trial court found that jurors engaged in misconduct when they ignored the instruction and discussed Solorio's failure to testify several times over the course of an hour during deliberations. ( *407Lavender, supra, 60 Cal.4th at p. 687, 181 Cal.Rptr.3d 28, 339 P.3d 318.) Each time, the foreperson admonished other jurors not to consider the topic, and they would move on to other issues before circling back to it.
We accept the court's credibility determinations and factual findings if supported by substantial evidence but independently assess whether prejudice arose. ( Avila, supra, 46 Cal.4th at pp. 726-727, 94 Cal.Rptr.3d 699, 208 P.3d 634 ;
*587Lavender, supra, 60 Cal.4th at p. 682, 181 Cal.Rptr.3d 28, 339 P.3d 318.)7 "When the record shows there was misconduct, the defendant is afforded the benefit of a rebuttable presumption of prejudice." ( Cissna, supra, 182 Cal.App.4th at p. 1116, 106 Cal.Rptr.3d 54.) "This presumption is provided as an evidentiary aid to the defendant because of the statutory bar against evidence of a juror's subjective thought processes and the reliability of external circumstances to show underlying bias." ( Ibid. ; § 1150, subd. (a).)
As one court recently put it, "[t]he law concerning prejudice lacks clarity." ( People v. Echavarria (2017) 13 Cal.App.5th 1255, 1264, 221 Cal.Rptr.3d 608 ( Echavarria ).) Whereas some cases focus on whether the presumption of prejudice arising from misconduct has been successfully rebutted , others consider whether the prejudice was sufficiently substantial to warrant reversal. ( Ibid. [collecting cases]; see People v. Von Villas (1995) 36 Cal.App.4th 1425, 1445, 43 Cal.Rptr.2d 233 (dis. opn. of Woods, J.).) As we discuss, it is also unclear whether in the context of the type of misconduct present here, we should, as the trial court did, consider the strength of trial evidence against the defendant in assessing prejudice.
Lavender involved the same type of jury misconduct. During deliberations, jurors commented on the defendants' failure to testify. ( Lavender , supra , 60 Cal.4th at p. 687, 181 Cal.Rptr.3d 28, 339 P.3d 318.) The Supreme Court explained that the presumption of prejudice arising from such misconduct "may be rebutted by an affirmative evidentiary showing that prejudice does not exist or by a reviewing court's examination of the entire record to determine whether there is a reasonable probability of actual harm resulting from the misconduct." ( Ibid. , italics added.) In another case involving the same type of jury misconduct, our high court stated that the presumption of prejudice is rebutted if a review of the entire record demonstrates no substantial likelihood of actual harm. ( People v. Leonard (2007) 40 Cal.4th 1370, 1425, 58 Cal.Rptr.3d 368, 157 P.3d 973 ( Leonard ).)
*408Neither Lavender nor Leonard actually considered the strength of the evidence on the question of guilt in assessing prejudice. Lavender remanded the matter to the trial court to resolve factual ambiguities in the record to assess prejudice. ( Lavender , supra , 60 Cal.4th at p. 693, 181 Cal.Rptr.3d 28, 339 P.3d 318.) It provided several factors that could on remand establish that no prejudice actually resulted from juror comments on the defendant's decision not to testify. ( Id. at pp. 687-692, 181 Cal.Rptr.3d 28, 339 P.3d 318, discussed post .) Nowhere did the opinion direct the trial court to consider the strength of the trial evidence against defendants in assessing prejudice. Likewise, Leonard found the presumption of prejudice rebutted because jurors did not draw adverse inferences from the defendant's decision not to testify. ( Leonard , supra , 40 Cal.4th at p. 1425, 58 Cal.Rptr.3d 368, 157 P.3d 973.) Leonard's review was confined to what happened during deliberations, not the trial evidence.
