Dooley, J.
¶ 1. Defendant, Donald Wigg, appeals a jury conviction for lewd and lascivious conduct. He claims that: (1) his constitutionally-based presumption of innocence was violated when the trial court permitted a police detective to repeatedly refer to the complainant as the “victim” during testimony; (2) the trial court abused its discretion in excluding testimony from his expert witness on how the interviews of the complainant failed to satisfy the scientifically-suggested protocol for best ensuring accurate interview responses; and (3) he is entitled to acquittal because the jury’s verdict convicting him of lewd and lascivious conduct, while acquitting him of sexual assault, is inherently inconsistent. While we find the trial court erred in permitting the police detective to refer to the complainant as the “victim” and in categorically excluding defense expert’s case-specific testimony, we hold both errors to be harmless. Additionally, we find no inconsistency in the jury’s verdicts. We therefore affirm the trial court’s decision.
¶ 2. In late January 1999, complainant B.M.Y., an eleven-year-old female, accompanied defendant, a thirty-five-year-old male, on a weekend ski vacation from their home state of Connecticut to Mount Snow in West Dover, Vermont. At the time, defendant’s brother was dating complainant’s mother, and the two families had recently taken a ski vacation together. On this occasion, complainant’s mother permitted her to accompany defendant alone. Complainant testified that she took a shower after she and defendant arrived at Mount Snow late in the day. After her shower, she claims that defendant removed her towel, rubbed lotion on her back and legs, and opened her vagina with his fingers and stuck his tongue inside her vagina. Later that same night, after watching television and going to bed, she further alleges that defendant unzipped her pajamas and sucked on her breasts.
¶ 3. B.M.Y. did not tell anyone about the incidents at Mount Snow until September 2000, when her mother’s new boyfriend confronted her after becoming suspicious that something inappropriate had occurred between B.M.Y. and defendant. After B.M.Y. disclosed the details of the trip to her mother’s boyfriend, he reported the incident to the Connecticut State Police. Trooper Francis Budwitz responded to the complaint and conducted an initial interview of B.M.Y. After finding the alleged assault occurred in Vermont, he turned the investigation over to Detective Bich Werner of the Dover, Vermont police
department. Detective Werner traveled to Connecticut and interviewed the complainant on two separate occasions in late 2000.
¶ 4. The State charged defendant with lewd and lascivious conduct with a minor in violation of 13 V.S.A. § 2602 for sucking on complainant’s breasts and with sexual assault on a minor in violation of 13 V.S
A.
§ 3252(a)(3) for inserting his tongue in her vagina.
¶ 5. At trial, the State presented seven witnesses, including complainant, her mother’s boyfriend, Trooper Budwitz and Detective Werner. The complainant testified at length to the nature of her relationship with defendant, the events leading up to the ski trip she took alone with defendant, what occurred during the trip, and why she delayed reporting the incidents of abuse. Trooper Budwitz and Detective Werner testified regarding their investigation, the procedures they utilized in interviewing the child complainant, and what the complainant revealed to them regarding the incidents of abuse. Detective Werner, the lead investigator on the case, repeatedly referred to complainant as the “victim” during his testimony. Despite defense’s timely objection, the trial judge permitted the detective to continue using the reference, ruling that use of the term was not highly prejudicial.
¶ 6. Defendant did not testify. He advanced his main defense — that the complainant’s accusations were untruthful — through cross-examination of each of the State’s witnesses and by calling Dr. Philip Kinsler, a psychologist qualified as an expert in interviewing children suspected of being victims of sexual abuse. In his proffer, defendant’s attorney explained that Dr. Kinsler would testify to the interview techniques that are most likely to produce accurate responses as well as those techniques most likely to produce inaccurate responses. Defendant also intended to have Dr. Kinsler testify about the investigatory interviews conducted by Trooper Budwitz and Detective Werner, in order to give an opinion on whether they were conducted in a manner that was consistent with these suggested techniques. After listening to the expert’s intended testimony in camera, the trial judge permitted him to testify about general interviewing techniques, but excluded testimony analyzing the specific interviews conducted by the police in this case.
