¶ 24. dissenting. E.W. bore the burden of demonstrating that he was in custody at the time that he made incriminating statements, and he failed to make this showing. The facts as found demonstrate that the interview took place at E.W.’s home; the interview was conducted by one officer, who did not communicate a belief in E.W.’s guilt or confront E.W. with evidence of the crime; E.W. was free to move around; the interview was terminated at several points to allow him to consult with his foster father; and no deceptive interview techniques were used. Under these circumstances, a reasonable fifteen-year-old in E.W.’s situation would have felt free to leave and terminate the interview. Therefore, I would affirm the court’s decision denying E.W.’s motion to suppress.
*323¶ 25. The pivotal question is whether E.W. was in custody at the time he made incriminating statements because police are not required to give Miranda warnings when suspects are not in custody. See Miranda v. Arizona, 384 U.S. 436, 444-45 (1966); State v. Garbutt, 173 Vt. 277, 282, 790 A.2d 444, 448 (2001) (“Suspects not in custody are not entitled to Miranda warnings.”).
¶ 26. As the majority states, the determination of custody requires an objective inquiry into the totality of the circumstances to ascertain whether a reasonable person in the defendant’s circumstances would feel free to refuse to answer and leave. State v. Pontbriand, 2005 VT 20, ¶ 11, 178 Vt. 120, 878 A.2d 227. The most important factor is whether police informed the defendant of the right to leave, but other factors are relevant, including the location of the interview, the extent the officer communicates his belief in the defendant’s guilt, how the interview was initiated, whether a reasonable person would feel free to leave, and whether the defendant is confronted with evidence of his guilt. State v. Sullivan, 2013 VT 71, ¶ 29, 194 Vt. 361, 80 A.3d 67 (citing State v. Muntean, 2010 VT 88, ¶ 19, 189 Vt. 50, 12 A.3d 518). The inquiry is meant to focus on whether the situation approximates an “ ‘incommunicado interrogation of individuals in a police-dominated atmosphere.’ ” State v. Willis, 145 Vt. 459, 475, 494 A.2d 108, 117 (1985) (quoting Miranda, 384 U.S. at 445).
¶27. In assessing these factors, the majority concludes that E.W. was in custody when he spoke with the officer primarily because: E.W. was not told that he was free to end the interview and leave; the officer confronted E.W. with suspicions and evidence of guilt; E.W.’s age made him more likely to believe he was not free to leave; and the setting of the interview in E.W.’s foster house and E.W.’s status being in the custody of the Department for Children and Families (DCF) made E.W. more likely to feel compelled to submit to police authority.
¶ 28. Apart from its conclusions, the majority’s analysis is flawed in two significant ways. First, we are required to accept the facts found by the trial court unless we conclude that its findings are clearly erroneous. State v. Oney, 2009 VT 116, ¶ 11, 187 Vt. 56, 989 A.2d 995. The majority claims to give the trial court’s findings deference, but nonetheless makes inferences from the record contrary to those findings. Second, although the majority recognizes that it is the defendant’s burden to prove “that he was ‘in custody’ and, therefore, entitled to Miranda warnings,” the ma*324jority makes inferences in E.W.’s favor that are not supported by record evidence. State v. LeClaire, 2003 VT 4, ¶ 15, 175 Vt. 52, 819 A.2d 719; accord Pontbriand, 2005 VT 20, ¶ 10.9 In fact, those inferences must be drawn against E.W., who had the burden of *325production as well as persuasion. With the exception of the majority’s first consideration — that E.W. was not explicitly told he could terminate the interview or leave — the rest of the majority’s analysis rests on facts not found by the trial court or found directly to the contrary, or inferences not supported by the evidence.
¶29. As to what the officer communicated to E.W. regarding E.W.’s guilt and whether the officer confronted E.W. with evidence of his guilt, the majority’s findings conflict with those of the trial court. The majority acknowledges that the trial court found there was “no evidence” that the officer communicated his subjective belief that E.W. was guilty of theft or that he confronted E.W. with evidence of his guilt. Yet, the majority weighs this factor in favor of custody, inferring from E.W.’s testimony that the officer asked about the location of the car that police did communicate a belief in E.W.’s guilt. In reviewing the trial court’s denial of a motion to suppress, we are required to follow the trial court’s findings. Further, where the burden rests on defendant, this Court cannot draw inferences in his favor about what was communicated when there is no direct evidence in support.