*588The parties have not cited, nor have we found, a case that evaluates prejudice from the type of misconduct we have here based on the strength of the evidence at trial. In suggesting that a review of the "entire record" is appropriate, Lavender and Leonard cite two cases involving altogether different forms of jury misconduct. In People v. Danks (2004) 32 Cal.4th 269, 8 Cal.Rptr.3d 767, 82 P.3d 1249, a juror's circulation of biblical verses during penalty phase deliberations was found not prejudicial given "compelling" penalty phase evidence presented. ( Danks , at pp. 305, 307, 8 Cal.Rptr.3d 767, 82 P.3d 1249.) In Hasson v. Ford Motor Co. (1982) 32 Cal.3d 388, 185 Cal.Rptr. 654, 650 P.2d 1171, a juror's reading a novel and doing crossword puzzles while evidence was being presented at trial was found not prejudicial given "overwhelming proof of liability" presented at trial. ( Hasson , at pp. 415, 417, 185 Cal.Rptr. 654, 650 P.2d 1171.)
We question whether we may, in this context, consider the strength of the trial evidence of Solorio's guilt. As Lavender recognized, "a defendant's decision not to testify[ ] [is] a topic that does 'not involve extra record material' and ... concerns a matter 'already obvious to the jurors.' " ( Lavender , supra , 60 Cal.4th at p. 692, fn. 2, 181 Cal.Rptr.3d 28, 339 P.3d 318.) Lavender seems to hint that the prejudice analysis may differ depending on the type of jury misconduct at issue. ( Ibid . ["We are not presented with, and therefore do not discuss, the proper analysis of a jury's improper receipt of information from 'extraneous sources.' " ] ) Here, the misconduct does not go to juror bias from considering extraneous information, but instead to a defendant's constitutional right against self-incrimination.
"[T]he purpose of the rule prohibiting jury discussion of a defendant's failure to testify is to prevent the jury from drawing adverse inferences against the defendant, in violation of the constitutional right not to incriminate oneself." ( Leonard , supra , 40 Cal.4th at p. 1425, 58 Cal.Rptr.3d 368, 157 P.3d 973.) A criminal defendant has a Fifth Amendment right "to remain silent 'unless he chooses to speak in *409the unfettered exercise of his own will' " -a right that would be meaningless if jurors could draw adverse inferences from his or her choice. ( Carter v. Kentucky (1981) 450 U.S. 288, 305, 101 S.Ct. 1112, 67 L.Ed.2d 241 ; Leonard , at p. 1424, 58 Cal.Rptr.3d 368, 157 P.3d 973.) Juror 11's declaration indicates that jurors drew precisely the adverse inference that the Fifth Amendment eschews, stating Solorio did not testify because "he knew he 'did it.' "8
Jury misconduct in this case went directly to the ultimate issue-Solorio's guilt for murder. In In re Stankewitz (1985) 40 Cal.3d 391, 220 Cal.Rptr. 382, 708 P.2d 1260, the Supreme Court cautioned that the presumption of prejudice is stronger when "the misconduct goes to a key issue in the case." ( Id. at p. 402, 220 Cal.Rptr. 382, 708 P.2d 1260 [juror's misconduct in stating during deliberations that he had been a police officer for 20 years and knew a robbery had taken place went to the ultimate issue of guilt for felony-murder, raising presumption of prejudice].) When the presumption of prejudice is stronger, the prosecution must put forth greater proof to rebut it. ( *589Echavarria , supra , 13 Cal.App.5th at pp. 1267-1268, 221 Cal.Rptr.3d 608.)
We will follow Lavender's approach and consider whether the prosecution rebutted the presumption of prejudice with evidence that no prejudice actually resulted. Applying each of the Lavender rebuttal factors, we conclude the prosecution has fallen short.