¶ 7. In its verdict, the jury acquitted defendant of the more serious sexual assault charge, but convicted him of the charge of lewd and lascivious conduct with a minor. This appeal followed.
¶8. Defendant first contends that the trial court violated his constitutionally-based presumption of innocence when it permitted the
lead police investigator, over objection, to repeatedly refer to the complainant as the ‘Victim” during his testimony. In overruling defendant’s objection, the trial court determined that the detective’s reference to the complainant as the “victim” was not “highly prejudicial” because “the jury knows [the state’s attorney] wouldn’t be here if the State didn’t believe this act occurred.” We agree with defendant’s contention that, where the commission of a crime is in dispute and the core issue is one of the complainant’s credibility, it is error for a trial court to permit a police detective to refer to the complainant as the “victim.” Based on the facts of this case, however, we do not find a violation of defendant’s rights, and hold that the error was harmless.
¶ 9. The trial court may exclude testimony if the probative value “is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury.” V.R.E. 403. The trial court has discretion to balance the factors in Rule 403, and “we will not disturb the trial court’s ruling absent a showing of an abuse of that discretion.”
State v. Webster,
165 Vt. 54, 56, 675 A.2d 1330, 1332 (1996). In criminal trials, the court’s discretion is limited, however, by defendant’s right to due process.
Id.
¶ 10. The first step in balancing is to consider whether the evidence has any probative value.
State v. Ogden,
161 Vt. 336, 341, 640 A.2d 6, 10 (1993). Although Detective Werner had to identify complainant as the alleged victim, the use of the term “victim” had no inherent probative value. See V.R.E. 401;
State v. LeClaire,
2003 VT 4, ¶ 14, 175 Vt. 52, 819 A.2d 719. Thus, with little or no weight on the probative value side of the balance, exclusion of the evidence was appropriate if the court found virtually any danger of unfair prejudice. We hold that there was a danger of unfair prejudice because the detective’s choice of language implied that he and the prosecution believed the complainant’s testimony, adding weight to the State’s case. When defendant objected, the court should have required the officer to use a more neutral term to identify the complainant. The court’s ruling that the officer could continue to use the term “victim” because it was not “highly prejudicial” misapplied the balancing that Rule 403 required. We cannot conclude that the court acted within its discretion.
¶ 11. We also conclude, however, that the error was harmless under the particular circumstances of this case. “‘Harmless error analysis requires the reviewing court to inquire if, absent the alleged error, it is clear beyond a reasonable doubt that the jury would have returned a guilty verdict regardless of the error.’ ”
State v. Wright,
154
Vt. 512, 519-20, 581 A.2d 720, 725 (1989) (quoting
State v. Hamlin,
146 Vt. 97, 106, 499 A.2d 45, 52 (1985)); see V.R.Cr.P. 52(a) (“Any error, defect, irregularity or variance which does not affect substantial rights shall be disregarded.”). After considering Detective Werner’s testimony in its entirety, examining the way in which he used the term “victim,” and considering the degree of danger of unfair prejudice, we conclude beyond a reasonable doubt that the jury would not have returned a different verdict had the detective used different and more neutral terminology.
¶ 12. In reaching our harmless error conclusion, we are mindful of the trial court’s view that the jury would normally expect the investigating officer, who is testifying for the State, to believe the complainant. In this case, however, the context of the detective’s testimony indicates that he was using a term he viewed as synonymous with complainant. See
Jackson v. State,
600 A.2d 21, 24-25 (Del. 1991) (explaining that “the term ‘victim,’ to law enforcement officers, is a term of art synonymous with ‘complaining witness’”). He never expressed an opinion that B.M.Y. was victimized or that defendant was guilty. He referred to defendant as the “suspect” again indicating that he did not intend to convey an opinion of defendant’s guilt. Even if the jury would expect the detective to believe B.M.Y., we do not believe that it would conclude that he expressed that opinion.
