OPINION
FIDEL, Judge.
Appellant William Edward Lundstrom was indicted on one count of first-degree
murder for the killing of his wife. He defended by asserting temporary insanity. He was tried before a jury and found guilty of manslaughter. His motion for new trial or judgment of acquittal was denied, and the court sentenced him to the presumptive term of 7.5 years of imprisonment with appropriate credit for pre-sen-tence incarceration.
Lundstrom presents three questions on appeal: (1) Did the trial court erroneously admit his self-incriminatory statements? (2) Did the trial court erroneously restrict his efforts to establish that the opinion of his medical witness on the issue of insanity was corroborated by the opinion of another doctor not called to testify at trial? (3) Was he prejudiced by an improper definition of clear and convincing evidence in a jury instruction concerning the defendant’s burden of proof to establish his defense of temporary insanity? Our answer to each question is no.
FACTS
Defendant and Tonya Lundstrom, the victim, were married for twelve years and had three children. Six months before her death, Tonya began seeing and became pregnant by another man. Lundstrom learned of their liaison and his wife’s pregnancy. On the morning of the shooting, Lundstrom, a Wickenburg resident, drove to Phoenix, where he bought a .38 caliber handgun. As he returned to Wickenburg, Lundstrom stopped to practice shooting. At 2:00 p.m. he went to the Frontier Inn in Wickenburg, the restaurant where Tonya worked; his wife was too busy to speak with him. He returned at 3:30 p.m.; his wife accompanied him to a back room to talk. After a brief discussion, Lundstrom produced the handgun and shot her five times. As Lundstrom waited outside the restaurant, Tonya died.
Lundstrom did not flee; he was arrested at approximately 3:40 p.m. Before he was arrested or read his rights, he volunteered to the police that he had shot his wife. He was held in custody until 7:31 p.m., when he was interviewed by Detective Ruben Madrid and Officer Verlin J. Modesitt. Lundstrom claims to have asked several police officers about Tonya’s condition during the approximately three hours of custody before this interview, but he was not told of her death until the interview began.
The interview lasted sixty-four minutes and included a twenty-seven minute break. At its start the police carefully informed Lundstrom of his rights. Lundstrom indicated that he understood them. However, when asked if he wished to waive his rights and proceed with the interview, he replied, “Can’t talk now.” Lundstrom began sobbing. The police provided tissue. They asked if he wanted something to eat or drink, and he declined. Then the following exchange occurred:
Officer Modesitt: Bill, Bill, you all done with this? ... Do you understand your rights that were read to you? Defendant: Yes.
Modesitt: You do understand, will you answer?
Defendant: Yes.
Modesitt: Now, do you want to talk to us now? Would you like to wait for a few minutes? Do you want to tell us something and get it off your chest?
Defendant: Let’s just get it over with.
Modesitt: Get it over with? Okay. Why don’t you, in your own words, tell us what happened.
Lundstrom proceeded to admit: (1) that he had gone to Phoenix to purchase a gun
in order to shoot Tonya, (2) that he went to the Frontier Inn for the purpose of shooting her, and (3) that he intended to shoot both his wife and her unborn baby. The rest of the interview concerned the defendant’s and victim’s marital disputes, defendant’s activities on the day of the shooting, and his recollection of the shooting and subsequent arrest. Interrogation concluded at 8:35 p.m.
VOLUNTARY SELF-INCRIMINATION
Lundstrom contends that the trial court erroneously admitted involuntary statements in violation of
Miranda v. Arizona,
384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966). He argues that his interrogators violated his
Miranda
rights by questioning him after he informed them that he could not speak and that this violation, exacerbated by his emotional state and his isolation by the police, rendered his statements involuntary and inadmissible.
Lundstrom was appropriately advised of his fifth amendment rights. The preliminary question is whether he then asserted them. Pursuant to
Miranda,
once a suspect has asserted his right to remain silent, the police must desist from questioning. The police in this case conducted an explicit interrogation and did so after Lundstrom’s initial declaration, “Can’t talk now.” But they waited an interval after this statement, offered Lundstrom additional time to collect himself, and only proceeded after Lundstrom’s indication that he understood his rights and wished the interview to proceed. Lundstrom at no time explicitly or implicitly indicated a “desire to deal with the police only through counsel.”
