Justice VOLLACK
announced the Judgment of the Court and delivered an Opinion
in which Chief Justice ROVIRA and Justice MULLARKEY join.
The defendant Roy Charles Stevens was convicted of two counts of sexual assault on a child, § 18-3-405, 8B C.R.S. (1986). The court of appeals affirmed his conviction. We granted certiorari to review the defendant’s assertion that the trial court erred in holding that hearsay statements admitted pursuant to the child hearsay statute, § 13-25-129, 6A C.R.S. (1987), were sufficiently corroborated to be admissible by the child-victims’ use of age-appropriate sexual terminology, sexual knowledge as demonstrated by their manipulation of anatomically correct dolls, and their behavioral changes. We affirm.
I.
Three children, J.L., E.E., and J.J., stated in July 1985 that the defendant, a daycare teacher hired in January 1985 by La Petite Academy (La Petite), had sexually assaulted them while they were attending the daycare center. J.L. and E.E. were about three and one-half years old; J.J. was almost four years old. The defendant was subsequently charged with three counts of sexual assault on a child.
None of the children testified at trial. The trial court ruled that J.L. and J.J. were incompetent to testify, and that E.E. was competent but unavailable.1 Pursuant to section 13-25-129, 6A C.R.S. (1987), the trial court permitted numerous individuals to testify regarding statements the children made to them implicating the defendant.
A.
We review separately the testimony surrounding the sexual abuse of J.L. and E.E.2
1.
J.L. first attended La Petite on July 22, 1985, and he was assigned to the defendant’s classroom. Shortly afterward, and before J.L. stopped attending La Petite two and one-half days later, J.L.’s mother noticed that J.L. began wetting and defecating in his pants even though he had only infrequent wetting “accidents” since he was toilet-trained more than ten months earlier. J.L. also began having nightmares, expressing fear of using the bathroom, and crying because “he said no one liked him.” J.L.’s mother noticed that although J.L. normally was an outgoing child, “he just all of a sudden at home wasn’t as rambunctious as he usually is.” Moreover, on occasion J.L. would point to men and yell, “I hate him.”
Shortly after his last day at La Petite, J.L. repeatedly stated he did not like “Mr. Steve,” which J.L.’s mother understood was the defendant. J.L.’s mother then asked J.L. in the presence of J.L.’s father, “[H]as anyone ever touched you in your private parts?” J.L. responded, “Mr. Steve put his finger in my bottom, and there was poop on his finger.” J.L. also stated that the defendant touched J.L.’s penis. The next day, J.L. stated that when the defen[949]*949dant put his finger on his “bottom,” the defendant said he had “cung” or “come.” J.L. told his mother and others that he was afraid of the defendant. On July 25, 1985, J.L. repeated his statements regarding the defendant to a doctor performing a physical examination on J.L. In addition, J.L. stated that the defendant had “peed” and it was “white.”
In August 1985, J.L. told Detective Mara Piluras during an interview that “Mr. Steve stuck his finger in his bottom and got poop all over.” Simultaneously, J.L. picked up an anatomically correct doll and “stuck his finger in the doll’s anus.” J.L. also indicated, using the doll, that the defendant had touched his penis.
Carolyn John, a psychiatric social worker who began providing therapy for J.L. in August 1985, testified that in the first session J.L. began playing with a toy kangaroo that had a baby kangaroo in its pouch and stated, “The baby kangaroo was scared, the baby kangaroo needed the diaper because he was afraid to go to the toilet.” During the second therapy session two months later, J.L. drew a dot on paper and said the dot represented a boy; the boy was yelling “help, help, save me, yelling for his mommy and daddy to save him.” During another session in November 1985, John told J.L. to tell her, “and not tell me make believe,” what had happened at school and what the defendant had done to him. Without further prompting, J.L. stated, “He stuck his finger in my bottom and he touched my pee pee.” After Ms. John provided J.L. with an anatomically correct boy doll, J.L. took the doll, pulled its pants down, and put his finger in the doll’s rectum. J.L. also grasped the doll’s penis with his fist. For months during the therapy sessions J.L. continued to talk about the defendant without prompting. John, who was qualified as an expert witness in the fields of social work and psychiatric social work, testified that J.L.’s symptoms were consistent with those of sexual assault victims, especially of a young age.
2.