¶ 13. The situation here is very different from that in
State v. Ayers,
148 Vt. 421, 535 A.2d 330 (1987),
the main case relied upon by defendant, where a prosecutor made several impermissible statements “that indicated a personal opinion that the defendant’s version of the events was false.”
Id.
at 424, 535 A.2d at 333. Based on these remarks, this Court found that “the jury could draw no other conclusion from the prosecutor’s remarks than that he thought the defendant was guilty.”
Id.
at 425, 535 A.2d at 333. We could not say that the improper statements had no influence on the jury verdict in
Ayers;
we reach a different conclusion in this case.
¶ 14. We have a similar view of
State v. Gokey,
154 Vt. 129, 574 A.2d 766 (1990), another case relied upon by defendant for the proposition that a witness’s use of the term “victim” constitutes reversible error. In
Gokey,
the expert witness testimony included a conclusion that the child complainant had been victimized: “ ‘[S]he had never been through anything else that would cause any signs of emotional disturbance or upset other than this victimization. There’s nothing else that happened to the child up to that point.’ ”
Id.
at 132, 574 A.2d at 767.
The comment on the complainant’s truthfulness, and thus defendant’s guilt, was much more direct, and it came from an expert witness who the jury might view as having special expertise in judging credibility. The main elements that determined the result in
Gokey
are not present here.
¶ 15. Defendant’s second claim is that the trial court erred in limiting the testimony of his expert witness, Dr. Kinsler. Defendant sought three types of testimony from Dr. Kinsler: (1) general testimony to describe the proper and improper methods of questioning children who may have been victims of sexual abuse and to explain why particular methods are proper or improper; (2) case-specific testimony as to whether the officers who interviewed B.M.Y. followed, or deviated from, the proper methods of questioning her; (3) case-specific testimony on the impact of any deviations from proper methodology on the credibility of B.MY.’s accusations against defendant. The trial court admitted the general testimony, that in (1) above, but refused to allow the case-specific testimony in categories (2) and (3), explaining that testimony in either category constituted an impermissible comment on witness credibility. On appeal, defendant claims that the exclusion of the testimony in the second category — whether the officers used proper methodology in questioning B.M.Y. — violated his constitutional rights under the Compulsory Process Clause and Due Process Clause. While we agree that the trial court erred in categorically excluding the expert’s case-specific testimony, we find the error harmless, and defendant’s constitutional claims to be without merit.
¶ 16. Although the State has not argued on appeal that the court erred in admitting the general expert testimony on proper methods of
questioning child sexual abuse victims, we address that issue to fully explain our decision on the case-specific testimony. Testimony from a qualified expert in the form of an opinion is admissible if it “will assist the trier of fact to understand the evidence or to determine a fact in issue.” V.R.E. 702. In our leading case of
State v. Catsam,
148 Vt. 366, 369-70, 534 A.2d 184, 187 (1987), we held that expert testimony of the psychological and emotional profile, and associated behavior of children who suffer post-traumatic stress disorder (PTSD) as victims of child sexual abuse, is admissible under Rule of Evidence 702. Our rationale was as follows:
The unique psychological effects of sexual assault on children place the average juror at a disadvantage in understanding the behavior of the victim. The confusion, shame, guilt, and fear that often result from such abuse may cause a victim to react and behave in a different manner from many other crime victims, especially when the sexual abuse victim is forced to testify to the acts in open court. Jurors who themselves have never experienced such emotions may be better able to assess the credibility of the complaining witness with the benefit of a better understanding of the emotional ante: cedents of the victim’s conduct provided by the expert testimony.
