SCHULTZ, Justice.
Defendant appeals from his conviction and sentence for indecent contact with a child. See Iowa Code § 709.12(2) (1983). He contends that the trial court erred in two respects: (1) in overruling his objections to the admission of expert testimony that children almost never lie about sexual abuse; and (2) in failing to grant a mistrial after a police officer made reference to the reputations of persons with whom defendant resided. We conclude that the trial court abused its discretion in admitting the challenged expert testimony, thereby depriving defendant of a fair trial. We there[92]*92fore reverse the conviction and remand for a new trial.
The State charged defendant with having indecent contact with an eight-year-old female victim (the complainant) one night in April 1984. The incident allegedly involved sexual molestation which occurred while the complainant’s five-year-old sister and she were asleep in a hide-a-way bed at the home of an acquaintance of their mother.
At trial the prosecution first called as witnesses the complainant, her sister, and other witnesses who gave testimony tending to support each element of the charge. The prosecution then elicited opinion testimony of two expert witnesses: Martha Davis, a prosecution witness who was the principal of the elementary school attended by the complainant; and Mary Jane Mosh-er, a child abuse investigator employed by the Iowa Department of Human Services whom the defendant called as a witness and the prosecution cross-examined. The essence of their challenged testimony was that children generally tell the truth when they report that they have been sexually abused. Defendant has not challenged the qualifications of these two expert witnesses, but he contends that their opinions concerning the general credibility of children reporting sexual molestation was not a proper subject for expert testimony.
Davis, the school principal, testified:
Q. During the course of your investigation and your work with sexually abused children, have you formed an opinion as to whether or not children lie about these particular kinds of incidents?
[[Image here]]
A. Yes, I have. First of all, I think it relates to two areas. One would be the statistics that I’ve learned of by going to special training sessions on this matter of child sexual abuse.
Statistically, the statistics show that children have not lied in this area. For example, we were given a statistic from Polk County that — I’m referring to the Polk County Child Sexual Abuse Model that is utilized in the case of an investigation — and out of about, I think it was, 75 cases, there was only one out of prosecuted cases where the child was not telling the truth.
And we’ve been given other statistics along this line. At our symposium last Wednesday that was emphasized again and again, children do not lie about this type of matter.
And the second part of my opinion has to do with my experiences, which has been the entire time I’ve been at [the elementary school], for three years, that there has not been a child who has lied about something like this. So I do believe the children.
Mosher, the child abuse investigator, had interviewed the complainant about this incident and was called as a witness by the defendant. On direct examination she related what the complainant had told her, a report somewhat inconsistent with the complainant’s trial testimony. On cross-examination of Mosher the prosecution developed the following testimony:
Q. Have you formed an opinion during the course of your investigations with regard to sexual abuse cases whether or not children in general tell the truth about these particular kinds of crimes?
A. Yes I have.
Q. And what is that opinion?
[[Image here]]
A. In all the years I’ve worked, which is sixteen, I have only had one child that lied to me about sexual abuse, and it is my opinion that it is very rare for a child to lie about this subject.
Q. Did you form your opinion based upon certain statistics or facts that you have at your recollection?
A. I did read a statistic based on the Giaretto Program in San Jose, which was one of the first programs in the sexual abuse investigation, and their finding was that perhaps one in 2,500 children who were interviewed did not tell the truth, which would make it exceedingly rare.
Defendant made timely objection that the testimony of Davis and Mosher was not a [93]*93proper subject for expert opinion testimony. Before either witness expressed an opinion about whether children in general truthfully report incidents of child abuse, defendant’s counsel objected by stating:
Your Honor, and I do not know what her opinion is, but there is no theory of evidence, there is no theory of law, and it is just directly prohibited by the Iowa Supreme Court to ask. No one can be an expert in that area. One cannot give an opinion in that area, your Honor.
Objecting to the testimony of Mosher, defendant’s counsel stated:
Your Honor, I would like to interpose an objection for the record. That is not the proper subject for expert opinion.
The Iowa Supreme Court has said in State v. Galloway [275 N.W.2d 736, 740 (Iowa 1979) (specially concurring opinion of seven justices)] that the jury, of course, can listen to facts, but it is the jury’s job to determine credibility, and it is not — it is — just simply is not a proper subject for expert testimony.