¶ 30. In weighing the totality of the circumstances, the content of the conversation between police and a suspect is important to show whether the “questioning created the kind of coercive environment indicative of police custody.” State v. Hieu Tran, 2012 VT 104, ¶ 16, 193 Vt. 148, 71 A.3d 1201 (quotation omitted). The conversation in this case bears no indication of coercion or police domination. The facts as found by the trial court and supported by the evidence show that the officer neither communicated his belief in E.W.’s guilt to E.W. nor confronted E.W. with evidence of his guilt. Further, there was no evidence that the questioning itself was coercive. Only one police officer was present, and the interview lasted at most an hour. There was no evidence that the police officer used any deceptive interview methods.10 The police officer did not dominate or control the atmosphere of the interrogation. To the contrary, E.W. witnessed the officer speak to his GAL and wait for the foster father to speak to the GAL twice before the officer spoke to E.W. at all. The police officer allowed the foster father not only to be present but to participate, and to *326speak privately with E.W. before and during the interview. Therefore, this factor weighs against a finding of custody.
¶ 31. Similarly, the physical setting does not support a showing that a custodial situation existed. This was a casual encounter at E.W.’s home in a location unlike the police-dominated atmosphere typical of a custodial situation, such as a police station or a police cruiser. See Sullivan, 2013 VT 71, ¶ 30 (considering fact that interview took place in defendant’s home as supporting conclusion that defendant was not in custody); In re D.A.C., 741 S.E.2d 378, 382 (N.C. Ct. App. 2013) (concluding juvenile not in custody based in part on facts that he was questioned outside his home in open area). The interview began in the foyer of the residence and moved outside to a less restrictive environment. During the questioning, E.W.’s freedom of movement was not restrained in any way. Rather than isolating E.W. in a room and insisting that he stay within sight, the officer allowed the interview to roam from house to porch to vegetable stand as circumstances dictated — from an enclosed space to a relatively unenclosed space. Additionally, E.W. was not taken into custody at the end of the interview.
¶ 32. The majority does not find these facts significant, holding that because E.W. was in DCF custody and living in a foster home, he did not draw comfort from the fact that the interview took place in a familiar setting or that he had family members close by. But this conclusion is based on conjecture and not record evidence. Certainly, E.W.’s status as a juvenile in DCF custody does not equate to being in police custody for purposes of whether Miranda applies. These are two very different concepts although the descriptive term used is the same. If defendant had been living with his natural parents, he would have been in their custody in the same way that he was in DCF custody. While DCF custody is a matter of who has control over care for the child, the custody inquiry under Miranda focuses on whether the suspect was “ ‘deprived of his freedom by the authorities in any significant way.’ ” Pontbriand, 2005 VT 20, ¶ 14 (quoting Miranda, 384 U.S. at 478). The majority recognizes that the concepts are different and yet weighs E.W.’s status as indicative of custody. Further, there was no particular evidence to suggest that E.W. was not comfortable in his foster home or with his foster family and so the location of the interview was less informal than for other suspects interviewed in their homes. When the facts are viewed objectively, *327the physical'setting of the interview leads to only one conclusion — that E.W. was not in custody during questioning.
¶ 33. The majority also weighs E.W.’s age as a factor indicating custody even though there is no evidence or findings to demonstrate whether the officer knew E.W.’s specific age, or whether E.W.’s age made him more likely to feel pressured to submit to police questioning and not feel at liberty to leave. Certainly, a suspect’s age is relevant to whether a situation was custodial. As the U.S. Supreme Court has held, “so long as the child’s age was known to the officer at the time of police questioning, or would have been objectively apparent to a reasonable officer, its inclusion in the custody analysis is consistent with the objective nature of that test.” J.D.B. v. North Carolina, 564 U.S. 261, 277, 131 S. Ct. 2394, 2406 (2011). The juvenile’s age is not dispositive, however, and must be weighed with all other relevant factors to determine if the situation was custodial. Id. at 277, 131 S. Ct. at 2406 (“This is not to say that a child’s age will be a determinative, or even significant, factor in every case.”).
¶ 34. To find it significant, the majority basically equates E.W. with the thirteen-year-old in J.D.B., and concludes that E.W. would not have felt free to withstand police questioning. There are several problems with this conclusion. First, the U.S. Supreme Court did not state that the thirteen-year-old in J.D.B. was automatically in custody by virtue of his age. The Court merely stated that the factor had to be considered along with others and that common sense should be used to determine how a juvenile’s age would affect his objective perception of the situation. Id. at 279-80, 131 S. Ct. at 2407 (explaining that police can use common sense to judge gradations of abilities of differently aged children). Second, the record demonstrates that the officer knew E.W. was in foster care, but no evidence was submitted to show that the officer knew E.W. was fifteen or that E.W.’s age was objectively apparent to the officer.11 Third, even assuming that the officer could determine that E.W. was fifteen, the record does not *328demonstrate that E.W.’s age made him more likely to perceive that he was in a custodial situation.
