Commonwealth v. Hoose , 467 Mass. 395 ( 2014 )

Spina, J.

In this case, the defendant, Robin Anthony Hoose, was convicted of two counts of murder in the first degree, one for the death of Irene Pierce on the theory of extreme atrocity or cruelty and one for the death of Frank Blanchard on the theories of extreme atrocity or cruelty and deliberate premeditation. On appeal, the defendant asserts error in (1) the judge’s denial of the defendant’s motion to suppress his statements to police, (2) the judge’s denial of the defendant’s motion for a change of venue, (3) the judge’s denial of the admission of certain “third-party culprit” evidence and of the defendant’s request for a third-party culprit instruction at trial, and (4) the judge’s grant of the Commonwealth’s motion to exclude expert testimony regarding the phenomenon of false confessions. The defendant also asks us to exercise our power under G. L. c. 278, § 33E, to vacate his convictions or to reduce them to murder in the second degree. We affirm the defendant’s convictions and decline to grant relief under G. L. c. 278, § 33E.

1. Background. We recite the facts the jury could have found at trial, reserving additional facts for our analysis of the issues raised on appeal. At approximately 12:30 a.m. on July 17, 2007, the defendant encountered Bill Goly and Goly’s girl friend, Emma Oliver, outside a tavern. Oliver testified at trial that she and Goly then accompanied the defendant to his apartment and there encountered the eventual victims, Oliver’s friend, Irene Pierce, and Pierce’s boy friend, Frank Blanchard.

While at the defendant’s apartment, a man in a red hat arrived to whom the defendant sold counterfeit drugs. The defendant, Goly, and Oliver then left Pierce and Blanchard at the defendant’s apartment, drove to the home of another individual, and obtained two bags of “crack” cocaine, which they smoked together. The three then returned to the defendant’s apartment sometime between 3 and 3:30 a.m., at which point Pierce and Blanchard informed the defendant that the man to whom he had sold the purported drugs earlier in the evening had returned to the apartment. The defendant became angry. He showed Blanchard where he kept a gun in the closet, and he showed the group a large knife, which he then placed on a dresser in the living room.

The defendant, Goly, and Oliver again left the defendant’s apartment and purchased another bag of crack cocaine, which the three used together. Sometime later, the three returned to the defendant’s apartment. Oliver went through the living room to use the bathroom. The lights were off in the living room, and she did not notice anything unusual in the apartment at that time. The three then spent some time on a porch outside the defendant’s apartment before Goly and Oliver departed between 4 a.m. and 4:30 a.m.

Sometime after Goly and Oliver left, the defendant went back inside his apartment. Although the defendant believed he had asked Pierce and Blanchard to leave his apartment earlier that night, the defendant found Pierce and Blanchard inside his apartment watching television in the living room. Blanchard was on a bed, and Pierce was on a couch wrapped in what the defendant described as his “good blanket.” When the defendant then noticed that his watch and ring were missing from where he had left them in the apartment, the defendant became extremely angry and agitated. He began screaming and flew into a violent rage. He hit Blanchard, rendering him unconscious, and then picked up a knife and stabbed Pierce repeatedly in the abdomen. When Blanchard began to regain consciousness, the defendant stabbed him repeatedly as well. Pierce died from loss of blood, and Blanchard died from multiple blunt and sharp force injuries.

Subsequently, the defendant covered Pierce’s body with a blanket and Blanchard’s face with a pillow. The defendant wrote in a notebook the date, “7/17/07,” and the words, “Sorry mom . . . Love yous. Wrong place, wrong time for them.” The defendant also cut himself on the neck and both arms with a box cutter. The defendant was later able to make his way to his mother’s apartment in the same building, and she accompanied him to the emergency room. The defendant was then admitted to the psychiatric ward of the Franklin Medical Center as a result of his self-inflicted injuries.

The following day, the defendant, with the assistance of his mother, contacted Detective Leon Laster of the Montague police department. The defendant had a long-standing relationship with Laster. The defendant indicated on the telephone that he wished to speak with Laster in person, stating that he had something “very serious” and “very big” to tell him. Laster proceeded to the hospital, and when he arrived, he noted dried blood on the defendant’s neck and bandages on his arm. Laster indicated that he was in a hurry and asked what the defendant needed to tell him. The defendant began to cry and, with his head in his hands, told Laster, “There are two dead bodies in my house.” Laster noted that the defendant did not appear to be under the influence of drugs or alcohol although he was crying and at times sobbing during their conversation. Laster then asked if the defendant had killed the people, and the defendant stated that he had. The defendant told Laster, “Well, I know I didn’t shoot them because I think I stabbed them.” The defendant also told the detective that the body of the female victim was in his living room where the defendant had covered the body with a sheet and the body of the male victim was on the bed. Later that day, the bodies of Pierce and Blanchard were discovered by police in the defendant’s apartment as he had described.

Upon his discharge from the hospital on July 19, the defendant was met by three police officers, including Detective Lieutenant John Gibbons of the State police, who was leading the investigation. At their request, the defendant accompanied police to the Greenfield police station. After reading the defendant Miranda warnings and obtaining a written waiver, police conducted a series of interviews that began at approximately 10:30 a.m. and lasted until approximately 3:50 p.m., when the defendant was placed under arrest. Throughout these interviews the defendant professed to not remembering either the events of July 17 or the statement he had made to Detective Laster the previous day. However, following his arrest, during booking at the police station, the defendant recalled much of what had occurred on July 17 and requested to speak with police once more. The defendant was again apprised of his Miranda rights as well as his right to prompt presentment and a probable cause hearing within twenty-four hours. See Commonwealth v. Rosario, 422 Mass. 48, 56 (1996); Mass. R. Crim. P. 3.1 (a) (3), 442 Mass. 1503 (2004). After waiving these rights, the defendant provided a detailed statement admitting to having killed both Pierce and Blanchard. All of the interviews of the defendant and his statements to police were recorded and admitted into evidence at trial.