Edwards v. Arizona,
451 U.S. 477, 484, 101 S.Ct. 1880, 1885, 68 L.Ed.2d 378, 386 (1981). There is a substantial difference between a defendant saying, “I can’t talk,” because he is momentarily too upset to do so and a defendant saying, “I won’t talk,” because he wants to consult a lawyer or wishes to avoid the risk of self-incrimination.
Cf. State v. Finehout,
136 Ariz. 226, 665 P.2d 570 (1983) (the court found a
Miranda
violation where police continued to interrogate after defendant stated, “I ain’t going to say any more.”) The record supports the conclusion that this defendant was in the first category, not the second, and that he waived his rights to remain silent and to consult counsel before submitting to interrogation.
We next consider whether, despite Lund-strom’s expressed willingness to be interviewed, the circumstances of his arrest and isolation created a coercive interrogation environment that rendered his statements inadmissible. The state had the burden of proving by a preponderance of the evidence that Lundstrom’s confession was voluntary.
State v. Knapp,
114 Ariz. 531, 538, 562 P.2d 704, 711 (1977). The United States Supreme Court has recently stated that confessions will only be excluded as involuntary when there is police misconduct causally related to the confession.
Colorado v. Connelly,
479 U.S. 157, 107 S.Ct. 515, 93 L.Ed.2d 473 (1986). Although personal circumstances (e.g., age, intelligence, mental or emotional status, physical condition) can augment the impression of involuntariness, they do not require suppression of a confession absent “the crucial element of police overreaching.”
Id.
at 163, 107 S.Ct. at 520, 93 L.Ed.2d at 482.
The trial court found no overreaching in this case. In its order denying the defendant’s motion to suppress, the court concluded “that there was no impermissible conduct by the law enforcement officers involved.” The record in our view supports that conclusion.
Mr. Lundstrom was not interrogated at extraordinary length. Confessions were deemed involuntary in
Ashcraft v. Tennessee,
322 U.S. 143, 64 S.Ct. 921, 88 L.Ed 1192 (1944), where the suspect was questioned continuously for thirty-six hours without sleep or rest, and in
Mincey v. Arizona,
437 U.S. 385, 98 S.Ct. 2408, 57 L.Ed.2d 290 (1978), where the defendant was subjected to a four hour interrogation
while incapacitated and sedated in an intensive care unit. Mr. Lundstrom’s sixty-four minute interrogation, by contrast, does not appear unwarranted.
Nor was Lundstrom kept in prolonged and deliberate isolation. In
Fikes v. Alabama,
352 U.S. 191, 77 S.Ct. 281, 1 L.Ed.2d 246 (1957), a confession was found involuntary because the suspect was isolated from everyone except his accusers for over a week. Lundstrom’s isolation, by contrast, lasted for little more than three hours, and the record provides no basis for concluding that the police acted inexpeditiously in approaching his interview.
Another factor in the determination of voluntariness is whether the police provided the defendant with basic amenities.
See, e.g., Payne v. Arkansas,
356 U.S. 560, 78 S.Ct. 844, 2 L.Ed.2d 975 (1958). In the present case the police made a commendable effort to make the defendant comfortable. They offered him food and drink and permitted him to rest when necessary. Lundstrom cites
Sample v. Eyman,
469 F.2d 819 (9th Cir.1972), as support for his proposition that, once his emotional state became apparent to the interrogating officers, they should have ceased questioning.
Sample
states that an upset defendant should be given time to recover his composure before continued questioning.
Id.
at 821. The police interviewers in this case did exactly that. When Lundstrom began weeping, they gave him time to recover, then asked him, “Bill, would you like to rest for about a half hour or so and then we’ll come back and talk to you? We want to make you as comfortable as possible.” When Lundstrom was capable of continuing, they did so. The police fulfilled the requirements of
Sample.
We will not overturn a trial court’s ruling on a motion to suppress in the absence of clear and manifest error.