Counsel’s objection that no person can be an expert in this area, and his specific reference to the Galloway opinion, alerted the trial court to the important concerns which we here express regarding this testimony. We believe that defendant properly preserved error on this matter at trial. See State v. Nimmo, 247 N.W.2d 228, 231 (Iowa 1976). The qualifications of these two witnesses is not in issue; the subject matter of their expert opinion testimony is our focus. The fundamental question we must answer is whether the prosecution met its burden to show that the subject matter of the testimony is admissible pursuant to Iowa Rule of Evidence 702:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education may testify thereto in the form of an opinion or otherwise.
We are generally committed to a liberal rule which allows opinion testimony if it will aid the jury in screening the properly admitted evidence to ascertain the truth. State v. Hall, 297 N.W.2d 80, 84 (Iowa 1980). When a trial court has exercised its discretion to admit expert testimony, we will reverse only if we find an abuse of that discretion and prejudice. Id. at 86; State v. Morrison, 323 N.W.2d 254, 256 (Iowa 1982). An abuse of discretion occurs when the trial court’s ruling is based on untenable grounds. Morrison, 323 N.W.2d at 256. Consequently, when a timely and specific objection under rule 702 challenges the propriety of expert testimony on a particular subject, the proponent has the burden to establish that the testimony is admissible.
With these general principles in mind, we examine the expert testimony to determine its purpose and effect. The State urges that the testimony “was offered merely as an aid to the jury in understanding the issue of the truthfulness of children, in general, who claim to have been sexually abused.” (Emphasis in original.) When viewed in light of the factual issues, this contention is unrealistic. The credibility of the eight-year-old child was a fighting issue between the parties. The gist of the principal’s testimony was that during her three-year period as principal no child had lied to her about being sexually abused; the investigator opined that in 16 years of work she had only one child lie to her about sexual abuse. The prosecutor’s obvious purpose in offering this expert testimony was to bolster the complainant’s credibility. We believe the effect was comparable to telling the jury that the complainant would not lie about matters concerning sexual abuse.
The issue presented in this appeal is whether expert testimony that a witness would not lie is admissible under rule 702. We have not been cited nor do we find any of our cases that are on point on this issue. Our case law does provide some assistance, however.
Initially, we observe that Iowa Rule of Evidence 702 is of recent origin. It was prescribed and reported by this court to the legislature and enacted in that form taking [94]*94effect July 1,1983. See 1983 Iowa Acts ch. 219; Iowa Code § 684.19. Rule 702 merely states our prevailing case law. Iowa R.Evid. 702 committee comment 1983.
We first examine Galloway, the authority defendant called to the court’s attention in his objection. A majority of this court in a special concurrence decided that the “trial court on retrial would be justified in sustaining an objection” to an offer of expert opinion testimony concerning the accuracy of eyewitness identification in general, even though the eyewitness’ identification was in question. Galloway, 275 N.W.2d at 741 (Reynoldson, C.J., concurring specially). We stated:
In the final analysis, jurors daily experience the fragility of their own memories. They know recollection fades with time and is affected by the relative significance of the incident. Probably most have experienced on several occasions their own or another’s misidentification in social or business relationships. Explanation of the scientifically identified mechanisms which bring about memory decay may be of academic interest, but it is of little aid to the jury in judging reliability of the particular eyewitness identification before them.
Id. We then commented on the cost and time which would be required by the admission of such opinions and concluded: “We should be slow to approve a rule which would inject another group of expensive forensic experts into an area historically reserved as a jury’s domain." Id. at 742 (emphasis added). The holding by the majority in Galloway allows a restriction on the use of expert opinions offered to attack the credibility of witnesses, but it falls short of deciding the admissibility of an expert opinion that directly comments on the truthfulness of a witness or whether the trial court would err as a matter of law in admitting such evidence.
Iowa has long been a leader in the liberal use of expert witnesses presenting opinion evidence. The seminal case in Iowa jurisprudence on the use of expert opinions is Grismore v. Consolidated Products Co., 232 Iowa 328, 5 N.W.2d 646 (1942). In Grismore we set forth liberal rules concerning the admissibility of expert opinion testimony and these rules are the genesis for most of our modern ease law in this area. Id. at 339-61, 5 N.W.2d 654-63. We also expressed guidelines for limitations on the admission of expert opinions when we stated:
Of course, we are not to be understood as saying that opinion testimony should be received as to all matters. It should be received only as to such matters as are the proper subjects of opinion testimony. No witness should be permitted to give his opinion directly that a person is guilty or innocent, or is criminally responsible or irresponsible, or that a person was negligent or not negligent, or that he had capacity to execute a will or deed.... But the reason is that such matters are not subjects of opinion testimony. They are mixed questions of law and fact.