2. Motion to suppress defendant’s statements to police. Prior to trial, the defendant moved to suppress his statements to police on the grounds that the statements were obtained in violation of his right to remain silent and his right to counsel under Miranda v. Arizona, 384 U.S. 436, 444-445 (1966). The defendant further argued that neither his multiple waivers of the Miranda rights nor his statements to police were voluntary and intelligent because he was experiencing extreme physical and emotional trauma stemming from his recent injuries and because he was under the influence of prescription medications that were not properly administered by the police. Following an evidentiary hearing, the judge denied the motion to suppress and issued written findings and rulings concluding that the defendant had voluntarily and knowingly waived his rights under Miranda and that his statements to police were made free from coercion and intimidation.

“In reviewing a ruling on a motion to suppress evidence, we accept the judge’s subsidiary findings of fact absent clear error,” and we defer to the judge’s determination of the weight and credibility to be given to oral testimony presented at a motion hearing. Commonwealth v. Contos, 435 Mass. 19, 32 (2001), quoting Commonwealth v. Eckert, 431 Mass. 591, 592-593 (2000). Where a judge’s findings are based on recorded evidence, we are in as good a position as the motion judge to evaluate that evidence. However, where, as here, the judge considered the recorded evidence in light of oral testimony at the motion to suppress hearing and made credibility determinations therefrom, we adhere to the normal standard of review. Commonwealth v. Clarke, 461 Mass. 336, 340-341 (2012). We conduct an independent review of the judge’s application of constitutional principles to the facts found. Commonwealth v. LeBeau, 451 Mass. 244, 254 (2008).

The judge concluded first that the defendant was not in custody at the police station until he gave his first written statement and was arrested, and therefore that the Miranda decision did not apply to the defendant’s statement at the hospital or at the police station prior to his arrest.¹ The judge further concluded that even if the defendant had been in custody at the police station prior to his arrest, the police did not undermine the defendant’s right to counsel, the defendant voluntarily waived his rights under the Miranda decision, and his statements were not the product of coercion or intimidation. There was no error in the judge’s denial of the defendant’s motion to suppress.

a. Right to counsel. Once an accused person invokes the right to counsel, all interrogation must cease. Edwards v. Arizona, 451 U.S. 477, 484-485 (1981). Commonwealth v. Morganti, 455 Mass. 388, 396 (2009), S.C., ante 96 (2014). The invocation should be clear enough that a “reasonable police officer in the circumstances would understand the statement to be a request for an attorney.” Id. at 397, quoting Davis v. United States, 512 U.S. 452, 459 (1994).

According to the judge’s findings, all of which are supported by the record, appointed counsel arrived at the Greenfield police station on July 19 while the defendant was being interviewed by police. Detective Lieutenant Gibbons, who was interviewing the defendant at the time, immediately informed the defendant that the attorney had arrived. The defendant expressed surprise when told of the attorney’s presence, asking who was paying for the representation and stating that he would meet with the attorney to find out “what [he] wanted.” The defendant and the attorney met privately for fifteen to twenty minutes after which the attorney told Gibbons that he had advised the defendant not to answer any more questions.² The attorney then left.

At this point, Gibbons returned to the interview room and informed the defendant that he did not want to know what the defendant and the attorney had discussed. Gibbons then read the defendant Miranda warnings again and asked how the defendant wished to proceed. The defendant stated that he wished to speak with Detective Laster. At this point, Laster joined the interview. The defendant was then provided with Miranda warnings once again, and he again waived his rights. Although the defendant still claimed not to recall either his conversation with Laster the previous day or the events of the July 17, the defendant requested paper and pen to write out a statement.

While the defendant was writing the statement, Officer Chad Sumner of the Greenfield police sat in the interview room with the defendant and assisted when the defendant asked how to spell certain words. After completing his written statement, the defendant made several comments expressing concern that the attorney he had met with earlier in the day would be “angry” that he had written a statement. In response, the officer stated, “Well, ultimately, you’re your own boss,” and, “Sometimes [attorneys] give good advice, but you’re your own man.”

In his amended motion to suppress, the defendant argued that he invoked his Miranda rights through his attorney’s statements to Gibbons following the attorney’s meeting with defendant. On appeal, the defendant further argues that his meeting with the attorney itself constituted an unambiguous invocation of his right to counsel, which was undermined both by Gibbons continuing to question the defendant after his meeting with counsel and by Sumner’s comments.

We defer to the judge’s findings, based on an assessment of the credibility of testimony at the motion to suppress hearing, that the attorney told Gibbons only that he had advised the defendant not to speak with the police, not that the defendant had invoked his right to remain silent or to have an attorney present during any further questioning. Indeed, even if the judge had found that the attorney directed the police not to question the defendant any further, we have held that any directive of an attorney need only be honored for so long as to allow the defendant to consult with the attorney. Commonwealth v. Vao Sok, 435 Mass. 743, 751-752 (2002). Furthermore, merely meeting with an attorney is insufficient to constitute an unambiguous invocation of a defendant’s right to remain silent or to have an attorney present during questioning. Miranda rights belong to the defendant, not to the attorney, and they are the defendant’s rights to invoke or waive on his own accord. Commonwealth v. Currie, 388 Mass. 776, 783 (1983). Where a defendant chooses to waive his right to counsel after having met with an attorney and receiving Miranda warnings again, there is no constitutional violation. Vao Sok, supra at 752. Currie, supra. Therefore, Gibbons did not undermine that right by returning to the interview room to ask the defendant how he wished to proceed.

We further agree that although Sumner’s comments regarding the assistance of an attorney would have been better left unsaid, the comments did not undermine the defendant’s right to counsel. The comments were made after the defendant had signed a second Miranda waiver form and drafted his first written statement. Therefore, the comments did not constitute inappropriate police pressure on the defendant’s decision whether to invoke or waive his rights. Additionally, the police honored the defendant’s rights under art. 12 of the Massachusetts Declaration of Rights by informing the defendant immediately that an attorney had arrived at the station and wished to consult with him and by ceasing questioning and enabling the defendant to meet with the attorney promptly and in private. See Commonwealth v. Mavredakis, 430 Mass. 848, 861 (2000). Therefore, the judge did not err in ruling that the police did not undermine the defendant’s right to counsel.

b. Voluntariness of waivers and statements. In denying the defendant’s motion to suppress, the judge also concluded that the defendant’s waivers of his Miranda rights and his statements to police were voluntary and intelligent. We agree with the judge’s assessment.