Id.
at 369, 534 A.2d at 187 (internal quotations and citation omitted). We have followed
Catsam
in numerous decisions, most recently expanding its reach to adult rape victims in
State v. Kinney,
171 Vt. 239, 250, 762 A.2d 833, 842 (2000). See also
State v. Leggett,
164 Vt. 599, 599-600, 664 A.2d 271, 271-72 (1995) (mem.);
State v. Gomes,
162 Vt. 319, 329-30, 648 A.2d 396, 403-04 (1994);
State v. Weeks,
160 Vt. 393, 399-403,628 A.2d 1262, 1265-67 (1993);
State v. Denny,
159 Vt. 262, 265, 617 A.2d 425, 427 (1992);
State v. Sims,
158 Vt. 173, 178-82, 608 A.2d 1149, 1152-55 (1991);
State v. Wetherbee,
156 Vt. 425, 430-37, 594 A.2d 390, 392-96 (1991);
Gokey,
154 Vt. at 133-37, 574 A.2d at 768-70;
State v. Hicks,
148 Vt. 459, 461-63, 535 A.2d 776, 777-78 (1987).
Gomes
has some similarity to this case because the State’s expert in that case also testified to whether the memories of young children “can be altered by leading or suggestive questions.” 162 Vt. at 329, 648 A.2d at 404.
¶ 17. Under the general theory of
Catsam,
a large majority of courts have held that the type of general expert evidence introduced in this case, explaining the proper and improper methods of examining children who may be victims of sexual assault, is admissible. See
State
v. Speers,
98 P.3d 560, 565-67 (Ariz. Ct. App. 2004);
State v. Malarney,
617 So. 2d 739, 740-41 (Fla. Dist. Ct. App. 1993) (per curiam);
Barlow v. State,
507 S.E.2d 416, 418 (Ga. 1998);
State v. Criqui,
2003 WL 22119226, *5 (Kan. Ct. App. 2003) (per curiam) (unpublished);
State v. Sloan,
912 S.W.2d 592, 596-97 (Mo. Ct. App. 1995) (per curiam);
State v. Sargent,
738 A.2d 351, 353-54 (N.H. 1999);
State v. Michaels,
642 A.2d 1372, 1384 (N.J. 1994);
State v. Gersin,
668 N.E.2d 486, 488 (Ohio 1996);
State v. Kirschbaum,
535 N.W.2d 462, 466-67 (Wis. Ct. App. 1995). We find the few decisions to the contrary unpersuasive or distinguishable. For example, the Maine court has held inadmissible expert testimony that children are suggestible and sometimes give the expected answer, especially in response to leading questions. The court held that this conclusion was common knowledge and, thus, testimony that would not aid the jury. See
State v. Ellis,
669 A.2d 752, 753-54 (Me. 1996). We conclude that the decision sets too high a barrier to admissibility, a barrier inconsistent with our analysis in
Catsam,
and fails to acknowledge the complexity in translating the general proposition into proper interview protocols.
¶ 18. Similarly, we distinguish the recent decision of the Washington Supreme Court in
State v. Willis,
87 P.3d 1164, 1168 (Wash. 2004), because the court acknowledged the general rule, but held the expert testimony unhelpful because it did not support that there were inaccuracies in the central allegations of the child victim. If anything, decisions like
Willis
support the conclusion we reach in this case.
¶ 19. We also agree with the trial court’s conclusion that evidence in the third category — how the improper questioning of B.M.Y. affected the reliability of her testimony — was inadmissible. Defendant sought admission of this testimony below, but has apparently abandoned his position here.
Again, the
Catsam
line of cases is helpful. In that case, the expert included in her opinion that children who suffer PTSD tend to tell the truth about incidents of sexual abuse and further testified that the victim in the case suffered from PTSD. We held that viewing the testimony as a whole, it “was tantamount to a direct comment that the complainant was telling the truth.”
Catsam,
148 Vt. at 370, 534 A.2d at 187. We concluded that “this type of evidence may unduly influence the jury’s judgment with regard to the truthfulness of
the complaining witness” and was categorically inadmissible.
Id.
at 371, 534 A.2d at 188.