E.E. began attending La Petite in September 1984. Although E.E. was not in the defendant’s classroom, E.E.’s mother at times left E.E. in the care of the defendant, who was one of the first teachers to arrive at La Petite in the morning. E.E.’s mother noticed during the summer of 1985 that E.E. “suddenly became very different than she had ever previously acted.” E.E. began having frequent inconsolable temper tantrums “for nothing in particular.” Although E.E. had been toilet-trained since the summer of 1984 and had had only two “accidents” since then, in the summer of 1985 E.E. began bed-wetting two to three times a week. On a few occasions, E.E. urinated on herself upon walking into the living room of her home. In addition, E.E. began having nightmares two to three times a week in June and July 1985. E.E. stated that she was afraid to go to the bathroom, and that she wanted her mother to go to the bathroom with her and stay with her while she was in the bathroom.
After E.E.’s mother learned of allegations of child abuse at La Petite, she asked E.E. if anyone had touched her “on your private parts where you go potty.” E.E. stated, “Yes, Mr. Stevens did.” E.E. stated that the defendant pulled down her panties. Subsequently E.E. indicated that the defendant had squeezed her buttocks and touched her in the vaginal area. E.E. stated that the defendant committed the acts while other children were playing outside and the defendant took her inside the daycare for a “timeout.” After E.E.’s first therapy session, E.E. told her mother without prompting, “Mr. Stevens rubbed his peanut in my face and peed on me.” E.E. also stated that she had to wipe her face off with paper towels. In January 1986, when E.E.’s mother was enrolling E.E. in another daycare center, E.E. said, “I don’t want a man teacher.”
In August 1985, Detective Piluras interviewed E.E. using anatomically correct dolls. While E.E. was identifying anatomical parts of a girl doll, E.E. volunteered in a matter-of-fact tone, “You know what, Mr. Stevens pulled my panties down.” When Piluras asked if anyone had ever touched her “body,” E.E.’s word for “vagina,” E.E. [950]*950stated that “Mr. Stevens” had with his hands. When Piluras asked if anything had been placed inside her, E.E. said the defendant had “put a bomb in her.”
In September 1985, E.E. began receiving therapy from Pat McKee, a social worker. McKee testified that E.E., who was verbal and “precocious,” was age-appropriate developmentally and behaviorally and could distinguish between reality, fantasy and lying. Over the course of several sessions, E.E. identified an anatomically correct doll as “Mr. Stevens” and stated that he did “bad things” to her. E.E. also stated that the defendant’s penis got hard, that the penis felt big and too big for her mouth, that the defendant “pee’d in my face,” and that the “pee” was “sticky and icky.” McKee stated that E.E. could indicate where the defendant touched her and that sometimes it “kind of tickled” and other times it hurt. McKee stated that while E.E. was trying to describe the incident, “[i]t was obvious she didn’t have all the words for all the things she experienced” at the time. McKee also testified, “This is a child who would be particularly difficult to coach. She has a mind of her own.” McKee, who was qualified as an expert witness in the fields of social work and applied psychotherapy, testified that E.E.’s statements were consistent with those of sexually abused children, and that E.E.’s atypical behavior during summer of 1985 was consistent with children who have experienced trauma.
B.
The jury found the defendant guilty of two counts of sexual assault on a child, and the defendant was sentenced to two concurrent terms of six years imprisonment.
On appeal, the defendant argued that the trial court erred in admitting the hearsay statements of J.L., E.E., and J.J. because the sexual assaults were not corroborated as required by subsection 13-25-129(l)(b)(II), 6A C.R.S. (1987). The court of appeals affirmed, holding that the record contained sufficient corroboration of the sexual assaults.
II.
The issue is whether J.L.’s and E.E.’s age-appropriate sexual terminology, manipulation of anatomically correct dolls, and behavioral changes constitute “corroborative evidence” of the sexual abuse of the children, as required by section 13-25-129, 6A C.R.S. (1987).
Section 13-25-129 provides in relevant part:
(1) An out-of-court statement made by a child ... describing any act of sexual contact, intrusion, or penetration ... performed with, by, on, or in the presence of the child declarant, not otherwise admissible by a statute or court rule which provides an exception to the objection of hearsay, is admissible in evidence in any criminal ... proceedings in which a child is a victim of an unlawful sexual offense ... if:
(a) The court finds in a hearing conducted outside the presence of the jury that the time, content, and circumstances of the statement provide sufficient safeguards of reliability; and
(b) The child either:
(1) Testifies at the proceedings; or
(II) Is unavailable as a witness and
there is corroborative evidence of the act which is the subject of the statement.
