MILKEY, J.

After a jury trial in Superior Court, the defendant was convicted of four counts of forcible rape of a child (G. L. c. 265, § 22A). The Commonwealth concedes that there were several violations of the first complaint doctrine at trial. The defendant objected to some of the errors, but not to others. The question we face is whether the combination of preserved and unpreserved errors requires a new trial. We conclude that it does. Background. The defendant is the father of the young girl who alleges that she was raped. We will refer to her as Becka. Becka's parents had separated, and during the relevant period, she lived principally with her mother, her mother's boyfriend (Elvin), and her two sisters. The defendant lived with his sister's family, and Becka visited the defendant and her paternal relatives at her aunt's house on weekends, two to three times per month. Becka's sisters, who have a different biological father, did not accompany her on those visits.

Becka testified that when she was eight, the defendant began anally raping her at her aunt's house, and that he continued to do this over several months' worth of visits. According to Becka, the rapes occurred either at night while the others in the house-hold were sleeping, or during the day when they were not home. While her testimony was brief, Becka provided some detail about the alleged acts, and she demonstrated to the jury, away from the witness stand, how her body was positioned while the acts occurred.

At the end of this graphic testimony, the prosecutor asked Becka if the man who raped her was in the courtroom. To the prosecutor's evident surprise, Becka answered, "No." The prosecutor then asked: "You don't see your dad in the courtroom today?" Becka again answered, "No." The prosecutor pressed her once more: "Your dad is not sitting over there in the courtroom?" At this point, Becka did identify her father, and the prosecutor prompted her to agree that he looked different because he did not "used to have glasses and a beard like that."

Defense counsel conducted only a limited cross-examination of Becka. Counsel elicited from her that while Becka referred to the defendant by his nickname, "Manico," she and her sisters called her mother's boyfriend Elvin, "Dad." Becka also acknowledged that Elvin moved out "not very long after this happened."

The Commonwealth called four additional witnesses, all of whom testified to events that occurred after the alleged incidents. The witness who immediately followed Becka was her older sister, Selena, who was Becka's close confidante and "secret keeper." Selena recounted a series of conversations that took place in March of 2004, while the mother, Becka, and Selena were traveling to school in the family minivan. According to Selena, Becka became extremely upset when their mother unexpectedly announced that Becka was to spend the upcoming weekend with the defendant. In response to questions from her mother and from Selena as to why she was so upset, Becka initially declined to explain why. After Selena continued to press, Becka, "whispering" into Selena's ear, told Selena that the defendant had raped her. While they were all still in the minivan, Selena eventually repeated Becka's allegations to their mother, and the mother was drawn into what was (at least in part) a three-way conversation.

After Selena testified, the mother gave her own account of the minivan conversations, over the defendant's objection that the mother could not serve as a "second first complaint witness." The two accounts were consistent, with Selena and the mother each providing some detail absent from the other's telling. The mother also testified about events that occurred in the aftermath of Becka's having revealed her allegations: that the mother repeated these allegations to Becka's great-grandmother, that the mother also passed them along to the State agency then known as the Department of Social Services (DSS), and that DSS directed Becka to come to the hospital.

The next witness was Dr. Nancy Miller, a pediatrician at the Family Advocacy Center at Baystate Children's Hospital. At the start of her testimony, Dr. Miller described the Family Advocacy Center as "a child advocacy center where we work with children and families who have been affected by child abuse and/or domestic violence." She explained the various ways that children are referred to the center, and she noted that she personally had examined "[s]everal thousand" children there. In response to the prosecutor's question as to why it was important for children to be examined at the Family Advocacy Center rather than at an emergency room, she responded without objection: "[W]e are experts in what we are doing."

Dr. Miller testified that she had met with Becka on April 26, 2004, a few weeks after Becka had reported the allegations to her sister and her mother. When Dr. Miller began to testify about what Becka had told her and shown her using anatomically correct dolls, the defendant objected. After the judge overruled the objection, Dr. Miller testified that Becka was able to demonstrate with the dolls "what happened to her." Dr. Miller then testified that she proceeded to conduct a physical examination of Becka during which she observed no signs of trauma, such as "tearing, redness, [or] bruising," in the relevant anatomical areas. Immediately thereafter, the prosecutor asked, "[D]id that surprise you?" Dr. Miller replied, "No," and when she started to respond to the prosecutor's follow-up question, "Why not?" the defendant objected again. At sidebar, defense counsel stated that the objection was based on the ground that the questioning was beginning to call for expert testimony and Dr. Miller had not been qualified as an expert. Defense counsel added that, regardless of whether Dr. Miller might qualify as an expert, the Commonwealth had not notified the defendant that it intended to call her as one and had not complied with the discovery rules regarding expert witnesses.