¶ 20. As above, this part of the
Catsam
holding has been reiterated and explained in a number of subsequent cases. See
Weeks,
160 Vt. at 403, 628 A.2d at 1267;
Sims,
158 Vt. at 178-79, 608 A.2d at 1152-53;
Wetherbee,
156 Vt. at 430-32, 594 A.2d at 392-93;
Gokey,
154 Vt. at 140, 574 A.2d at 771. Exclusion of expert testimony that a complainant’s testimony is unreliable because of defects in the initial questioning of her is consistent with
Catsam
and the cases that have followed it. The cases from other jurisdictions are consistent with this view. See
Speers,
98 P.3d at 566;
Barlow,
507 S.E.2d at 418;
Criqui,
2003 WL 22119226, at *4.
¶ 21. This leaves us with the testimony in the second category, the exclusion of which defendant challenges on appeal. Defendant argues that even under the
Catsam
cases, he should have been allowed to ask his expert whether the police officer interviews of B.M.Y. were tainted by improper questioning techniques. The court faced this exact issue in
United States v. Rouse,
111 F.3d 561, 570 (8th Cir. 1997), and noted:
A qualified expert may explain to the jury the dangers of implanted memory and suggestive practices when interviewing or questioning child witnesses, but may not opine as to a child witness’s credibility. That leaves a troublesome line for the trial judge to draw — as the expert applies his or her general opinions and experiences to the case at hand, at what point does this more specific opinion testimony become an undisguised, impermissible comment on a child victim’s veracity?
Id.
at 571. The trial court in this case faced a similar dilemma. In response, it ruled:
It’s up to the jury to determine how the scientific literature applies to the questions that were asked of [B.M.Y.]____By having Dr. Kinsler testify to how his knowledge of the scientific bases applies to this particular case, he would be telling the jury how to apply the scientific information to the case, and that’s the jury’s province____
¶ 22. We conclude that the court’s rationale does not support exclusion of the testimony in the second category. Again, one of the cases in the
Catsam
line is helpful. In
State v. Gokey,
the expert witness testified to the profile of a sexually-abused child, including as well the child’s description of the sexual abuse, the abuser’s name, and a
detailed account of her interview with the child. 154 Vt. at 132, 574 A.2d at 767. This Court found that the evidence was admissible in part, and inadmissible in part, explaining the distinction as follows:
[T]he State introduced evidence of the profile of a sexually abused child to assist the jury in understanding three facts about the child’s behavior. First, according to her mother, the child did not tell her of the December 1986 incident of abuse until about two weeks later. Second, the child claimed that defendant had abused her on many other occasions prior to the December incident, but she had not told her mother until January 1987. Third, although the point was disputed, the child continued to visit the defendant’s house, where her playmate lived, often remaining there when defendant was the only adult present. We hold that these facts in evidence, which might be considered anomalous or unusual, justified the admission of expert testimony for the limited purpose of showing the jury that the child’s behavior in these respects was consistent with the behavior of child sexual abuse victims generally. Without such testimony, the jury would be disadvantaged and might base its deliberations upon misconceptions.
While limited profile evidence was thus permissible, here the expert exceeded the limits. At most, she ought to have been permitted to describe to the jury evidence on the tendency of sexually abused children to delay reporting incidents of abuse and to continue relationships with their abusers, to give her opinion whether the child’s behavior was consistent with this evidence, and to explain the basis for that opinion. Her graphic retelling of the child’s description of the events, her comments about the child’s ability to distinguish truth from falsehood, and her conclusory remarks about “this victimization” were all beyond the pale — unduly prejudicial and unjustifiable under our rules of evidence and our cases.
Id.
at 137-38, 574 A.2d at 770. The important point about
Gokey
is that the Court did not exclude all case-specific expert testimony. It allowed the expert witness to testify that the complainant’s conduct was consistent with the profile, excluding only the testimony that suggested that the expert believed the statements of the complainant. Testimony is not inadmissible simply because it allows the jury to draw inferences that support or weaken the complainant’s credibility. See
Gomes,
162
Vt. at 330, 648 A.2d at 404 (“‘The fact that the jury, if it believes the expert’s testimony, may draw inferences which would tend to bolster the victim’s credibility does not make the evidence inadmissible.’” (quoting