State v. Smith,
136 Ariz. 273, 275, 665 P.2d 995, 997 (1983). We find no error in the admission of defendant’s statements.
DISCLOSURE OF BASIS FOR EXPERT TESTIMONY
Defense psychologist Michael Bay-less, Ph.D., testified at trial that, at the time of the killing, Lundstrom was suffering a “brief reactive psychosis” and was unable to “understand and really decipher the nature and the quality of his behavior.” Bayless attempted at several points to state that his opinions were substantiated by those of Leonardo Garcia-Bunnel, M.D. Dr. Garcia, a psychiatrist for the Correctional Health Services at the Maricopa County Jail, assigned Lundstrom to a psychiatric ward following the killing; he was not called to testify at trial. The second issue on appeal arises from efforts of defense counsel to place the opinions of Dr. Garcia before the jury.
During direct examination, defense counsel asked Bayless if he had conferred with other doctors about the defendant. Bay-less indicated that he had consulted with several, including Dr. Garcia. When defense counsel questioned Bayless about the basis for his opinions, the prosecutor requested a bench conference and informed the court that Bayless, when interviewed, had advised the prosecution that he had
not
relied on another doctor’s opinion. The prosecutor stated his concern that the defense was improperly trying to “backdoor” the opinion of Dr. Garcia into evidence. Garcia had not been listed as a witness. The court found this objection premature.
Bayless proceeded to testify that, when composing his written report about the defendant’s mental condition, he had not relied on anything but the grand jury transcript, the medical examiner’s report, the police report, and his own expertise. Later, however, in an unresponsive answer on direct examination, Bayless volunteered that his view of defendant was substantiated by Dr. Garcia. The prosecution objected and was sustained. In chambers thereafter, defense counsel acknowledged Bay-less’s initial indication that he had not relied upon the opinions of other doctors in preparing his report. Counsel argued, however, that Bayless had later spoken with Dr. Garcia, now relied in part on that conversation, and should be permitted to advise the jury that Garcia, like himself, viewed Lundstrom as having suffered a
“brief reactive psychosis” at the time of the crime.
The court refined its ruling, stating: [Y]ou can bring out that he was relying upon an opinion of Dr. Garcia, as long as you don’t attempt to bring out what the opinion was, and don’t attempt to bootstrap it in after the fact by asking him whether his opinion is consistent with or the same as Dr. Garcia’s opinion.
Back before the jury, Bayless concluded his testimony on direct examination by stating that his opinion, at the time of trial, was based in part on Garcia’s opinion. This was not, however, the last to be heard of Dr. Garcia. Under cross-examination about the point when he believed Lund-strom to have lost contact with reality, Bayless unresponsively added that Dr. Garcia concurred with his opinion. The prosecutor once again objected and moved that the testimony be stricken. The court denied the motion, stating that the prosecutor had invited the answer by failing to conduct his cross-examination by leading questions. Thereafter, on redirect, when Bay-less started once again to refer to Garcia, defense counsel cut him off.
In conference before final argument, the court instructed defense counsel, Thomas A. Thinnes, not to refer to Garcia’s opinion. Counsel responded without objection, “I didn’t intend to.” However, during closing argument on the following day, Thinnes stated,
Mr. Birkemeier asked the doctor, Ph.D. Bayless, what he had done. He asked a rather broad question of what he had done regarding the formulation of his opinion, and the testimony that he was giving in this case. Dr. Bayless answered that he had checked with Dr. Garcia, and Dr. Garcia concurred in his opinion.
The prosecution objected. A bench conference ensued. The court stated, “I thought this was something I mentioned not to get into.” Thinnes argued that, because the court had declined to strike Bayless’s ultimate reference to Garcia’s concurrence, the testimony was appropriate for discussion in closing argument. The court disagreed, struck counsel’s statement, and instructed the jury to disregard it. Later, out of the presence of the jury, the court commented that it should have stricken the testimony about Garcia, but had felt at the time that to do so would only draw it additional attention. The court denied the defendant's motion for a mistrial.
During deliberations, the jury asked for a transcript of Bayless’s testimony. After the court responded that no transcript was available, but that the testimony of all doctors could be read to the jury in open court, the jury replied, “We are only interested in Dr. Bayless’s allusions to other physicians’ opinions relating to Mr. Lundstrom’s mental condition at the time immediately following 2-25-85. If that’s not possible, we’ll rely on our notes and memories.” When the court declined to limit a reading to selected portions of the Bayless testimony, the jury chose to have no testimony reread. It then returned the verdict.