Id. at 361, 5 N.W.2d at 663. Expert opinions regarding whether a witness testified truthfully bear similarity to expressing an opinion on the defendant’s guilt or innocence. Typically, the truthfulness of the victim or the accused bears heavily upon, and is intertwined with, the guilt or innocence of the accused. Both matters, credibility of a witness and the determination of the guilt or innocence of the accused, are reserved solely for the fact finder. Does this similarity require a court to exclude expert opinion testimony regarding the truthfulness of a witness?
Although some of our other cases aid us in addressing this issue, they do not provide a complete answer. In State v. Galloway, 167 N.W.2d 89, 93 (Iowa 1969), the criminal defendant signed a waiver for taking a polygraph test and agreed that the results could be admitted into evidence. Because of the defendant’s prior consent, we held admissible the polygraph operator’s opinion that the five answers given by the defendant in reply were untrue. Id. at 94. However, we noted that the operator could not be asked or permitted to answer [95]*95directly whether the defendant was involved in the shooting, even though defendant denied involvement in the shooting in one of his five replies. Id. We stated that “[i]f defendant was involved in the shooting in this case he was involved in a murder, in other words, guilty.” Id. We reversed a drug delivery conviction in State v. Oppedal, 232 N.W.2d 517, 524 (Iowa 1975), when an officer opined that a quantity of marijuana seized from defendant was possessed by him “with intent to deliver,” an essential element of the crime. We stated that the officer’s opinion was “tantamount to permitting the witness to testify that he had an opinion as to the ultimate fact of defendant’s guilt or innocence.” Id. We have placed similar limitations on other opinion testimony. See Nimmo, 247 N.W.2d at 230; State v. Swartz, 244 N.W.2d 553, 555 (Iowa 1976); State v. Ogg, 243 N.W.2d 620, 621 (Iowa 1976); State v. Droste, 232 N.W.2d 483, 488-89 (Iowa 1975); State v. Horton, 231 N.W.2d 36, 38 (Iowa 1975).
These cases reveal that we generally will not admit expert opinion testimony that essentially passes on the guilt or innocence of the defendant. These inadmissible opinions go a step beyond merely aiding the fact finder in understanding the evidence and actually invade the exclusive domain of the jury, that is, the determination of the guilt or innocence of the accused. Additionally, expert opinions on the truthfulness of a witness should generally be excluded because weighing the truthfulness of a witness is a matter reserved exclusively to the fact finder.
Other jurisdictions hold that it is improper in criminal cases to permit a witness, expert or otherwise, to express an opinion as to whether another witness is telling the truth. See, e.g., State v. Sullivan, 2 Conn. Cir. 412, 199 A.2d 709, 713 (1964) (impermissible for doctor to opine as to reliability of lay witness’ opinion regarding defendant’s sobriety); State v. Perkins, 444 N.E.2d 878, 879 (Ind.App.1983) (psychiatrist not allowed to testify that it was his “strong impression” that defendant was telling the truth); Commonwealth v. Carter, 9 Mass.App. 680, 403 N.E.2d 1191, 1193 (1980) (doctor’s opinion that mildly retarded child was telling the truth held objectionable); State v. Williams, 34 N.C.App. 408, 238 S.E.2d 668, 672 (1977) (psychiatrist’s opinion that he considered defendant reliable properly excluded); Beishir v. State, 522 S.W.2d 761, 765 (Mo.1975) (expert testimony that a witness is not truthful inadmissible); State v. Correra, 430 A.2d 1251, 1255 (R.I.1981) (“No witness, expert or otherwise, may testify that another witness or the defendant is lying or faking. This is an area within the exclusive province of the jury.”); Hopkins v. State, 480 S.W.2d 212, 220-21 (Tex.Crim.1972) (psychiatric evidence not admissible regarding credibility of witness); contra State v. Kim, 64 Haw. 598, 645 P.2d 1330, 1334-37 (1982) (not an abuse of discretion to permit a child psychiatrist to testify that he found complainant’s account of alleged rape to be believable); People v. Cowles, 246 Mich. 429, 224 N.W. 387, 388 (1929) (medical doctor’s opinion in rape case that complainant “was a pathological falsifier, a nymphomaniac, and a sexual pervert” admissible for impeachment purposes). Nevertheless, expert opinions as to organic or mental disorders of a witness are admissible for the purpose of challenging the witness’ credibility. Jones v. State, 232 Ga. 762, 208 S.E.2d 850, 853 (1974); see also Mangrum v. State, 227 Ark. 381, 299 S.W.2d 80, 84 (1957); People v. Neely, 228 Cal.App.2d 16, 39 Cal.Rptr. 251, 253 (1964).