State v. Hicks,
148 Vt. at 461, 535 A.2d at 778)).
¶ 23. To the extent that they address the issue, the decisions from other jurisdictions allow expert testimony in the second category. See
Speers,
98 P.3d at 567 n.3 (limiting expert’s testimony to general evidence on the dangers of improper interviewing techniques “and discussion of the particular practices employed in the instant case”);
Barlow,
507 S.E.2d at 417 (admitting expert testimony to show “whether the interviewing techniques actually utilized were proper”);
Criqui,
2003 WL 22119226, at *5 (finding trial court erred in excluding testimony of “the problems that [the expert] perceived with the interviewing techniques used in this case”);
Sloan,
912 S.W.2d at 597 (error to exclude expert testimony that questions used to examine child were inappropriate);
Michaels,
642 A.2d at 1384;
State v. Ungerer,
1996 WL 362804, *4 (Ohio Ct. App. 1996) (unpublished). As the New Jersey Supreme Court stated in
Michaels,
“[ejxperts may thus be called to aid the jury by explaining the coercive or suggestive propensities of the interviewing techniques employed, but not of course, to offer opinions as to the issue of a child-witness’s credibility, which remains strictly a matter for the jury.” 642 A.2d at 1384.
¶ 24. We stress that the trial court made a blanket decision to exclude the evidence in the second category and did not conduct a specific balancing decision under Rule of Evidence 403. Based on the above decisions and
Gokey,
we conclude that a categorical exclusion of such evidence was erroneous. To the extent the exclusion decision could have been justified in whole or in part by balancing the probative value of the evidence in this case against factors that weighed against admission under Rule 403, it is sufficient for us to note that no balancing occurred. See
State v. Shippee,
2003 VT 106, ¶ 13, 176 Vt. 542, 839 A.2d 566 (mem.) (explaining that while “[t]he discretion of the trial court is broad when reaching a decision based on the balancing test under Rule 403,” error occurs when the defendant can show that the court completely withheld its discretion).
¶ 25. This brings us to the heart of our decision — whether the summary exclusion of the evidence in the second category was harmless in this case. As we said above, we can find harmless error only if it is clear beyond a reasonable doubt that the jury would have returned a guilty verdict regardless of the error.
Wright,
154 Vt. at 519-20, 581 A.2d at 725. Because our harmless error standard does not vary
whether the error is labeled as constitutional or not, it makes no difference whether the court’s refusal to allow the expert to testify about the interviewing techniques used in this case rose to the level of constitutional error as defendant alleges.
State v. Carter,
164 Vt. 545, 555, 674 A.2d 1258, 1265 (1996).
¶ 26. Typically, a harmless error analysis focuses on the evidence of guilt in the record.
Wright,
154 Vt. at 520, 581 A.2d at 725. This case presents a different posture because the effect of the exclusion of the evidence was
not
that the jury was unaware of its content. Evidence of the interviews was admitted through testimony of the interviewers and B.M.Y., and the jury heard about proper interview techniques from the expert. In addition, the cross-examination and closing argument connected the two types of evidence to attack the reliability of the interviews. Although we recognize that counsel’s questions and closing argument are not evidence for the jury to consider, they may be considered in evaluating whether exclusion of the testimony was harmless. See
Delaware v. Van Arsdall,
475 U.S. 673, 684 (1986) (listing among factors for determining whether exclusion of potentially damaging cross-examination was harmless as “whether the testimony was cumulative, the presence or absence of evidence corroborating or contradicting the testimony of the witness on material points, [and] the extent of cross-examination otherwise permitted”).
¶ 27. The jury had the testimony of defendant’s expert on proper and improper interviewing techniques to evaluate the quality of the officers’ interviews of B.M.Y. During cross-examination, defense counsel probed each of the interviewers regarding the interviewing techniques and revealed in what ways their methods were deficient. During Trooper Budwitz’s testimony, he admitted that he had no training in investigating sexual abuse cases, that B.MY.’s mother was present during her interview, that the interview was not recorded and that he followed no interview protocol. Detective Werner testified that he used leading questions during B.M.Y.’s interview, B.MY.’s mother was present during the beginning of her interview, and B.M.Y.’s mother answered questions for her. In addition, B.M.Y.’s statement to Trooper Budwitz and the transcript of Detective Werner’s interview were admitted into evidence. Thus, the content of the interviews were admitted as evidence for the jury to consider.