A defendant’s waiver of Miranda rights must be knowing, intelligent, and voluntary. Commonwealth v. Gaboriault, 439 Mass. 84, 89 (2003). Similarly, due process requires that a criminal defendant’s statements be voluntarily made, as a product of rational intellect and free will, and not as a result of “inquisitorial activity” by the government such as coercion or threats. Commonwealth v. Walker, 466 Mass. 268, 276-277 (2013). Commonwealth v. Selby, 420 Mass. 656, 662 (1995).

Although we inquire separately into the voluntariness of the defendant’s waiver of Miranda rights and the voluntariness of the statements, both inquiries require us to examine the totality of the circumstances surrounding the making of the statements to ensure that the defendant’s will was not overborne. See Walker, 466 Mass. at 276-277; Commonwealth v. Scoggins, 439 Mass. 571, 576 (2003); Selby, 420 Mass. at 663. Factors relevant to this analysis include, but are not limited to, “promises or other inducements, conduct of the defendant, the defendant’s age, education, intelligence and emotional stability, experience with and in the criminal justice system, physical and mental condition, the initiator of the discussion of a deal or leniency (whether the defendant or the police), and the details of the interrogation, including the recitation of Miranda warnings.” Commonwealth v. Mandile, 397 Mass. 410, 413 (1986).

In thorough written findings, all of which are supported by the record, the judge properly applied this analysis in concluding that the defendant voluntarily waived his Miranda rights and that his statements were made free from coercion or intimidation. Based on the evidence presented at the suppression hearing, including the testimony of Detective Lieutenant Gibbons and recordings of each interview of the defendant at the police station, the judge found that the police made no promises or other improper inducements, and that the defendant was a forty-four year old, self-sufficient man with some high school education who was highly experienced in dealing with law enforcement and the criminal justice system. The judge also found that the defendant was interviewed in an office setting, that the police did not employ trickery or coercion to overcome the defendant’s will, that the tone of questioning was not hostile or heated, and that the defendant was advised fully of his Miranda rights, both verbally and in writing more than once. The defendant was also immediately informed of his attorney’s arrival at the police station and given a full opportunity to consult with him. Finally, the judge found that the defendant was given frequent breaks to smoke and to use the bathroom, that the defendant was not under the influence of alcohol, that his medications were not withheld from him, and that the medications did not adversely affect his thinking or ability to act volitionally.

Although the defendant contends that his waivers and subsequent statements were involuntary in part because his medications were improperly administered by the police, the record supports the judge’s conclusion that the medications were not withheld and that the police made an effort to administer the medications as prescribed throughout the day.³ The record also supports the judge’s conclusion that the defendant did not manifest or report any ill effects after consuming the medication as prescribed.

In sum, the totality of the circumstances surrounding the questioning of the defendant supports the judge’s findings that the defendant’s multiple waivers of his Miranda rights and subsequent statements to police were voluntary, knowing, and intelligent. We affirm the judge’s denial of the defendant’s motion to suppress.

3. Motion for change of venue. Prior to trial, the defendant moved for a change of venue on the basis that publicity surrounding the case was so pervasive and prejudicial in content that it had tainted the jury pool in the community and would deprive the defendant of his right to a trial by an impartial jury. The defendant renewed this motion orally throughout jury selection and in writing on the last day of jury selection.

A judge may transfer a case for trial if there exists in the community where the prosecution is pending so great a prejudice against the defendant that the defendant may not be able to obtain a fair and impartial trial by jury. Mass. R. Crim. P. 37 (b) (1), 378 Mass. 914 (1979). A judge has substantial discretion in deciding whether to grant a motion for a change of venue based on pretrial publicity, and we review such a decision for an abuse of discretion. Commonwealth v. James, 424 Mass. 770, 775 (1997).

Here, the judge denied the defendant’s initial motion in a written ruling on the basis that pretrial publicity was not so extensive or inflammatory as to taint the entire jury pool and that, based on the totality of the circumstances, the judge believed that it would be possible to empanel a fair and impartial jury using thorough voir dire procedures. She deferred judgment on the renewed motions during jury selection to determine whether an impartial jury could in fact be seated, and she denied the defendant’s renewed motion on the final day of jury selection once the jury had been empanelled. The defendant argues on appeal that these rulings denied him of his right to a fair trial as guaranteed by art. 12 and by the Sixth Amendment to the United States Constitution. We disagree.

In a motion for a change of venue, the defendant has the burden to establish the “solid foundation of fact” necessary to support a grant of the motion. Commonwealth v. McCowen, 458 Mass. 461, 476 (2010), quoting Commonwealth v. Clark, 432 Mass. 1, 6 (2000). To establish prejudice stemming from extensive pretrial publicity or settled community opinion, the defendant must show either presumptive prejudice or actual prejudice. Commonwealth v. Toolan, 460 Mass. 452, 462 (2011). The defendant failed to establish either.

a. Presumptive prejudice. Presumptive prejudice occurs when the jury pool in the community has been so tainted by pretrial publicity that the entire venire may be presumed prejudiced regardless of the specific voir dire procedures utilized. Id. at 463. However, prejudice is only presumed in truly extraordinary circumstances where the “trial atmosphere [is] utterly corrupted” by media coverage. Skilling v. United States, 561 U.S. 358, 380 (2010), quoting Murphy v. Florida, 421 U.S. 794, 798-799 (1975). See Commonwealth v. Entwistle, 463 Mass. 205, 221 (2012), cert. denied, 133 S. Ct. 945 (2013). Two factors play a central role in creating the presumption of prejudice. Toolan, supra at 464. Commonwealth v. Morales, 440 Mass. 536, 540, 541 (2003). First, the nature of the pretrial publicity, specifically whether it is both extensive and sensational, is a highly significant factor. Toolan, supra. Second, whether the judge was in fact able to empanel jurors who appear impartial is the factor of “primary importance.” Id.