(2) If a statement is admitted pursuant to this section, the court shall instruct the jury that it is for the jury to determine the weight and credit to be given the statement and that, in making the determination, it shall consider the age and maturity of the child, the nature of the statement, the circumstances under which the statement was made, and any other relevant factor.
Section 13-25-129, similar to child hearsay statutes in other states,3 permits courts to [951]*951admit hearsay statements of child declar-ants under specific guidelines.
The enactment of section 13-25-129 represents the legislature’s attempt to ensure that child abusers do not go free merely because the prosecutor is unable to obtain witnesses to the abuse other than the child who is unable to testify about the abuse. See People v. Diefenderfer, 784 P.2d 741, 746-51 (Colo.1989); see also State v. Bellotti, 383 N.W.2d 308, 316 (Minn.Ct. App.1986) (in enacting child hearsay statute, Minnesota legislature “did not want to allow child abusers to escape conviction merely by choosing victims” who are unavailable to testify because of their age or otherwise), review denied (1986); Note, The Testimony of Child Victims in Sex Abuse Prosecutions: Two Legislative Innovations, 98 Harv.L.Rev. 807, 808, 817 (1985) (child hearsay exception statutes attempt to remedy “unusual difficulties” of prosecuting crimes against children).4
Under the terms of section 13-25-129, a child’s hearsay statements are admissible only after the trial court makes two findings. The court first must find that the time, content, and circumstances of the child’s hearsay statement provide sufficient safeguards of reliability. If the child is not available as a witness, the court must also find there is corroborative evidence of the sexual abuse that is the subject of the statement.
Section 13-25-129 does not define “corroboration,” nor does the Washington child hearsay statute, upon which Colorado’s statute was based. See Wash.Rev.Code § 9A.44.120 (1989); Note, supra, at 811. The rules of statutory construction require that we assign the technical or particular meaning to words and phrases that have acquired such a technical or particular meaning. § 2-4-101, IB C.R.S. (1987). “Corroboration” has long been used in the context of evidentiary law, see, e.g., People v. Garcia, 789 P.2d 190 (Colo.1990), and section 13-25-129 is found in the statutory provisions concerning evidence, see title 13, article 25 (titled “Evidence—General Provisions”). Accordingly, we find it appropriate to define “corroboration” in the context of evidentiary law. See N. Singer, 2A Statutes and Statutory Construction §§ 46.05, 47.29 (Sands 4th rev. ed. 1984).
In that context, we have approved a definition of “corroborative evidence” that comprises evidence, direct or by proof of surrounding facts and circumstances, that tends to establish the fact sought to be proved. See Davis v. People, 176 Colo. 378, 382, 490 P.2d 948, 950 (1971); accord State v. Swan, 114 Wash.2d 613, 622, 790 P.2d 610, 615 (1990); State v. Hunt, 48 Wash.App. 840, 849-50, 741 P.2d 566, 571-72, review denied, 109 Wash.2d 1014 (1987); State v. Allen, 157 Ariz. 165, 176, 755 P.2d 1153, 1164 (1988); Black’s Law Dictionary 311 (5th ed. 1979).
As we have construed section 13-25-129’s “corroborative evidence” requirement, the trial court is granted broad dis[952]*952cretion to determine whether it should consider proffered corroborative evidence, and to determine how much weight it should give the evidence. In this sense, the corroboration requirement is similar to our evidentiary rules’ minimal requirements for admission of any “relevant evidence.” See CRE 401 & 402 (all relevant evidence, defined as evidence having “any tendency to make the existence of any fact” more or less probable than it would be without the evidence, is admissible except as otherwise provided). The discretion afforded trial courts in determining what evidence is “corroborative” is justified and necessitated because sexual abuse frequently is committed “ ‘with the greatest possible stealth and secrecy.’ ” People v. District Court of El Paso County, 776 P.2d 1083, 1085 n. 1 (Colo.1989) (quoting State v. Smith, 16 Utah 2d 374, 376, 401 P.2d 445, 447 (1965)). Most sexual-abuse crimes consist of petting, exhibitionism, fondling, and oral copulation, Comment, supra note 4, at 1750, and children rarely resist their attackers. Id. Thus, seldom can acts of sexual abuse be physically corroborated, and usually the only evidence available directly implicating the attacker is the child’s hearsay statements. See District Court of El Paso County, 776 P.2d at 1085 n. 1. Because of the difficulty in obtaining corroborative evidence, we recognize that a “ ‘stubborn insistence on corroboration that is impossible to obtain’ ” should not defeat the essential purposes of the child hearsay statute. Swan, 114 Wash.2d at 641, 790 P.2d at 624 (quoting Jones, 112 Wash.2d at 493, 772 P.2d at 500); see also Nelson v. Farrey, 874 F.2d 1222, 1230 (7th Cir.1989), cert. denied, — U.S. -, 110 S.Ct. 835, 107 L.Ed.2d 831 (1990) (“[H]ow ironic it would be if the child molester could use the trauma inflicted on his victim as the fulcrum for levering his way to freedom.”).