The first rebuttal factor considers whether jurors drew adverse inferences of guilt from the defendant's decision not to testify. ( Lavender , supra , 60 Cal.4th at pp. 689-690, 692, 181 Cal.Rptr.3d 28, 339 P.3d 318.) Comments of mere "wonderment and curiosity" are normally innocuous. ( People v. Hord (1993) 15 Cal.App.4th 711, 727, 19 Cal.Rptr.2d 55 ( Hord ).) "It is natural for jurors to wonder about a defendant's absence from the witness stand." ( People v. Loker (2008) 44 Cal.4th 691, 749, 80 Cal.Rptr.3d 630, 188 P.3d 580 ( Loker ).) " '[M]erely referencing that they wish he would have testified is not the same as drawing negative inferences from the absence of testimony.' " ( Leonard , supra , 40 Cal.4th at p. 1425, 58 Cal.Rptr.3d 368, 157 P.3d 973.) Thus, for example, in Leonard the court found no prejudice from the jury's brief discussion where there were no adverse inferences expressed regarding the defendant's decision not to testify. ( Ibid. )
"When comments go beyond natural curiosity and their content suggests inferences from forbidden areas, the chance of prejudice increases." ( Hord , supra , 15 Cal.App.4th at p. 728, 19 Cal.Rptr.2d 55.) "For example, if a juror were to say, 'The *410defendant didn't testify so he is guilty,' or 'we will have to find the defendant guilty of the greatest charges to ensure he will be adequately punished,' the comments go beyond mere curiosity and lean more toward a juror's drawing inappropriate inferences from areas which are off limits." ( Ibid. ) That is the box we find ourselves in-Juror 11's declaration makes it clear that jurors were drawing adverse inferences of guilt. Thus, the first rebuttal factor was not met. ( Lavender , supra , 60 Cal.4th at p. 690, 181 Cal.Rptr.3d 28, 339 P.3d 318 ; Hord , at p. 728, 19 Cal.Rptr.2d 55.)
The second rebuttal factor considers the length of discussion about the topic. ( Lavender , supra , 60 Cal.4th at p. 692, 181 Cal.Rptr.3d 28, 339 P.3d 318 [noting disputed factual issue as to whether the issue was merely mentioned or discussed at greater length].) In Hord , the court found the presumption rebutted where the juror merely made an "oblique remark about a party not saying anything to protect himself" and there was no apparent lengthy discussion into inappropriate areas. ( Hord , supra, 15 Cal.App.4th at pp. 727-728, 19 Cal.Rptr.2d 55.) "Transitory comments" about the topic are usually innocuous, "particularly when a comment stands alone without any further discussion." ( Ibid. ; see Leonard , supra , 40 Cal.4th at p. 1425, 58 Cal.Rptr.3d 368, 157 P.3d 973 ["brief discussion" of the topic was not prejudicial].) The fact that only a few jurors recall any comment on the topic may tend to "indicate[ ] that the discussion was not of any length or significance." ( Avila , supra , 46 Cal.4th at p. 727, 94 Cal.Rptr.3d 699, 208 P.3d 634.)
Here, the court made a factual finding based on Juror 11's testimony that the topic was raised "several times over the course of about an hour within the 12 to 13 hours of deliberations." The trial court did not make a factual finding as to how many jurors participated in the discussion, but it noted that Juror 11 put the number at six or seven jurors, whereas eight jurors had no recollection of any *590discussion on the topic.9 Although the inferences are not entirely consistent, on this record we are unable to conclude that the discussion about the defendant's choice not to take the stand was so transitory or brief that it rebutted the presumption of prejudice.
The third rebuttal factor-the crux of Lavender -considers whether jurors were reminded not to consider the defendant's decision not to testify. ( Lavender , supra , 60 Cal.4th at p. 690, 181 Cal.Rptr.3d 28, 339 P.3d 318.) If the foreperson promptly reminded jurors when the improper statement was made, that would "(in the absence of objective evidence to the contrary) ... constitute strong evidence to rebut the presumption of prejudice." ( Ibid. ; see Loker , supra , 44 Cal.4th at p. 749, 80 Cal.Rptr.3d 630, 188 P.3d 580 [prompt admonition rebutted presumption of prejudice]; Hord , supra , 15 Cal.App.4th at p. 711, 19 Cal.Rptr.2d 55 [same]; Avila , supra , 46 Cal.4th at pp. 725-727, 94 Cal.Rptr.3d 699, 208 P.3d 634 [same; the topic came up once in the guilt phase and once in the penalty phase after the verdict but the juror was admonished both times].)