Here, the defendant failed to establish that the publicity leading up to the trial was both extensive and sensational. We have held that publicity is not extensive where the nature of the coverage becomes more factual and the frequency of coverage decreases in the time period between the crimes and jury empanelment. Morales, 440 Mass. at 541. In support of his initial motion for a change of venue, the defendant submitted twelve newspaper articles that appeared either in print or on the Web site of The Recorder, a newspaper local to the Greenfield area.⁴ These articles identified the deaths of Pierce and Blanchard as homicides having occurred in the defendant’s apartment and contained references to the defendant’s criminal record, including his guilty plea in 2000 to a charge of assault and battery by means of a dangerous weapon for slashing the throat of a seventeen year old woman with a kitchen knife. The articles also referenced the defendant’s confession to police. Of the twelve articles, five were published in July, 2007, in the days following the discovery of the victim’s bodies and five more were published in the six months following, between August, 2007, and January, 2008. The final two were published in April, 2008. The defendant later supplemented his renewed motion for a change of venue with four additional articles published between June, 2009, and January, 2010, which reported on various pretrial hearings and reiterated the allegations against the defendant relating to the killings of Pierce and Blanchard and the defendant’s confession. We agree with the judge’s conclusion that these articles did not constitute pervasive publicity because they appeared in a small number of local news sources and the intensity of the reporting decreased over time with no articles appearing between January, 2010, and the time of the judge’s ruling in April, 2010. See Morales, supra.

Furthermore, the first requirement for presumptive prejudice is met only if the publicity is both extensive and sensational. Morales, 440 Mass. at 540. Commonwealth v. Colon-Cruz, 408 Mass. 533, 551 (1990). Publicity is sensational when it contains emotionally charged material that is gratuitous or inflammatory, rather than a factual recounting of the case. Toolan, 460 Mass. at 464 n.19. See United States v. Angiulo, 897 F.2d 1169, 1181 (1st Cir. 1990). Although references to the defendant’s past criminal convictions may not have cast him in the most favorable light, we have held that fact-based publicity, even that which contains references to charges pending against the defendant, a confession, or a prior criminal record, is not the sort of sensational publicity that would give rise to a presumption of prejudice. Morales, supra. See Angiulo, supra (reports of charges against defendant coupled with references to him as “reputed crime figure” and “mafia boss” were not so sensational as to contribute to presumption of prejudice). Therefore, we agree that the defendant failed to establish that the pretrial publicity surrounding his case was both extensive and sensational.

Additionally, the defendant failed to establish the second factor in the presumptive prejudice analysis because it was not practically impossible for the trial judge to empanel an impartial jury. See Toolan, 460 Mass. at 464. We have measured this practical impossibility by looking to the percentage of the venire that was dismissed for cause as a result of prejudice from exposure to pretrial publicity. Morales, 440 Mass. at 541. In cases where a high percentage of the venire is found to be prejudiced as a result of pretrial publicity, a shadow of doubt may be cast over the remaining venire members such that the prejudice of the remaining venire members may be presumed. Angiulo, 897 F.2d at 1181-1182. Although we have not identified a specific percentage of dismissals that will trigger the presumption, we have held that dismissal of as much as forty-two per cent of the venire is not sufficient to give rise to a presumption that the entire venire was tainted by pretrial publicity. Commonwealth v. Angiulo, 415 Mass. 502, 515 (1993). In this case, the trial judge estimated that less than twenty per cent of the venire was dismissed for cause as a result of exposure to pretrial publicity.⁵ Moreover, of the sixteen seated jurors, only two had been exposed to any pretrial publicity at all. After a thorough voir dire, the judge concluded each could put aside what he or she had read and remain impartial. A judge is entitled to accept a juror’s statements of disinterest. See James, 424 Mass. at Ill. The record reflects that it was possible for the trial judge to empanel a jury that appeared impartial. Thus, the defendant failed to meet the requirements for presumptive prejudice.

b. Actual prejudice. The defendant likewise failed to establish actual prejudice that deprived him of his right to an impartial jury. To demonstrate actual prejudice, a defendant must show that, in the totality of the circumstances, pretrial publicity deprived the him of his right to a fair and impartial jury. Morales, 440 Mass. at 542. In a case that has been the subject of pretrial publicity, the voir dire procedures utilized by the judge are particularly important. Toolan, 460 Mass. at 466-467. There is no indication in the record before us that the jurors seated in this case were prejudiced by pretrial publicity. As discussed supra, only two seated jurors had been exposed to any pretrial publicity at all, and both were subjected to a thorough individual voir dire. Moreover, the judge took ample steps to guard against any prejudice stemming from the media’s coverage of this case leading up to the trial. See Morales, 440 Mass. at 542-543; Colon-Cruz, 408 Mass. at 552. In denying the defendant’s original motion for a change of venue, the judge stated that she would be willing to reconsider a change of venue request should the empanelment of impartial jurors prove impossible, and she conducted thorough individual voir dire of each potential juror. She permitted counsel to suggest additional questions of individual jurors during both the prehminary voir dire of the whole venire and the individual voir dire of each venire member, and she granted the defense an additional preemptory challenge. Throughout jury selection and the trial, the judge admonished potential and seated jurors not to discuss the case with anyone under any circumstances and not to come into contact with any local news accounts related to the case. She also inquired each time potential or seated jurors returned to court whether anyone had come into contact with any information related to the case and noted their responses for the record. Thus, in light of the steps taken by the trial judge to safeguard the defendant’s right to an impartial jury, we conclude that the defendant failed to show actual prejudice necessitating a change of venue.

4. Third-party culprit evidence and jury instruction, a. Third-party culprit evidence. At trial, the defendant sought to admit in evidence a knife that was seen in the days following the killings in a cellar hole on Chapman Street, which is located approximately one-half mile to one mile from the site of the killings. The defendant also sought to question the medical examiner as to whether this knife was consistent with the wounds of the two victims. The judge refused to admit the knife in evidence on the ground that the testimony of three witnesses who observed the knife in the hole some days after the killings was insufficient to connect the knife to the crime scene or the victims’ injuries. On appeal, the defendant argues that the judge’s ruling deprived him of his right to present a defense that a third party committed the killings.⁶ We disagree.