We hold that “corroborative evidence,” as contemplated by section 13-25-129, is any evidence, direct or by proof of surrounding facts and circumstances, that tends to establish the act described by the child in the statement occurred.
Section 13-25-129’s requirement of “corroboration” of the alleged sexual abuse should not be confused with the statute’s separate requirement of “safeguards of reliability.” 5 Indicia of reliability and guarantees of trustworthiness are constitutionally required prior to the admission of hearsay statements to preserve the sixth amendment’s basic interest in requiring “confrontation” — even though an accused cannot directly confront the hearsay declarant. See Idaho v. Wright, — U.S. -, 110 S.Ct. 3139, 111 L.Ed.2d 638 (1990). The corroboration requirement is best considered not as a traditional index or guarantee of a statement’s reliability, but as a method of objectively establishing the commission of a sexually abusive act.
C.
Because the defendant argues that any corroborative evidence properly considered by the trial court pursuant to section 13-25-129 was insufficient under the statute to support admission of the children’s hearsay statements, the final subsidiary question we must address before we reach the substance of the defendant’s allegations of error is what quantum of corroborative evidence the prosecution must produce to warrant admission into evidence of a child’s hearsay statements. At the outset we note that this court has held that section 13-25-129 is more restrictive of admission of hearsay statements than is the federal Constitution. Diefenderfer, 784 P.2d at 748. When a hearsay declarant is unavailable to testify at trial, the federal Constitution requires only that admission of the hearsay statements be preceded by a showing that the statements contain particularized guarantees of trustworthiness or indicia of reliability. Wright, 110 S.Ct. at 3146 (citing Ohio v. Roberts, 448 U.S. 56, 65, 100 S.Ct. 2531, 2538, 65 L.Ed.2d 597 (1980)). Section 13-25-129 requires not only that a child’s hearsay statements concerning sexual [953]*953abuse contain indicia of reliability, but also that the sexual abuse be corroborated. Unlike Ohio v. Roberts ’ minimum confrontation-clause standards for admission of hearsay statements, section 13-25-129’s corroboration requirement is not a constitutional prerequisite. See Roberts, 448 U.S. at 65-66, 100 S.Ct. at 2538-2539; Graham, Indicia of Reliability and Face to Face Confrontation: Emerging Issues in Child Sexual Abuse Prosecutions, 40 U. of Miami L.Rev. 19, 58 (1985); Comment, supra note 4, at 1766; but see Note, supra, at 812, 819-22.
The corroboration requirement, however, admittedly serves the important role of ensuring that the sexual abuse alleged by the child actually occurred and, at least indirectly, safeguarding against wrongful conviction. See Swan, 114 Wash.2d at 622, 790 P.2d at 615; Jones, 112 Wash.2d at 493-495, 772 P.2d at 499; Note, supra, at 820; see also Hunt, 48 Wash.App. at 847-849, 741 P.2d at 571 (corroboration requirement protects against fabricated or imagined allegations that defendant cannot test with cross-examination). Section 13-25-129’s corroboration requirement is more accurately said to reflect the legislature’s concern that in the typically emotionally charged child sexual abuse cases “the trier of fact may be too willing to convict an accused sexual offender ... on the basis of evidence of alleged out of court statements of children that describe socially repugnant sexual contact.” Graham, supra, at 58; accord Hearings on S.B. 11 Before the Senate Comm. on the Judiciary, 54th Gen. Assembly, 1st Session, Audio Tape No. 83-3, at hr. 14, min. 41 (Jan. 19, 1983) (statement of Senator Wells); District Court of El Paso County, 776 P.2d at 1085 n. 1; Jones, 112 Wash.2d at 493-495, 772 P.2d at 499 (child’s allegations of sexual abuse can have powerful emotional impact on jury); Bulkley, supra note 3, at 36.