*411By contrast, a prompt reminder would be in sufficient to rebut the presumption of prejudice if there were objective circumstances indicating that the reminder was ineffective. ( Lavender , supra , 60 Cal.4th at p. 692, 181 Cal.Rptr.3d 28, 339 P.3d 318.) As the Supreme Court explained in Lavender , "a persistent refusal [by the jury] to follow the court's instructions would tend to confirm the prejudicial effect of the misconduct claimed by defendants." ( Ibid . )10 For example, in Cissna , a juror's daily conversations with a nonjuror about topics that included the defendant's failure to testify was prejudicial in the face of repeated instructions by the trial judge not to discuss the case. ( Cissna , supra , 182 Cal.App.4th at pp. 1111, 1119, 106 Cal.Rptr.3d 54.) Likewise, an agreement among jurors to disregard the court's instructions would be prejudicial misconduct requiring reversal. ( Krouse v. Graham (1977) 19 Cal.3d 59, 81, 137 Cal.Rptr. 863, 562 P.2d 1022.)
Here, the trial court found that jurors repeatedly discussed Solorio's decision not to testify. Despite admonitions each time the topic was brought up, jurors circled back to the topic "several times." There was, therefore, "objective evidence establishing a basis to question the effectiveness of the reminder." ( Lavender , supra , 60 Cal.4th at p. 687, 181 Cal.Rptr.3d 28, 339 P.3d 318.) Where repeated reminders were given that Solorio's lack of testimony could not be considered, we cannot say they were sufficient to rebut the presumption of prejudice.
To the extent it is appropriate to consider the overall strength of the evidence against Solorio presented at trial, we would still find the presumption unrebutted. If the trial evidence were so overwhelming, why would the jury repeatedly discuss their view that Solorio must be guilty because he did not take the stand? Given the nature of improper statements here-that Solorio did not testify because he was guilty-and the repeated revisiting of this matter over a one-hour period despite *591admonitions, we would find at least a "reasonable probability" that actual harm resulted from the misconduct even if we were to consider the "entire record." ( Lavender , supra , 60 Cal.4th at p. 687, 181 Cal.Rptr.3d 28, 339 P.3d 318 ; cf. People v. Nesler (1997) 16 Cal.4th 561, 583, 66 Cal.Rptr.2d 454, 941 P.2d 87 ["repeated references" to improper subjects during deliberations established a substantial likelihood of prejudice].)
We recognize that "[j]urors, like all human beings are imperfect" and that "the criminal justice system must not be rendered impotent in quest of an ever-elusive perfection." ( Lavender , supra , 60 Cal.4th at p. 681, 181 Cal.Rptr.3d 28, 339 P.3d 318 ; In re Carpenter (1995) 9 Cal.4th 634, 654, 38 Cal.Rptr.2d 665, 889 P.2d 985.) Nevertheless, on the record before us, we find no basis to find that the *412prosecution rebutted the presumption of prejudice. Solorio is entitled to a new trial due to prejudicial juror misconduct, and we reverse his conviction for retrial. ( Stankewitz , supra , 40 Cal.3d at p. 402, 220 Cal.Rptr. 382, 708 P.2d 1260.)11
DISPOSITION
The judgment is reversed and the matter is remanded for a new trial.
WE CONCUR:
HUFFMAN, Acting P. J.
IRION, J.
As of January 1, 2017, California Rules of Court, rule 8.90 (Rule 8.90 ) became effective. Rule 8.90(b) requires appellate courts to "consider referring to" certain individuals "by first name and last initial or, if the first name is unusual or other circumstances would defeat the objective of anonymity, by initials only," in order to protect those individuals' privacy. Although the rule lists victims, not witnesses, as subject to the rule, we refer to the two eyewitnesses in this case by their first names and last initials, and thereafter by their first names only, to protect their privacy.