A defendant’s right to create reasonable doubt through the presentation of so-called “third-party culprit” evidence is a “time-honored method” of defending against criminal charges. Commonwealth v. Rosa, 422 Mass. 18, 22 (1996). Because the right to raise a defense is a due process right of constitutional dimension, we review the judge’s ruling independently. Commonwealth v. Silva-Santiago, 453 Mass. 782, 804 n.26 (2009).

Trial judges are permitted broad discretion in determining whether to exclude evidence that a third party committed the crime. Commonwealth v. Bizanowicz, 459 Mass. 400, 418 (2011). In a case such as this where the proffered evidence is not hearsay, the evidence may only be admitted if it has a rational tendency to prove the issue the defense raises, and it is not “too remote or speculative.” Id., quoting Silva-Santiago, 453 Mass. at 801. Here, the judge was correct in concluding that admitting the knife in evidence likely would have required significant speculation on the part of the jury because the knife had not been shown to relate in any way to the killings or the location of the killings. The defendant had established that the knife was found in a hole on the same street on which a woman resided who may have harbored some jealousy toward Pierce. However, this evidence alone fails to establish a sufficiently close connection between the knife and the murders to require admission of the knife in evidence. See Commonwealth v. Fitzpatrick, 463 Mass. 581, 595-596 (2012).

Furthermore, the judge did not foreclose entirely the defendant’s ability to assert a third-party culprit defense. Although the judge refused to admit the knife in evidence and precluded the defendant from questioning the medical examiner about the knife or from arguing with regard to the knife in closing argument, the judge also refused the Commonwealth’s request to strike all testimony related to observation of the knife on Chapman Street in the days following the killings. The defendant also was able to present testimony regarding the man in the red hat to whom the defendant had sold counterfeit drugs, along with evidence that blood belonging to an individual other than the defendant or either of the victims was found in the apartment on the pillow covering the face of Blanchard and that investigators never determined the identity of this additional blood source. The defendant further emphasized these facts throughout his closing argument. Therefore, the judge did not err in refusing to admit the knife in evidence, and the defendant was not deprived of his constitutional right to present a defense.

b. Third-party culprit instruction. On appeal, the defendant also argues that because he was permitted to introduce certain third-party culprit evidence at trial, the judge erred in failing to provide the jury with a “third-party culprit instruction.” Specifically, the defendant requested that the judge instruct the jury that if they were to determine that a third party could have committed the murders, they could not find the defendant guilty unless they found beyond a reasonable doubt that the third party did not commit the crimes. The defendant also requested an instruction that the Commonwealth has the burden to prove that the defendant perpetrated the crime alone. The judge refused to give these instructions, stating that the instructions provided to the jury had conveyed adequately the Commonwealth’s burden in the context of the facts of the case. We agree that the instructions regarding the Commonwealth’s burden to prove each element of the crime, including the identity of the perpetrator, were constitutionally sufficient.

We consider the constitutional adequacy of jury instructions based on the “over-all impact” of the instructions as interpreted by a reasonable juror. Commonwealth v. Farley, 443 Mass. 740, 745 (2005), quoting Commonwealth v. Sellon, 380 Mass. 220, 231-232 (1980). Generally, a judge is not required to provide an instruction, or to use the specific wording offered by the defendant, so long as the charge, taken as a whole, adequately conveys the legal issue to the jury. Commonwealth v. Pires, 453 Mass. 66, 71 n.3 (2009). See, e.g., Commonwealth v. Walker, 460 Mass. 590, 614 (2011). We review jury instructions with regard to the Commonwealth’s burden of proof in a criminal case to determine whether the instructions, taken as a whole, make clear the Commonwealth’s burden to prove each element of the crime beyond a reasonable doubt. Farley, supra. We have never held that admission of third-party culprit evidence requires a judge to provide the jury with a third-party culprit instruction. See Commonwealth v. Gomes, 459 Mass. 194, 207-208 (2011); Farley, supra at 747. Rather, we have held that instructions in the third-party culprit context are not error where they adequately convey to the jury that the Commonwealth’s burden includes the obligation to prove beyond a reasonable doubt that the defendant committed the crime. Gomes, supra. Farley, supra at 747. We also have held that even when a defendant raises a third-party culprit defense, the Commonwealth does not have the burden to prove beyond a reasonable doubt that some third party is not guilty of the charged crime. Farley, supra at 745-746. Cf. Commonwealth v. Salemme, 395 Mass. 594, 601 (1985) (where two parties have equal opportunity to commit murder and where prosecution does not pursue joint venture theory, prosecution may need to present evidence ruling out one of two parties in order to prove guilt of other beyond reasonable doubt).

Here, the judge provided the jury with thorough and comprehensive instructions regarding the Commonwealth’s burden of proof. Furthermore, the lack of a specific third-party culprit instruction did not, as the defendant contends, negate the third-party culprit evidence presented by the defendant. A jury is presumed to follow the judge’s instructions. Commonwealth v. Degro, 432 Mass. 319, 328 (2000). Commonwealth v. Watkins, 425 Mass. 830, 840 (1997). Thus, if the evidence of the man in the red hat or the presence of an additional source of biological evidence at the crime scene had raised a reasonable doubt in the minds of the jury that the defendant committed the crime, the jury could not, according to the judge’s instructions, have convicted the defendant. See Farley, 443 Mass. at 747. Therefore, the absence of a third-party culprit instruction was not error.

5. Exclusion of expert testimony. At trial, the defendant proffered the testimony of Allison Redlich, Ph.D., a professor in the School of Criminal Justice at the University of Albany, State University of New York, who has conducted and published several studies regarding confessions and police interrogation techniques. Dr. Redlich also coauthored a 2010 research review that she described in voir dire as standing alone among peer-reviewed publications because it has been approved by the American Psychological Association as a whole. Known as a “White Paper,” this article indorses the scientific study of false confessions in the field of psychology. See Kassin, Drizin, Grisso, Gudjonsson, Leo, & Redlich, Police-Induced Confessions: Risk Factors and Recommendations, 34 J. Law & Hum. Behav. 3 (2010) (White Paper).