After counsel debated these issues at length, the judge rejected all of the defendant's arguments and allowed Dr. Miller to testify freely as an expert witness. She proceeded to recount in detail why she was not surprised to observe that Becka did not show any signs of physical abuse, supporting her views with references to unspecified scientific studies. As but one example, she testified, without further objection, that "most of the time, eighty-five percent or more of the time, when we look at scientific studies of children who have been sexually abused, they will have normal examinations, that is, due to the nature of the alleged acts and the physiology of the issue involved."

Defense counsel undertook only a brief cross-examination of Dr. Miller, totaling less than six pages of transcript. Counsel did not challenge Dr. Miller's credentials or conclusions and largely focused on getting clarification that Dr. Miller was not asserting that she had observed physical signs that positively demonstrated a sexual assault but only that the absence of such signs did not rule it out.

On redirect, the prosecutor returned to the subject of what Becka had communicated to Dr. Miller. In response to the prosecutor's questions, Dr. Miller repeatedly stated that when Becka had described what had happened to her, she was speaking about the defendant and not Elvin (even though it was Elvin whom Becka called "Dad").

The Commonwealth's final witness was a social worker at DSS. On direct examination, she testified about the investigatory practices that DSS used both in general and in this case. Although she stated that she had interviewed Becka as part of this process, the social worker did not repeat the substance of those conversations. On cross-examination, defense counsel elicited from her that, when Becka had been asked whether she knew the difference between a truth and a lie, she had answered, "No." The social worker also acknowledged that at the time of the alleged incidents, the mother had an open and "supported" case file at DSS for physical abuse and neglect, and that Elvin "apparently" did as well. On redirect, the social worker explained that she had asked Becka a series of follow-up questions designed to determine that she could discern the difference between the truth and a lie, and the social worker indicated that Becka could do so. The social worker also stated that she had not investigated Elvin for any sex-related allegations.

After the Commonwealth rested, the defendant called only one witness, his niece (Becka's cousin), who lived in the house where and when the rapes allegedly occurred. The main subject of her testimony was the sleeping arrangements in the house. She testified on direct examination that when Becka stayed at the house, Becka slept with her in her bedroom, while the defendant slept in the living room. On cross-examination, however, the niece abandoned that position and acknowledged that the defendant sometimes slept in the bedroom with Becka.

Discussion. Selena testified as the Commonwealth's designated first complaint witness. see Commonwealth v. King, 445 Mass. 217, 242-245 (2005). The defendant raises no claims of error about Selena's testimony but challenges instead the additional complaint testimony that came in through the mother, Dr. Miller, and the DSS social worker. Of the seven individual violations of the first complaint doctrine that the defendant alleges, the Commonwealth concedes that five of them were error, and it does not contend that any of this complaint testimony could have been admitted on alternative grounds. See Commonwealth v. Arana, 453 Mass. 214, 229 (2009); Commonwealth v. Monteiro, 75 Mass. App. Ct. 489, 494-495 (2009). The Commonwealth nevertheless argues that the various violations do not amount to reversible error, because "they did not prejudice the defendant (for those [errors] preserved) or give rise to a substantial risk of a miscarriage of justice (for the unpreserved)."

We agree with the Commonwealth that many of the errors, including some of the ones the defendant has chosen to highlight, are of little moment (at least when they are viewed individually). For example, the mother's separate account of the minivan conversation ultimately added minimal, if any, force to the Commonwealth's case regardless of whether it should have been admitted. In addition, the mother's testimony that she repeated Becka's allegations to her great-grandmother is more notable for its irrelevancy than its prejudicial effect.

However, other errors are more serious. Of particular concern is the testimony of Dr. Miller about what Becka had told and showed her. As the Commonwealth now concedes, it was error to admit Dr. Miller's testimony about this, and the defendant raised a timely objection. Therefore, we can affirm only if we are convinced "that the error did not influence the jury, or had but very slight effect." Commonwealth v. Flebotte, 417 Mass. 348, 353 (1994), quoting from Commonwealth v. Peruzzi, 15 Mass. App. Ct. 437, 445 (1983). For the reasons that follow, we cannot reach that conclusion.