Solorio also based his new trial motion on one juror's hearing difficulties but does not raise that issue on appeal.
Further references are to the Evidence Code, unless otherwise specified. Section 1150, subdivision (a) provides: "Upon an inquiry as to the validity of a verdict, any otherwise admissible evidence may be received as to statements made, or conduct, conditions, or events occurring, either within or without the jury room, of such a character as is likely to have influenced the verdict improperly. No evidence is admissible to show the effect of such statement, conduct, condition, or event upon a juror either in influencing him to assent to or dissent from the verdict or concerning the mental processes by which it was determined." (See People v. Cissna(2010) 182 Cal.App.4th 1105, 1116, 106 Cal.Rptr.3d 54 (Cissna ) ["No evidence may be presented concerning the subjective reasoning processes of a juror that can neither be corroborated nor disproved...." ].)
The court found a material disputed fact because Juror 6's declaration did not reference any discussion on defendant's decision not to testify. In addition, Juror 11's declaration did not address how many jurors participated in the improper discussion, how long the discussion lasted, whether jurors were admonished not to consider defendant's decision not to testify, and if so, whether discussions on the topic ended after the admonishment.
In Tafoya, a juror discussed the Catholic church's opinion about the death penalty with a priest and presented that opinion to the jury during penalty phase deliberations. (People v. Tafoya, supra, 42 Cal.4th at p. 190, 64 Cal.Rptr.3d 163, 164 P.3d 590.) The court considered whether, on review of the "entire record," there was "a substantial likelihood of juror bias." (Id. at p. 192, 64 Cal.Rptr.3d 163, 164 P.3d 590.)
Avila involved the same type of jury misconduct at issue here-discussion of defendant's decision not to testify. The trial court misread Avila in concluding that jurors in the case had discussed the topic "three or four times." As we read it, two jurors recalled the issue coming up once in the guilt phase-"the offending juror was immediately reminded he could not consider this factor and the discussion ceased." (Avila, supra, 46 Cal.4th at p. 727, 94 Cal.Rptr.3d 699, 208 P.3d 634.) Thereafter, in the penalty phase, "the comment was not made until after the verdict was reached and the bailiff contacted." (Ibid. ) Again, "the offending juror was reminded that this consideration was not permitted." (Ibid. )
The People question Juror 11's account that the topic was discussed more than once. However, the testimony of Juror 11 alone is sufficient to support the court's finding that the topic came up "several times" and that the offending jurors were promptly admonished each time. (In re Marriage of Mix(1975) 14 Cal.3d 604, 614, 122 Cal.Rptr. 79, 536 P.2d 479.) We therefore defer to this finding. (Cissna, supra, 182 Cal.App.4th at p. 1118, 106 Cal.Rptr.3d 54 ; Avila, supra, 46 Cal.4th at pp. 726-727, 94 Cal.Rptr.3d 699, 208 P.3d 634.)
The trial judge noted that although Juror 11's declaration had not been received into evidence, he had read it, and "obviously it affects my consideration of the testimony by Juror Number 11...." Because it appears the court considered this declaration in making its findings, we consider it on appeal. We disregard, however, Solorio's citation to declarations deemed inadmissible by the trial court.
Juror 7 corroborated Juror 11, testifying that "about half the jurors" engaged in the discussion.
In Lavender, there was an unresolved factual question as to whether there was "objective indication that one or more jurors was unable or unwilling to follow the court's instructions once reminded of them." (Lavender, supra, 60 Cal.4th at pp. 691-692, 181 Cal.Rptr.3d 28, 339 P.3d 318.) The court directed the trial court on remand to address this factor in making its findings.
Solorio also challenges the trial court's exclusion of a defense expert's testimony attempting to quantify Ramos's risk for drug-induced irrational violence at the time of the shooting. (See generally People v. Wright(1985) 39 Cal.3d 576, 582-584, 217 Cal.Rptr. 212, 703 P.2d 1106.) Because we reverse based on juror misconduct, we have no need to address this additional argument.