Dr. Redlich had not examined or treated the defendant, and she did not intend to opine either on the defendant’s mental state during the police interview or on the veracity of the statements the defendant had made to police. Rather, the defense proffered that Dr. Redlich’s testimony would be limited to the fact that false confessions do occur, that they are an area of scientific study, and that there are certain factors related to interrogation methods and the internal disposition of individual suspects that have been identified as commonly occurring among false confessions. The defendant argued that such testimony would be helpful to the jury in assessing the reliability of the defendant’s statements to police.⁷

The Commonwealth moved to exclude Dr. Redlich’s testimony on the ground that it did not meet the standard for the admission of expert testimony set forth in Commonwealth v. Lanigan, 419 Mass. 15, 26 (1994), because the research underlying the testimony lacked a sufficient empirical basis. After hearing argument on the Commonwealth’s motion, the judge decided to take testimony from the expert witness in voir dire to determine whether such testimony should be admitted for the jury’s consideration.⁸ The judge limited the subject of the voir dire to the witness’s proffered opinion regarding the concept of false confessions as an area of scientific research, the factors that may contribute to such confessions, and any connection this information would have to the defendant’s case and the jury’s assessment of the evidence.

At the hearing, Dr. Redlich testified that false confessions exist and have been the subject of hundreds, if not more than 1,000, research studies. She also described various methodologies that are used to conduct scientific research on false confessions. For example, proven⁹ false confessions have been analyzed to identify commonalities among them in order to identify risk factors that may give rise to a false confession. Additionally, laboratory experiments have been conducted to confirm that humans engage in certain counterintuitive behaviors such as admitting to wrongdoing that they have not in fact undertaken. Moreover, for the purposes of seeking to establish a potential incidence rate of false confessions, “self-report” surveys have been conducted among convicted offenders asking whether they have ever given a false confession and under what circumstances. Dr. Redlich explained that as a result of this research, a taxonomy of false confession “types” has been identified¹⁰ and certain factors have been linked to proven false confessions.¹¹ Based on her review of the defendant’s recorded statements, Dr. Redlich also opined on those false confession factors that were present in the defendant’s case.¹² Additionally, Dr. Redlich acknowledged that no studies have been conducted comparing the prevalence of these factors among false confessions to either “true” confessions or all confessions. Furthermore, Dr. Redlich testified that the studies based on proven false confessions that have been used to identify these relevant factors were based on a sample size of approximately 150 to 200 proven false confessions.

At the conclusion of the Lanigan hearing, the judge ruled that the expert testimony could not be admitted in evidence because the principles and methods on which Dr. Redlich’s opinions were based had not been shown to be sufficiently reliable to go before the jury. The judge emphasized that the research studies that identified the factors linked to false confessions were based on a limited sample size of proven false confessions and that no research or information would come before the jury regarding how frequently such factors also may be present in true confessions. The defendant argues on appeal that this ruling deprived him of his constitutional right to present a defense. We review rulings on the admission of expert testimony for an abuse of discretion, see Commonwealth v. Robinson, 449 Mass. 1, 5 (2007), and we affirm.

The proponent of scientifically based expert testimony must establish, among several foundational requirements, that the expert’s opinion will be based on principles and methods that are reliable. Commonwealth v. Barbosa, 457 Mass. 773, 783 (2010), cert. denied, 131 S. Ct. 2441 (2011) (summarizing five foundational requirements for admission of expert testimony). Lanigan, 419 Mass. at 26. Reliability may be established either by demonstrating that the principles and methods generally are accepted in the relevant scientific community or by applying the factors set forth in the United States Supreme Court’s decision in Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 592-594 (1993). Commonwealth v. Patterson, 445 Mass. 626, 640-641 (2005) (“Where general acceptance is not established . . . , a full Daubert analysis provides an alternate method of establishing reliability”). The judge is the gatekeeper of the evidence and is responsible for making the threshold determination that the expert opinion is sufficiently reliable to go before the jury. Commonwealth v. Shanley, 455 Mass. 752, 761-762 (2010). The judge’s gatekeeping function in the context of expert testimony applies in addition to the judge’s general duty to exclude evidence that is irrelevant or for which the probative value is substantially outweighed by the risk of unfair prejudice, confusion, or waste of time. See Commonwealth v. Bonds, 445 Mass. 821, 831 (2006); Patterson, supra at 639 n.10; Mass. G. Evid. §§ 401, 403 (2014).

On appeal, the defendant argues in part that the judge abused her discretion in excluding this testimony for lack of reliability because in other contexts we have permitted the admission of expert testimony based on social science research that similarly lacks the type of prevalence data at issue here. For example, in Shanley, 455 Mass. at 766, we upheld the admission of expert testimony regarding the phenomenon of dissociative amnesia, or repressed memories, of child sexual abuse. There we acknowledged that the research on which the expert’s testimony relied had been criticized as lacking “known error rates or standardization,” but we concluded that the trial judge did not abuse his discretion in admitting the testimony, which was based on “a wide collection of clinical observations and a survey of academic literature.” Id.

Particularly in the case of psychological or behavioral sciences, lack of prevalence data alone may not be sufficient to justify a ruling that the theory is unreliable within the meaning of Lanigan. See Shanley, 455 Mass. at 766. See also Mass. G. Evid. § 702 (2014). However, unlike in Shanley, where the expert testimony was based on the witness’s own clinical observations, his review of thousands of studies on human memory, and his review of eighty-five studies specific to the dissociative amnesia relevant in that case, here the expert estimated that she had read over 250 studies on false confessions but that only four of those studies were based on a sample of proven false confessions and that the total sample size of proven false confessions among these studies is at most 150 to 200 individuals.¹³ See Shanley, supra at 764.