The State maintains that the expert opinions aided the jury in understanding the evidence and did not invade the jury’s function of weighing witness credibility. Basically, the State urges that child sexual abuse cases are different and deserve special treatment because jurors may not have special knowledge about the truthfulness of children who claim someone has sexually abused them. The State urges in its brief: “Certainly, a professional child sexual abuse investigator and an elementary school principal are in a better position to evaluate the truthfulness of such allega[96]*96tions that those of us who rarely, if ever, are personally involved in such cases.”
The court in State v. Kim, 64 Haw. 598, 645 P.2d 1330 (1982), essentially adopted the State’s contention. In that case defendant was charged with raping his thirteen-year-old step-daughter and the credibility of the complaining witness was in issue. On appeal, the court held the trial court did not abuse its discretion in allowing a child psychiatrist to opine that the witness’ conduct, along with her mental and emotional conditions, were approximately the same as other child rape victims and that he found the witness’ account to be believable. Id. 645 P.2d at 1333-38. The court concluded that the opinions assisted the jury to understand the evidence and the probative value of the testimony outweighed its prejudicial effects. Id. at 1336.
Recent cases concerning the admissibility of expert evaluation of the complainant as a method of proving forcible rape shed some light on the issue in the present case. In one case, expert testimony that the complaining witness was suffering from post-traumatic stress disorder known as “rape trauma syndrome” was admissible as evidence that a forcible assault did take place. State v. Marks, 231 Kan. 645, 647 P.2d 1292, 1299 (1982). However, a majority of the courts have rejected this reasoning and hold that it is error to admit such opinions. See, e.g., People v. Bledsoe, 36 Cal.3d 236, 203 Cal.Rptr. 450, 460, 681 P.2d 291, 301 (Cal.1984) (expert opinion as to rape trauma syndrome was inadmissible to prove that rape occurred because of unreliability); State v. Saldana, 324 N.W.2d 227, 230-31 (Minn.1982) (error to admit expert opinion on rape trauma syndrome because evidence does not assist in fact-finding function and has not reached a level of reliability that exceeds the common sense evaluation of jury deliberations); State v. Taylor, 663 S.W.2d 235, 241 (Mo.1984) (error to admit expert’s opinion of rape trauma syndrome as it is implied opinion that victim told truth in describing the rape, there is a risk the jury will regard opinion as dispositive on the issue of consent, and the possibility of prejudice outweighs its probative value). In Taylor the Missouri court described the evidence as a direct attack on credibility and an invasion on the function of the jury as follows:
Clearly, the psychiatrist’s specific statement that the victim did not fantasize the rape was an express opinion about her credibility, and his entire testimony that the victim suffered from rape trauma syndrome carried with it an implied opinion that the victim had told the truth in describing the rape. Further, the psychiatrist’s testimony that he was specifically trained to evaluate verbal and nonverbal responses lends a special reliability to his opinion of the victim’s credibility.... The jury was competent to determine the victim’s credibility; therefore, Dr. Amanat’s testimony designed to invest scientific cachet on the critical issue was erroneously admitted. Otherwise, trials could degenerate to a battle of experts expressing opinion on the substance of witness’ veracity.