¶28. In addition, Dr. Kinsler testified about proper interviewing protocols. He explained to the jury that certain techniques make an interview more rehable. He highlighted use of a neutral tone, allowing children to express themselves in their own words, using open-ended
questions, not asking leading questions and not employing bribes or rewards. Dr. Kinsler also explained that an interview conducted by an authority figure may influence the interviewee’s response. In addition, he¡ described how repetition of interviews increases the chance of inaccurate responses.
¶ 29. Finally, in closing argument, defense counsel linked the evidence from the expert testimony to the testimony about the interviews. He pointed out that the interviews were conducted by male police officers — authority figures that would demand respect. He stressed that the interviews contained leading questions and did not follow a reasonable protocol. Additionally, he argued that the interviewers employed suggestive questioning techniques and bribery.
¶ 30. Consequently, the loss caused by the exclusion was that the jury did not hear the defense criticism of the officers’ questioning of B.M.Y. from the mouth of the expert witness. Admittedly, expert testimony might have had a greater impact on the jury. The question we must answer, however, is whether we have a reasonable doubt that the jury verdict would have been different if the critical evaluation of the officers’ questioning methods came from the expert rather than from defense counsel’s cross-examination and argument based on the expert’s general testimony.
¶ 31. We are aided in our evaluation of the harmless error question by two decisions from other jurisdictions. The first is
Rouse,
which is generally indistinguishable from this case. As in this case, the trial court in that case allowed general expert testimony about proper methods of interviewing children alleged to be victims of sexual assault, but excluded case-specific evidence criticizing the methods and techniques of questioning actually used in investigating the case. Apparently, a majority of the three-judge panel agreed that the exclusion was erroneous, but a majority also agreed that the error was harmless. 111 F.3d at 572. It cited three factors in reaching its conclusion: (1) the jury heard the general expert testimony and the closing argument in which defense counsel relied upon the expert testimony to criticize the interviews of the complainants; (2) the jury heard evidence of the investigative techniques actually used, “learned of the social influences affecting the victims at the time they accused their uncles of sexual abuse” and observed the testimony of the complainants; and (3) the complainants’ trial testimony was consistent with the story they told before they were interviewed by law enforcement personnel and entered therapy.
Id.
In general, the same factors are present here. As described, Dr. Kinsler generally testified about proper interviewing
techniques, and defense counsel relied on this testimony during closing argument and cross-examination to question the techniques used in this case. Further, the jury heard about the contents of each interview. Finally, B.M.Y.’s first disclosures of sexual abuse were made to her mother’s boyfriend before she was interviewed by police, and her trial testimony was consistent with this initial disclosure.
¶32. Similarly, in
Commonwealth v. Allen,
665 N.E.2d 105, 109 (Mass. App. Ct. 1996), the trial court admitted expert testimony “regarding proper and improper interview techniques to be employed with children alleged to be the victims of sexual abuse,” but refused to allow the defense expert to directly testify about the questions employed in the interviews of the complainants. The court expressed that this testimony would “impermissibly comment on the children’s credibility.”
Id.
The appeals court found no error in that the expert was allowed to testify generally about improper interviewing techniques and the defendant was permitted to introduce information about the interviews.
Id.
at 109-10; see also
Criqui,
2003 WL 22119226, at *5 (concluding that exclusion of all expert testimony on improper interviewing techniques was harmless, in part, because defense attorney “was able to bring the alleged interviewing errors to the jury’s attention” in cross-exam and closing argument).