Moreover, the judge’s determination as to the admissibility of this expert testimony also may have been influenced by her assessment of the probative value of such evidence in the context of the facts of the case that had been presented to the jury. For example, only some of the dispositional factors and almost none of the situational factors that have been identified in proven false confession cases were present in the defendant’s case. Specifically, the defendant likely was suffering some pain and discomfort stemming from his recent self-inflicted injuries, and his history of substance abuse, depression, and a severe brain injury may have made him more vulnerable or prone to suggestion, but the defendant was not a juvenile during the interrogation nor does the record indicate that he suffers from any intellectual disabilities. Additionally, there is no indication in the record of the police using a false evidence ploy or offers of leniency during the interrogation. Furthermore, although the interrogation extended throughout most of the day, the length did not implicate either of the two concerns underlying that factor, specifically social isolation or sleep deprivation. The defendant was permitted multiple breaks from the interrogation, thereby providing brief respite from the inherent stress of the process. He also was not isolated from all forms of social support in that he was permitted to speak with his attorney and his mother, and at the defendant’s request Detective Laster, whom the defendant saw as a friend, was permitted to participate in the interview. Additionally, the interview did not likely result in sleep deprivation as it did not last overnight, and there was no indication that the defendant had not slept the night before while he was in the hospital.

Furthermore, aside from proffering the expert testimony, the defendant’s only attempt to attack directly the substantive reliability of his statements to police occurred in closing argument. The defendant emphasized in closing that he was ill and in pain during the interrogation, and he made efforts to cast doubt on the voluntariness of his statements. However, the defendant did not, for example, present evidence of the subsequent recantation of his statements to police, or testify that he fabricated his statements to police as a result of police coercion or other motives. See, e.g., People v. Page, 2 Cal. App. 4th 161, 177-179, 185-186 (1991) (admission of expert testimony related to factors that may influence defendants to make false confessions was proper where defendant testified on his own behalf and asserted that confession was product of false evidence ploys and other suggestions by interrogators). Nor did the defendant present the testimony of a treating physician who could opine that the defendant was suffering from a psychiatric disorder during the interrogation that would cause him to make false statements. See, e.g., United States v. Shay, 57 F.3d 126, 133-134 (1st Cir. 1995) (error to exclude proffered expert testimony from psychiatrist that defendant suffered from recognized mental disorder that caused him to make false statements inconsistent with his own self-interest).

Thus, in light of the limited number of false confession factors present in this case, combined with the lack of evidence before the jury calling into question the veracity of the defendant’s statements, the judge may have concluded that the proffered expert testimony was not relevant and would have distracted or confused the jury by giving rise to speculation based on facts and assumptions not in evidence. See Commonwealth v. Jones, 464 Mass. 16, 19-20 (2012); Commonwealth v. Beausoleil, 397 Mass. 206, 217 (1986). Therefore, it was not an abuse of discretion to exclude the expert testimony where the judge could have concluded also that the testimony’s probative value was outweighed by its potential to confuse or distract the jury.

We do acknowledge, however, that the phenomenon of false confessions is a growing area of psychological and social science research, and we are mindful that false confessions have been demonstrated to occur even in the context of serious crimes, including murder. See White Paper, supra at 5. Therefore, we do not foreclose the possibility that under appropriate circumstances this sort of expert testimony could be relevant to a defendant’s case and helpful to a jury. See Robinson, 449 Mass. at 6-7. We also acknowledge that the state of the research on false confession has progressed significantly since we ruled in Robinson that such expert testimony properly was excluded as unreliable. See id. at 2 (trial at which expert testimony was proffered took place in 1997). See generally White Paper, supra, and sources cited. Although the precise number of false confessions or prevalence data regarding false confession factors may never be identified, as research regarding factors present in proven false confessions continues to progress, information regarding these factors may be helpful in certain circumstances. For example, should a case arise in which a defendant attacks directly the veracity of his or her statements to police and where several of the false confession factors thus far identified are present, it may be appropriate for expert testimony of the sort proffered here to be taken into consideration.¹⁴ Therefore, we leave further development of this issue for another day.

6. Review under G. L. c. 278, § 33E. We have reviewed the entire record and discern no basis to grant relief under G. L. c. 278, § 33E.

Judgments affirmed.

1 We agree that the defendant was not in custody at the hospital when he requested that Detective Laster come to see him. However, the defendant may have been in custody for the entire time at the police station because although he was given the choice whether to accompany police to the station, a reasonable person in the defendant’s position, knowing he had admitted to two killings the day before, and being met on discharge from the hospital by three police officers, may have felt he had no choice but to go with them. Additionally, the defendant was driven to the station by police, and while there, although permitted breaks on request, he was escorted throughout the station. Furthermore, the initial series of interviews terminated with his arrest. See Commonwealth v. Sneed, 440 Mass. 216, 220 (2003) (whether defendant is in custody is objective inquiry based on defendant’s reasonable belief). However, because we agree with the judge’s determination that each of the defendant’s Miranda waivers was voluntary, as were his statements to police, even if he was in custody for the whole of his time at the police station, there was no error in denying the defendant’s motion to suppress.

2 The attorney and Gibbons gave conflicting accounts of this conversation at the motion to suppress hearing. The judge credited Gibbons’s recollection of the conversation because it was more specific and concluded that although the attorney’s intent may have been to inform the police that the defendant had invoked his right to remain silent, what the detective reasonably understood the attorney to be communicating was that he had advised the defendant not to answer any more questions.

3 The record indicates that the defendant’s anxiety medication was not administered precisely as prescribed. However, we defer to the judge’s determination based on evidence presented at the motion to suppress hearing that any improper dosing was the result of a discrepancy between the prescription issued at the hospital and the dosage dispensed by the pharmacy. We also agree with the judge’s finding that the defendant was permitted to take the prescribed medications that were picked up from the pharmacy by his mother, and that the police did not withhold his medication. Furthermore, the defendant presented no expert testimony at the motion to suppress hearing regarding the effects of this medication in either proper or improper doses. Therefore, we agree that in the totality of the circumstances any inaccuracies in the administration of the defendant’s medication were not sufficient to render his waivers or his statements involuntary.

4 According to the findings of the judge on the motion for a change of venue, articles published on the Web site of The Recorder are available only by subscription.

5 Although the defendant argues on appeal that other jurors should have been dismissed for cause as a result of their exposure to pretrial publicity, mere exposure to pretrial publicity or knowledge of the case is not sufficient to require the judge to dismiss a juror for cause. Morales, 440 Mass. at 542. Colon-Cruz, 408 Mass. at 551. Thus, the fact that less than twenty per cent of the total venire was dismissed for cause is an acceptable basis on which to assess this second factor.