As is generally true in prosecutions of this sort, the Commonwealth's case ultimately rested on the credibility of a single eyewitness. see Commonwealth v. Mendrala, 20 Mass. App. Ct. 398, 402 (1985), quoting from Commonwealth v. Gardner, 350 Mass. 664, 667 (1966) ("As is often the case where there are charges of rape or sexual assault, the question of guilt or innocence rests in large part `upon whether the jury believed the victim's version of what happened or the defendant's]'"). In this context, allowing multiple witnesses to testify that an alleged victim relayed her allegations to them by itself creates potential prejudice. see Commonwealth v. Stuckich, 450 Mass. 449, 457 (2008) ("Repetition of the narrative tends to enhance the credibility of the complainant to the prejudice of the defendant"). See also Commonwealth v. Monteiro, 75 Mass. App. Ct. at 497 ("In a case such as this one, which turned on credibility, there is a particularly high probability of prejudice from the admission of duplicative complaint evidence").

Here, the Commonwealth did more than merely elicit that Becka had repeated her allegations to multiple parties. Most importantly, although Dr. Miller never specifically was asked whether she believed Becka, her testimony effectively communicated to the jury that she did. As noted, supra, Dr. Miller was allowed to testify that Becka was able to demonstrate with the dolls "what happened to her." Moreover, directly thereafter, the prosecutor pointedly asked Dr. Miller whether the absence of physical trauma "surprise[d]" her. Dr. Miller's answering, "No," and her explanation of why she was not surprised by the absence of physical trauma necessarily conveyed that she believed Becka's allegations to be true.

The potential prejudice caused by Dr. Miller's implicitly vouching for Becka's allegations was significantly amplified by the fact that Dr. Miller was called to testify as an expert witness. see Commonwealth v. Quincy Q., 434 Mass. 859, 872 (2001), quoting from Commonwealth v. Richardson, 423 Mass. 180, 186 (1996) ("The danger of vouching `is greater where the witness is testifying as both a direct witness and an expert, particularly where the witness offers fresh complaint testimony'"). This court, and the Supreme Judicial Court, long and frequently have cautioned about the potential dangers of using the same witness as both an expert and a percipient witness in criminal cases, especially in sexual assault cases. See, e.g., Commonwealth v. McCaffrey, 36 Mass. App. Ct. 583, 590-594 (1994), and cases cited. As we said in McCaffrey, id. at 591, "[t]here appears to be particular risk of undue influence when an expert's testimony can be construed as impliedly supporting the truthfulness of a victim in sexual abuse prosecutions." Where Dr. Miller had been put before the jury as an expert who had examined "[s]everal thousand" child abuse victims, her conveying that she believed that Becka was telling the truth bolstered Becka's credibility to a degree that cannot be over-looked. We are unable to conclude with confidence that this "did not influence the jury, or had but very slight effect." Commonwealth v. Flebotte, 417 Mass. at 353.

Other errors, which were not preserved, were not of great import, but neither were they trivial. For example, the testimony about the Commonwealth's investigatory process (which came in through the mother and the DSS social worker) at least to some extent established "[t]he fact that the Commonwealth brought its resources to bear on this incident[, which] creates the imprimatur of official belief" in the alleged victim. Commonwealth v. Stuckich, 450 Mass. at 457 (concluding that testimony about the government's sexual assault victim interview process ordinarily should be excluded because such testimony "is unnecessary and irrelevant to the issue of the defendant's guilt, and is extremely prejudicial"). see Commonwealth v. Monteiro, 75 Mass. App. Ct. at 493-494. Our conclusion that a new trial is required is reinforced by the volume of the errors present. "Even though not all of the testimony challenged on appeal was the subject of objection at trial, we conclude that appellate relief is warranted based upon the aggregate errors in the case, regardless of the applicable standard of review." Id. at 497.

Finally, some comments are appropriate regarding the defendant's contention that the Commonwealth failed to disclose that it intended to call Dr. Miller as an expert witness. The full extent to which the defendant can make out such a claim cannot be determined on the record before us. To the extent that the defendant wanted to rely on this argument, he should have sought to develop the relevant facts in his motion for a new trial. However, regardless of whether any such claim might have provided an independent ground for reversal, we note that our review of the record leaves us with the unmistakable impression that, for whatever reason, defense counsel was unprepared to engage Dr. Miller as an expert witness, despite the fact that her expert testimony went to the heart of one of the defendant's principal defenses. This is one of the considerations that precludes us from concluding that "on the whole record we are satisfied the jury could find their way and the defendant was well defended and fairly tried." Commonwealth v. O'Brien, 35 Mass. App. 827, 833-834 (1994) (Kaplan, J.) (where "[t]he defense conducted a strong cross-examination of the [child's treating] psychotherapist," no reversible error found as to defense claim that, "to preserve clear lines, the prosecution should have brought in an expert other than one who served as therapist to the child").

In sum, we conclude that the errors require a new trial. Judgments reversed. Verdicts set aside.