¶ 35. Common sense demonstrates that a fifteen-year-old juvenile is more responsible than a younger child and able to understand whether a situation has created a “restraint on freedom of movement of the degree associated with a formal arrest.” Sullivan, 2013 VT 71, ¶ 29 (quotation omitted) (explaining that custody is created when suspect is arrested or placed under conditions approximating arrest). Vermont law recognizes this fact, treating juveniles differently depending upon the nature of the alleged crime and the age of the juvenile. A juvenile of fourteen or fifteen years of age is ordinarily treated as an adult if charged with one of twelve major, violent crimes. See 33 V.S.A. §§ 5102(2) (C)(ii), 5204(a). A juvenile of sixteen or seventeen years of age may be treated as an adult for any crime. Id. § 5204(a). Here, given the other circumstances — that E.W. was questioned by one police officer in a familiar setting and with family members close by — there is nothing to indicate that a reasonable fifteen-year-old in E.W.’s circumstances would perceive that he was under arrest and not free to leave.12 See In re N.J., No. COA13-53, 2013 WL 5460091, at *7 (N.C. Ct. App. Oct. 1, 2013) (unpub.) (concluding that reasonable fifteen-year-old in juvenile’s situation would not have believed he was under arrest and therefore situation was not custodial).
¶ 36. The majority makes an additional error by stating that “the presence or absence of an independent adult may have some bearing on whether E.W. was in custody.” Ante, ¶ 13. In In re E.T.C., 141 Vt. 375, 449 A.2d 937 (1982), this Court held that a juvenile cannot voluntarily and intelligently waive his rights against self-incrimination and to counsel without first consulting with an impartial adult, who is genuinely interested in the juvenile’s welfare, independent from the prosecution, and aware of the juvenile’s rights. Id. at 379, 449 A.2d at 940. Our cases have made it clear, however, that the ability to consult with an interested adult becomes relevant only if the situation is custodial *329to then determine whether a juvenile has voluntarily and intelligently waived his rights. State v. Piper, 143 Vt. 468, 473, 468 A.2d 554, 557 (1983). There is no right to consultation for noncustodial police questioning because this could “cause unwarranted and prejudicial delay in investigatory situations when time is often of the essence.” Id. In this case, the majority improperly considers the presence of an independent adult in its analysis of whether custody existed.
¶ 37. To the extent that the presence and involvement of E.W.’s foster father was relevant at all, it was relevant to the physical setting of the interview, and when considered as part of that factor weighs against a finding of custody. See Muntean, 2010 VT 88, ¶ 23 (explaining that physical setting of interview important to custody determination because it influences reasonable person’s feeling on whether he or she is at liberty to leave). The trial court found the foster father was “genuinely interested” in E.W.’s welfare and was an active participant in the questioning. He consulted with E.W. prior to the interview and interrupted the interview when it appeared E.W. might incriminate himself to speak privately with E.W. on several occasions. In addition, although the foster father “spoke about honesty” in private with E.W., he “did not instruct [E.W.] that he had to speak to the trooper, nor did he tell [E.W.] that ‘[E.W.] should do the right thing.’ ” This discussion with E.W. did not create a coercive police environment. See In re D.A.C., 741 S.E.2d at 383 (concluding that fact that juvenile’s parents told him to speak honestly did not create custodial interrogation because it did not create government coercion). These facts, if anything, would demonstrate to a reasonable juvenile in E.W.’s situation that he was free to terminate the interview at any time.
¶ 38. The majority discounts the importance of the presence of the foster father by relying on facts contrary to the court’s findings. The trial court found, based on the evidence, that the foster father did not tell defendant to do the right thing in the interview. Despite this finding, the majority emphasizes testimony to the effect that the foster father discussed honesty with E.W. and the difficulty of doing the right thing. Ante, ¶ 21.