6 We note that during trial, defense counsel argued that the knife was relevant only to show that the police investigation was inadequate, thus pursuing a so-called “Bowden defense.” See Commonwealth v. Fitzpatrick, 463 Mass. 581, 596-597 (2012), citing Commonwealth v. Bowden, 379 Mass. 472, 485-486 (1980). Although the same evidence often may be used to support a third-party culprit defense and a Bowden defense, these two defenses are “logically (and legally) distinct.” Commonwealth v. Silva-Santiago, 453 Mass. 782, 800 (2009). Therefore, we will treat any error with respect to the judge’s failure to admit the knife found on Chapman Street as unpreserved insofar as the defendant argues for the first time on appeal that the knife was proffered to support a third-party culprit defense. Furthermore, although the defendant does not argue on appeal that the judge’s decision to exclude the knife from evidence was error under Bowden, we consider that argument pursuant to our power under G. L. c. 278, § 33E, and we conclude that the judge’s exclusion of the knife did not foreclose improperly the defendant’s right to assert a defense that the police investigation was inadequate. Contrast Bowden, supra at 486. As the judge concluded, the second knife had not been connected to the victims or the crime scene in any way and presented too great a risk of prejudicing the Commonwealth by diverting the jury’s attention to collateral matters. See Commonwealth v. Bizanowicz, 459 Mass. 400, 414 (2011). Furthermore, the defendant was permitted to elicit testimony and to cross-examine Commonwealth witnesses for the purposes of creating the inference that the police investigation was inadequate and that the Commonwealth’s evidence was therefore unreliable and incomplete. See Silva-Santiago, supra at 801. Additionally, a significant portion of the defendant’s closing argument emphasized areas of the investigation that the police did not pursue.

7 Defense counsel acknowledged that this expert testimony may be relevant also to the jury’s assessment of the voluntariness of the defendant’s statements to police. However, based on false confession research, circumstances may exist where a false confession is nevertheless voluntary. See Kassin, Drizin, Grisso, Gudjonsson, Leo, & Redlich, Police-Induced Confessions: Risk Factors and Recommendations, 34 J. Law & Hum. Behav. 3, 14 (2010) (White Paper). Thus, a constitutional challenge to the admission of a defendant’s statements based on voluntariness is distinct from a defense based on the theory that the statements are substantively false and should not be credited by the jury.

8 It was wise for the judge to conduct such a hearing in this case; we have approved of this practice in similar circumstances where the proffered testimony relates to the “evolving nature of scientific and clinical studies of the brain and memory and the controversy surrounding those studies.” Commonwealth v. Shanley, 455 Mass. 752, 763 n.15 (2010).

9 Confessions are classified as “proven” false when it is later determined that either (1) no crime has been committed, such as when a death is determined to have in fact been the result of natural causes; (2) it was physically impossible that the defendant was present during the crime, such as when the defendant is discovered to have been incarcerated at the time of the crime; (3) scientific testing excludes definitively the convicted offender through a determination that critical biological evidence, such as semen in a rape case, does not belong to the defendant; or (4) the true perpetrator is identified and linked to the crime through scientific evidence or by the revelation of critical details of the crime that could only have been known by the true perpetrator. See White Paper, supra at 5.

10 False confessions have been classified into three categories: “voluntary,” “compliant,” and “internalized.” White Paper, supra at 14-15. A voluntary false confession occurs without prompting from authorities, such as when an individual seeks to gain notoriety by confessing to a high-profile crime or to deflect attention from a loved one who is the true culprit. Id. at 14. Compliant false confessions occur when a suspect knows he or she did not perpetrate the crime but acquiesces to pressure imposed in the interrogation context and comes to believe that the short-term benefit of a confession (e.g., escaping the interrogation, obtaining a reduced sentence, or receiving some other implied reward) outweighs the long-term risk of conviction. Id. Finally, “internalized” false confessions arise where a suspect is prone to suggestion or otherwise vulnerable and comes to accept that he or she in fact perpetrated the crime, often because the suspect has developed a profound distrust of his or her own memories. Id. at 15.

11 These factors have been grouped into two categories: “situational” risk factors and “dispositional” risk factors. White Paper, supra at 16, 19. Situational factors may include the length of the interrogation, including whether it gives rise to sleep deprivation for the suspect and whether it results in complete isolation from sources of support. Id. at 16. Situational factors also include certain interrogation techniques such as confronting a suspect with false evidence of involvement in the crime (the “false evidence ploy”) and the “minimization” tactic, which has been shown to imply promises of leniency to the suspect. Id. at 16-17, 18. Dispositional factors are those vulnerabilities of a suspect including age (specifically whether the suspect is a juvenile), as well as mental impairment, such as in the form of intellectual disability or mental illness. Id. at 19-21. In the case of mental illness, psychological disorders have been linked to “faulty reality monitoring, distorted perception, impaired judgment, anxiety, mood disturbance, poor self-control, and feelings of guilt,” which could contribute to the making of a false confession. Id. at 21.

12 Dr. Redlich indicated that substance use and prior brain injuries are the sort of internal factors that could render a suspect more vulnerable in an interrogation context. She also identified as relevant the fact that the defendant had just been released from the psychiatric hospital following a suicide attempt and that he was in pain from his self-inflicted injuries, and she noted that depression has been linked to false confessions in that it gives rise to fatigue, hopelessness, and a foreshortened sense of the future. She also explained that the defendant’s statement to police could be termed “non-generative” in that it did not lead the police to identify any additional evidence corroborative of the defendant’s guilt, such as the location of the murder weapon.

13 The White Paper, supra at 5, appears to confirm this estimate, describing a study of 125 cases of proven false confessions as “the largest sample ever studied.”

14 See, e.g., Leo, Drizin, Neufeld, Hall, & Vatner, Bringing Reliability Back In: False Confessions and Legal Safeguards in the Twenty-First Century, 2006 Wis. L. Rev. 479, 522, 531-532 (proposing threshold assessment by judge of substantive reliability of confessions prior to admission in evidence to ensure probative value outweighs potential unfair prejudice).