Dr. Bayless’s testimony with respect to Dr. Garcia boils down to this: (1) He composed his written report on Lundstrom’s mental condition without knowledge of or reliance upon Dr. Garcia’s opinion. (2) He became aware of Dr. Garcia’s views before the trial. (3) Learning of Dr. Garcia’s views did not alter Bayless’s opinion. (4) However, he felt that Garcia’s views “substantiated” his own. (5) Garcia’s opinion was therefore part of the data which underlay the opinion Bayless expressed at trial. Lundstrom claims that under these circumstances he should have been allowed to examine Dr. Bayless about Dr. Garcia’s substantive opinion and to refer to it in closing argument and that he was prejudiced by the trial court’s restrictions.
This controversy arises under Rules 703 and 705, Arizona Rules of Evidence.
Rule 703 states:
Bases of Opinion Testimony by Experts: The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to him at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the
subject, the facts or data need not be admissible in evidence.
Rule 705 states:
Disclosure of Facts or Data Underlying Expert Opinion:
The expert may testify in terms of opinion or inference and give his reasons therefor without prior disclosure of the underlying facts or data, unless the court requires otherwise. The expert may in any event be required to disclose the underlying facts or data on cross-examination.
Defendant argues that Rule 703 permits otherwise inadmissible “facts or data” to be admitted as a basis for expert opinion, if “of a type reasonably relied upon by experts,” that Bayless’s reference to Garcia’s opinions meets this requirement, and that the defense should accordingly have been allowed to draw from Bayless the hearsay opinions of Dr. Garcia.
If Dr. Bayless had clearly established at trial that he had
relied
on the opinion of Dr. Garcia in forming his own opinion about the defendant’s mental condition, his recitation of Garcia’s opinion would have been appropriate.
See Hernandez v. Faker,
137 Ariz. 449, 454, 671 P.2d 427, 432 (App.1983) (medical opinions of a nontestifying doctor may amount to “facts and data” upon which a testifying medical expert may reasonably rely and are therefore admissible to show the basis of the testifying expert’s opinion). The threshold question is whether Dr. Bayless relied. Reliance was a preliminary question for the trial judge to decide under Rule 104(a), Arizona Rules of Evidence. (Preliminary questions concerning ... the admissibility of evidence shall be determined by the court____)
Reliance in the context of this case was a mixed question of fact and law. The trial judge properly approached it pursuant to Rule 104 by permitting defense counsel to question Bayless on the subject. Bayless answered vaguely and variously. From his answers, the trial court might appropriately have made the factual finding that Bay-less did not actually rely on Garcia’s opinions, but only found those opinions a bolster to views already fully formed. Such a finding would have justified precluding examination of Bayless in regard to Garcia’s opinion.
See Sharman v. Skaggs Co.,
124 Ariz. 165, 602 P.2d 833 (App.1979), in which Division Two of this court held that an expert witness cannot discuss the opinions of other experts when he has not relied on them in formulating his own.
The trial court apparently found, however, that Bayless, by the time of trial, had relied on Garcia’s opinions at least to some degree. This factual finding leads to an overlapping legal question: whether this was reliance of a type disclosable within the contemplation of Rules 703 and 705. The rationale for including the opinions of non-testifying health care providers among disclosable “facts and data” underlying the opinions of testifying medical experts is found in the advisory committee’s note to Federal Rule 703, incorporated by reference within the comment to Arizona’s Rule 703. The federal committee wrote:
[A] physician in his own practice bases his diagnosis on information from numerous sources and of considerable variety, including statements by patients and relatives, reports and opinions from nurses, technicians and other doctors, hospital records, and X rays. Most of them are admissible in evidence, but only with the expenditure of substantial time in producing and examining various authenticating witnesses. The physician makes life-and-death decisions in reliance upon them. His validation, expertly performed and subject to cross-examination, ought to suffice for judicial purposes.