¶ 39. It may be that if E.W. had put on a thorough and complete case to meet his burden, the record would support as a matter of law that he was in custody and should have been provided Miranda warnings such that his confession would be *330suppressed. As the majority admits, “very little testimony was adduced about [the interview’s] specific content or progression.” Ante, ¶ 6. The foster father was a witness and was present throughout the interview. Defense counsel could have elicited the missing testimony. On the record before us, however, E.W. failed to demonstrate that he was in custody when he made the statements he now seeks to suppress. E.W. failed to show that he was subjected to an environment that was coercive, police-dominated, or approximating an arrest. To the contrary, the facts demonstrate that E.W. was questioned in a familiar setting by one officer, who did not use coercive interview techniques and neither confronted E.W. with evidence of guilt nor expressed a belief in E.W.’s guilt, E.W. was free to move around and consult with others, and E.W. was not taken into custody. Given these facts, the court .properly denied the motion to suppress, and I would affirm.
The issue of which party bears the burden of proving a defendant was in custody and therefore entitled to a Miranda warning has not been squarely decided by the U.S. Supreme Court. Federal courts generally hold that the defendant bears the burden of proving he was in custody at the time the statements were made. See, e.g., United, States v. Jorgensen, 871 F.2d 725, 729 (8th Cir. 1989) (affirming court’s denial of defendant’s motion to suppress because defendant “failed to demonstrate that he was subjected to custodial interrogation”); United States v. Lawrence, Nos. 88-2056, 88-2086, 88-2087, 88-2109 and 88-2135, 1989 WL 153161, at *5 (6th Cir. Dec. 18, 1989) (per curiam) (stating that defendant has burden of demonstrating that he was entitled to receive Miranda warnings because he was in custody); United States v. Charles, 738 F.2d 686, 692 (5th Cir. 1984) (holding that defendant bears burden of demonstrating statements were obtained during custodial interrogation), overruled on other grounds by United States v. Bengivenga, 845 F.2d 593 (5th Cir. 1988) (en banc); United States v. de la Fuente, 548 F.2d 528, 533-34 (5th Cir. 1977) (explaining that defendant bears burden of production and persuasion in suppression hearing in absence of well-defined exceptions); United States v. Artis, No. 5:10-cr-15-01, 2010 WL 3767723, at *4 (D. Vt. Sept. 16, 2010) (“The weight of authority appears to hold that a defendant bears the burden of establishing that he or she was subjected to custodial interrogation in order to establish a constitutional violation as the basis for suppression of evidence.”). But see United States v. Charbonneau, 979 F. Supp. 1177, 1181 (S.D. Ohio 1997) (stating government has burden of proving that Miranda warnings were not necessary).
In addition to this Court, other state courts also agree that the burden rests on the defendant. See State v. Vitale, 497 A.2d 956, 963 (Conn. 1985) (holding defendant has initial burden of showing conversation was custodial interrogation); State v. James, 225 P.3d 1169, 1172 (Idaho 2010) (‘We join the vast majority of courts that have considered the issue and hold that the burden of showing custody rests on the defendant seeking to exclude evidence based on a failure to administer Miranda warnings.”); Smith v. State, 974 A.2d 991, 1003 (Md. Ct. Spec. App. 2009) (explaining that defendant bears burden of proving Miranda is applicable); Wilkerson v. State, 173 S.W.3d 521, 532 (Tex. Crim. App. 2005) (explaining that until defendant demonstrates that statement was product of custodial interrogation, State has no burden to show compliance with Miranda).
To be clear, once custody is established, the State bears the burden of demonstrating that a suspect received notice of his Miranda rights and waived them voluntarily. See Colorado v. Connelly, 479 U.S. 157, 168-69 (1986) (holding that State bears burden of proving waiver of Miranda rights by preponderance of evidence); Smith, 974 A.2d at 1003 (explaining that once defendant demonstrates that Miranda is applicable burden is on State to show that its requirements were satisfied); State v. Caron, 155 Vt. 492, 506, 586 A.2d 1127, 1135 (1990) (“The State bears a heavy burden of proving a knowing and intelligent waiver of Miranda rights . . . .”).
The record does not show whether the police officer was carrying a weapon when he interviewed E.W.
The majority explains that even though the officer did not know E.W.’s precise age, this is immaterial because the officer was aware that E.W. was a minor in foster care. This fact alone, however, simply means that E.W. was less than eighteen years old. 33 V.S.A. § 5103(c) (explaining that jurisdiction of family court over child in delinquency or child-neglect proceeding generally extends to child’s eighteenth birthday).
This does not amount, as the majority claims, to a conclusion that E.W.’s age was not “relevant to his perception as to whether he was free to terminate the officer’s questioning.” Ante, ¶ 19. Certainly, E.W.’s age is one relevant factor to consider. The question is whether it is a significant factor. Here, there is simply nothing in the record to indicate that E.W.’s age was known to the officer or that E.W.’s age was particularly significant in this case.