¶33. We recognize that this case does not involve the more typical kind of harmless error where the weight of the prosecution evidence is so overwhelming that no single evidentiary error is likely to make a difference. See
State v. Oscarson,
2004 VT 4, ¶ 32, 176 Vt. 176, 845 A.2d 337 (emphasizing that the most important factors are the strength of the State’s case and the weight of the evidence in issue). The analysis in
Rouse
and
Allen
is consistent with our harmless error precedents, particularly in giving central weight to the cumulative nature of the excluded evidence. See
Oscarson,
2004 VT 4, ¶ 32;
State v. Crannell,
170 Vt. 387, 397, 750 A.2d 1002, 1012 (2000);
State v. Fisher,
167 Vt. 36, 41-42, 702 A.2d 41, 45 (1997);
State v. Venman,
151 Vt. 561, 569, 564 A.2d 574, 579 (1989). The excluded testimony was cumulative in the sense that the protocols and substance of the interviews had already been presented to the jury. Overall, we agree that the testimony of the expert witness linking the evidence already presented was highly unlikely to change the result. By rejecting the sexual assault charge, the jury was clearly aware of the reliability questions surrounding B.M.Y.’s testimony. We cannot conclude that more specific testimony from the expert witness — drawing conclusions that were obvious based on his earlier testimony, and the cross-examination and
argument of defense counsel — had a sufficient possibility of changing the result to undermine defendant’s conviction for lewd and lascivious conduct. We therefore hold that the erroneous exclusion of the expert’s case-specific testimony critiquing the interview techniques used in questioning B.M.Y. was harmless beyond a reasonable doubt.
¶ 34. Defendant’s final claim is that the jury’s verdict of guilt on the lewd and lascivious conduct charge is inconsistent with its acquittal on the charge of sexual assault of a minor. In support of this argument, defendant asserts that because each charge is based on conduct that occurred during the same night during a ski trip in Vermont, and the basis for each is the testimony of B.M.Y., the jury could not have acquitted on one charge and convicted on the other. Transparently, defendant asserts, the jury rendered a compromise verdict.
¶ 35. Essentially, defendant attacks the jury’s fact-finding. This Court applies a clearly erroneous standard in reviewing a jury’s findings of fact and will reverse only if the evidence is insufficient to support the finding.
State v. Martin,
145 Vt. 562, 572, 496 A.2d 442, 449 (1985).
¶ 36. In making his argument, defendant relies almost exclusively on
State v. Crepeault,
167 Vt. 209, 704 A.2d 778 (1997), where this Court reversed a jury verdict of guilt on an aggravated sexual assault charge because it was inconsistent with the jury’s acquittal of the defendant on the three other charges in the information. The key factor that led to reversal in
Crepeault
was that the aggravated sexual assault charge did not have a separate factual basis, but depended on acts specified in the other three counts.
Id.
at 214, 704 A.2d at 782. The jury found the defendant not guilty on the other three counts, but convicted defendant of the fourth count of aggravated sexual assault.
Id.
In reversing, this Court explained that “it is readily apparent that the general charge of aggravated sexual assault ... was not submitted to the jury as a separate and independent offense,” but was based “upon the assumption that defendant might be convicted of one or more of the sexual acts alleged in the first three counts, and was included solely to carry the additional serious-bodily-injury enhancement.”
Id.
at 212, 704 A.2d at 781.
¶ 37. The situation in this case is very different. Although they allegedly occurred during the same night, and involved the same victim, the counts the jury considered here were separate and independent. See
State v. Carpenter,
155 Vt. 59, 64, 580 A.2d 497, 500 (1990) (finding no inconsistency in jury’s decision to convict defendant of aggravated assault but acquit him of sexual assault where the two
offenses required different elements and were independent of one another). In convicting defendant of lewd and lascivious conduct, the jury necessarily determined beyond a reasonable doubt that he had committed the specific act of sucMng on the complainant’s breasts. In acquitting defendant of sexual assault as alleged, the jury found that defendant had not committed the separate act of inserting his tongue in her vagina. Even though both findings were based primarily on the jury’s evaluation of one witness’ credibility, the complainant, the jury was free to believe her in part and disbelieve her in part. See
State v. Pelican,
160 Vt. 536, 541, 632 A.2d 24, 27 (1993) (clarifying jury’s role as arbiter of credibility);
State v. Fairbanks,
123 Vt. 298, 300, 187 A.2d 335, 336 (1963) (noting that the trier of fact is the judge of a witness’ credibility). Neither finding is clearly erroneous, and there is no ground to upset the verdicts.
Affirmed.