In the application of this rationale, a distinction might be drawn between building-block opinions and reinforcing opinions. An example of the first would be a family physician’s reliance upon a radiologist’s X-ray interpretation as a building block for ultimate diagnosis. An example of the second is provided by this case. Dr. Bayless, at most, found reinforcement in Dr. Garcia’s concurrence with his ultimate conclusion about the defendant. The first category of opinion was clearly contemplated by Rules 703 and 705 as a disclosable basis for
the opinion of a testifying witness. Whether the second category constitutes reliance within the meaning of Rules 703 and 705 is a question unbriefed by counsel. Though we identify it, we need not resolve it in order to decide the case. Rather, we assume, for purposes of deciding, that the trial court correctly treated Dr. Bayless’s usage of the opinion of Dr. Garcia as a form of reliance appropriate for disclosure pursuant to Rules 703 and 705.
Assuming the validity of this finding, the court too narrowly restricted the testimony of Dr. Bayless, when it ruled, “You can bring out that he was relying upon an opinion of Dr. Garcia, as long as you don’t attempt to bring out what the opinion was.” If Bayless relied within the meaning of Rules 703 and 705, he should not have been confined to stating the fact of his reliance; rather, he should have been permitted to specify that he based his opinion in part on Dr. Garcia’s conclusion that the defendant was suffering a brief, reactive psychosis when the killing occurred.
See Hernandez v. Faker,
137 Ariz. at 451, 671 P.2d at 429 (trial court properly permitted a medical expert witness to disclose as facts or data underlying his opinion the reported
opinion
of another doctor not called to testify at trial).
Any error in the trial court’s ruling was harmless, however, for two reasons. First, the defendant was not entitled to have the jury give substantive weight to Dr. Garcia’s opinion. In
Lynn v. Helitec Corp.,
144 Ariz. 564, 568, 698 P.2d 1283, 1287 (App.1984), we stated:
Facts or data, not admitted or inadmissible, on which an expert may reasonably rely may be revealed to the trier of fact
not as substantive evidence
but to show the basis of an expert’s opinion. Udall & Livermore, Law of Evidence § 23 (2d ed. 1984 Pocket Part.)
(Emphasis added). In
State v. Lessen,
130 Ariz. 1, 7, n. 1 633 P.2d 410, 416, n. 1 (1981), our supreme court likewise stated:
Even if otherwise inadmissible hearsay underlying an expert opinion is introduced into evidence, it has only “the limited purpose of disclosing the basis for the opinion____” Comment to Rule 703, Arizona Rules of Evidence. The hearsay underlying the expert opinion has no substantive value.
Hickok v. G.D. Searle & Co.,
496 F.2d 444, 447 (10th Cir.1974).
Dr. Bayless informed the jury more than once of the facts and data he relied on in formulating his opinions. He stated those opinions in careful detail. Even if the trial court had altogether precluded Bayless from informing the jury that the basis of his views had expanded when he learned of and considered the concurrence of Dr. Garcia, the pertinence of this additional information was so slight that the impact of such preclusion would have been harmless beyond a reasonable doubt.
Yet the trial court did not ultimately preclude such evidence, and this leads to our second basis for holding any error to be harmless: Dr. Bayless finally succeeded in the course of cross-examination in informing the jury that Dr. Garcia’s opinions substantiated his own.
Defense counsel next argues, however, that merely getting the fact of Garcia’s concurrence before the jury did not suffice; rather, he should have been entitled to address the issue during closing argument. Again, because the opinion of Dr. Garcia, if admissible at all, was entitled to no substantive consideration, we conclude that the inability of defendant's counsel to reiterate the fact of Garcia’s concurrence is of minimal significance and does not require reversal.
JURY INSTRUCTIONS
Lundstrom had the burden of proving his insanity to the jury by clear and convincing evidence. A.R.S. § 13-502. In an instruction requested by Lundstrom’s counsel, the trial court defined clear and convincing evidence for the jury as follows:
To be “clear and convincing”, evidence should be “clear” in the sense that it is certain, plain to the understanding, unambiguous; and “convincing” in the sense that it is so reasonable and per
suasive as to cause you to believe it.