663 S.W.2d at 241 (citations omitted).
Also helpful are those cases that examine the nature of admissible expert opinions in child sexual abuse cases. Although the California court in Bledsoe held that an opinion that a child suffered rape trauma syndrome was inadmissible to prove a rape occurred, expert testimony on the after effects of rape, such as the emotional and psychological trauma that a victim suffers, may be admissible. 681 P.2d at 301. In a sexual conduct case involving a six-year-old child, the Minnesota court found no abuse of discretion in the admission of expert testimony regarding the characteristics or traits typically observed in sexually abused children. State v. Myers, 359 N.W.2d 604, 608-10 (Minn.1984). In Myers, the expert’s opinion that the child’s allegations were truthful was properly admitted by the trial court because the defendant had waived his right to object to this testimony by “opening the door” in his cross-examination. Id. at 611. The court indicated, however, that it would reject such testimony regarding the truth or falsity of a witness’ allegation [97]*97about a crime because the “expert’s status may lend an unwarranted ‘stamp of scientific legitimacy’ to the allegations.” Id. The Oregon court allowed expert testimony as to the reaction of young children that have been raped by a family member and that the complainant’s behavior was typical of most victims. State v. Middleton, 294 Or. 427, 657 P.2d 1215, 1219-21 (1983). The court then stated: “We expressly hold that in Oregon a witness, expert or otherwise, may not give an opinion on whether he believes a witness is telling the truth.” Id. 657 P.2d at 1221. Likewise, the Washington court allowed an expert to give statistics that supported her opinion that a child’s delay in reporting sexual abuse was not unusual, but held that it was error for the court to allow statistics showing that in 85 to 90 percent of cases children are molested “by someone they already know.” State v. Petrich, 101 Wash.2d 566, 683 P.2d 173, 180 (1984). In Petrich the defendant was the grandfather of the child and the court believed, as to the latter statistics, that the potential for prejudice was significant compared to its minimal probative value. Id. The South Dakota court held that it was reversible error to allow an expert to testify that the child had obtained knowledge about sexual matters through experience with the defendant. State v. Logue, 372 N.W.2d 151, 156-57 (S.D.1985). The Logue court reasoned that although the information may be helpful to the jury to understand the evidence its probative value was substantially outweighed by the prejudicial effect of misleading the jury, and the trial court clearly abused its broad discretion by admitting such testimony. Id. at 157. In summary, it seems that experts will be allowed to express opinions on matters that explain relevant mental and psychological symptoms present in sexually abused children. Unlike the holding in Kim, though, most courts reject expert testimony that either directly or indirectly renders an opinion on the credibility or truthfulness of a witness.
Expert opinion testimony is admissible pursuant to Iowa Rule of Evidence 702 if it “will assist the trier of fact to understand the evidence or to determine a fact in issue.” The ultimate determination of the credibility or truthfulness of a witness is not “a fact in issue,” but a matter to be generally determined solely by the jury. An exception to this would be where the defendant is charged with perjury. Consequently, we conclude that expert opinions as to the truthfulness of a witness is not admissible pursuant to rule 702. As we indicated, the effect of the expert opinions in this case was the same as directly opining on the truthfulness of the complaining witness.
Further reasons support this conclusion. We do not believe that an expert’s testimony that young children do not lie about sexual matters is particularly helpful to the jury. It is within the common knowledge of a juror that these children would have to have some type of sexual experience, instruction or exposure in order to acquire the necessary knowledge to make such accusations and testify accordingly. Indeed, one court noted that: “A young child is unlikely to fabricate a graphic account of sexual activity because such activity is beyond the realm of his or her experience.” State v. McCafferty, 356 N.W.2d 159, 164 (S.D.1984).
We understand and recognize the State’s concern about the sexual abuse of children. These cases are difficult to prosecute because of the age of the victims and the lack of eyewitnesses. Such crimes are indeed detestable and society demands prosecution of these abusers. However, a sexual abuse charge alone carries a large stigma on the accused and conviction provides a serious penalty. In interpreting our Rules of Evidence, we must not only be aware of the needs of society but also the defendant’s right to a fair trial.
A witness is not permitted to express an opinion as to the ultimate fact of the accused’s guilt or innocence. Oppedal, 232 N.W.2d at 524. In this case the trial court admitted expert testimony relating to the truthfulness of the complaining witness. We believe the effect of the opinion [98]*98testimony was to improperly suggest the complainant was telling the truth and, consequently, the defendant was guilty. We conclude the opinion testimony crossed that “fine but essential” line between an “opinion which would be truly helpful to the jury and that which merely conveys a conclusion concerning defendant’s legal guilt.” Horton, 231 N.W.2d at 38. In this instance admission of the expert opinion constitutes error. See Nimmo, 247 N.W.2d at 230; Swartz, 244 N.W.2d at 555; Ogg, 243 N.W.2d at 621; Oppedal, 232 N.W.2d at 524. As noted earlier, adequate objection was made for us to reverse on these grounds.
We hold that the trial court erred in overruling the objection to the testimony of the expert witnesses Davis and Mosher. The court clearly abused its discretion and committed reversible error by admitting opinion testimony prejudicial to defendant. Accordingly, the case is reversed and remanded for a new trial.
REVERSED AND REMANDED.
All Justices concur except HARRIS, WOLLE, and LARSON, JJ., who dissent, and LAVORATO, J., who takes no part.