In
State v. Renforth,
155 Ariz. 385, 746 P.2d 1315 (Ariz.App.1987), (review granted, CR-0199, Ariz.Sup.Ct. Nov. 17, 1987), this court held that definition to be fundamental error. We stated:
[A] party who has the burden of proof by clear and convincing evidence must persuade the jury that his or her claim is highly probable. This standard is more exacting than the standard of preponderance of the evidence, but less exacting than the standard of proof beyond a reasonable doubt. We believe that a jury could be adequately instructed in some such terms as these. To define clear and convincing evidence as “certain” and “unambiguous,” by contrast, requires an unwarranted intensity of proof.
Id.,
746 P.2d at 1318.
The
Renforth
decision was issued during the pendency of this appeal. Thereafter, defendant moved for and was granted the opportunity to challenge the definition of clear and convincing evidence in the present case. Both parties filed supplemental memoranda on the subject.
For the reasons stated in
Renforth,
we conclude that the trial court incorrectly defined “clear and convincing evidence”. We note that the supreme court has limited review in
Renforth
to the question whether this definitional error constituted fundamental error. We need not attempt to anticipate the supreme court’s resolution of that question, because, assuming arguendo that the definition constituted fundamental error, we hold on the record in this case that the error was harmless.
Fundamental error need not be reversible when there is substantial evidence in the record to support the verdict and it can be said that the error did not, beyond a reasonable doubt, contribute significantly to the verdict.
State v. Henley,
141 Ariz. 465, 468, 687 P.2d 1220, 1223 (1984), quoting
State v. Sorrell,
132 Ariz. 328, 330, 645 P.2d 1242, 1244 (1982).
In
Renforth
we found the court’s incorrect definition to be not only fundamental but reversible error. In this case, however, the court supplemented its definition with language lacking in the
Renforth
instructions:
The defendant’s burden of proving that he is not responsible for criminal conduct by reason of insanity in no way alters the state’s burden of proving the defendant guilty beyond a reasonable doubt. The burden of proof beyond a reasonable doubt is a greater burden of proof than the burden of proof by clear and convincing evidence.
Although this language, in our view, did not expunge the court’s definitional error, it rendered the error harmless. The word “certain,” as we indicated in
Renforth,
may be understood to mean “not to be doubted as a fact: indisputable” and “given to or marked by complete assurance and conviction, lack of doubt, reservation, suspicion, or wavering through or as if through infallible knowledge or perception____” Webster’s Third New International Dictionary of the English Language, Unabridged (1976). Defining clear and convincing evidence in terms of certainty could mislead a jury into believing that the burden of proof by clear and convincing evidence equals or exceeds the burden of proof beyond a reasonable doubt. In the present case, however, by instructing the jury that the clear and convincing burden was a lesser burden, the trial court relieved the risk of such a misunderstanding.
In
State v. Turrentine,
152 Ariz. 61, 68, 730 P.2d 238, 245 (App.1986), a case in which a criminal defendant unsuccessfully claimed insanity, Division Two found that the trial court had incorrectly defined “clear and convincing evidence,” but that its definitional error did not require reversal. The trial court had gone on to advise the jury of the standard’s intermediacy,
requiring “more than a preponderance of the evidence but less than proof beyond a reasonable doubt.” Division Two found this accurate portion of the instruction sufficient to alleviate any error in the remainder.
Here, likewise, we find that the accurate portion of the instruction relieves the inaccurate remainder of the necessity of reversal. Our conclusion is supported by the fact that the deputy county attorney did not exploit the court’s definitional error. He simply and appropriately argued, “You must be convinced. You must be persuaded that the defendant was insane.” Finally, we find that there is substantial evidence in the record to support the verdict. Accordingly, we conclude “that the error did not, beyond a reasonable doubt, contribute significantly to the verdict.”
State v. Henley,
141 Ariz. at 465, 468, 687 P.2d 1220, at 1223 (1984).
CONCLUSION
In summary we find (1) that the record supports the trial court’s conclusion that the defendant’s confession was voluntary and admissible; (2) that the defendant was not prejudiced by the trial court’s evidentia-ry rulings concerning the basis of Dr. Bay-less’s opinions; and (3) that the error in the trial court’s definition of clear and convincing evidence was harmless under the circumstances of this case. The conviction and judgment of the trial court are affirmed.
BROOKS, P.J., and HAIRE, J., concur.