Accordingly, the quantum of evidence required must reflect the legislature’s concern that convictions for child sexual abuse must not be based on child hearsay admitted into evidence solely on facts that only tepidly suggest the existence of sexual abuse. Rather, the quantum of corroborative evidence needed to support admission of a child’s hearsay statement must be enough to induce a person of ordinary prudence and caution conscientiously to entertain a reasonable belief that the sexual abuse that is the subject of the child’s hearsay statement occurred. See People v. Ayala, 770 P.2d 1265, 1267 (Colo.1989); cf State v. Spronk, 379 N.W.2d 312, 314 (S.D.1985) (child hearsay statute’s corroboration requirement satisfied by evidence that “in some substantial degree” tends to affirm that alleged act of sexual abuse occurred); Hunt, 48 Wash. App. at 849-850, 741 P.2d at 571-72 (child hearsay statute’s corroboration requirement satisfied by “evidence of sufficient circumstances” that would support logical and reasonable inference that alleged act of abuse occurred). The evidence need not be sufficient to support a conviction or send the case to the jury. See Hunt, 48 Wash.App. at 848-849, 741 P.2d at 571. This quantum-of-evidence standard is consistent with the proper balance we have struck between “protecting the defendant’s right to confront an adverse witness and the prosecution’s right to present reliable, otherwise unobtainable evidence to the jury.” District Court of El Paso County, 776 P._2d at 1090.
Moreover, the trial court may examine the evidence as a whole to determine whether sufficient corroboration exists to warrant admission of child hearsay statements under section 13-25-129. Diefenderfer, 784 P.2d at 751; accord Swan, 114 Wash.2d at 641, 790 P.2d at 624; see also In the Interest of O.E.P., 654 P.2d 312, 319 (Colo.1982) (totality of circumstances underlying child’s hearsay statements demonstrated that trial court did not err in admitting statements under excited utterance hearsay exception); State v. Sorenson, 143 Wis.2d 226, 246, 421 N.W.2d 77, 85 (1988) (court must evaluate totality of factors in residual hearsay exception to determine whether hearsay statement should be admitted).
[954]*954III.
We now reach the question of whether the corroborative evidence adduced in this case before the trial court meets the requirements we have set forth above. The defendant objects to the validity of three types of evidence that the court of appeals, pursuant to section 13-25-129, 6A C.R.S. (1987), found corroborative of the sexual abuse of the children: age-appropriate sexual terminology, manipulation of anatomically correct dolls, and behavioral changes. The defendant argues that the evidence did not conform to the requirements of section 13-25-129, and that the trial court’s use of the evidence to support the admission of the children’s hearsay statements violated his sixth amendment right to confrontation. We do not agree.
At the outset, we must identify the appropriate standard of review for a challenge to a trial court’s ruling on the existence and sufficiency of corroborative evidence under section 13-25-129. The trial court’s ruling on the existence and sufficiency of corroborative evidence under the statute necessarily comprises mixed questions of fact and law, and as such the trial court enjoys broad discretion to determine whether the corroborative evidence may support the admission of hearsay statements under the statute. Cf. Pigford v. People, 197 Colo. 358, 361, 593 P.2d 354, 357 (Colo.1979) (trial court has substantial discretion on whether to admit evidence of prior criminal transaction). Thus, we will not overturn a trial court’s ruling on corroborative evidence brought pursuant to subsection 13-25-129(l)(b)(II) unless the court has abused its discretion. See People v. District Court of El Paso County, 776 P.2d 1083, 1090 (Colo.1989); People v. Gillispie, 767 P.2d 778, 781 (Colo.App.1988), cert. denied (1989); People v. Hise, 738 P.2d 13, 16 (Colo.App.1986), cert. denied (1987).6 Courts in other states that have enacted child hearsay statutes almost identical to section 13-25-129 also apply the abuse-of-discretion standard. See, e.g., State v. Petry, 524 N.E.2d 1293, 1296 (Ind. Ct.App.1988); State v. Swan, 114 Wash.2d 613, 665-67, 790 P.2d 610, 637-38 (1990).
The court of appeals’ only reference to age-appropriate sexual terminology occurred in its discussion of the corroborative evidence it found in the record. The court of appeals stated, “Also, each child gave similar reports, in a language appropriate to their age, to the several adults who had interviewed them.” People v. Stevens, No. 87CA0320, slip op. at 3 (Colo.App. Jan. 12, 1989) (unpublished opinion) (emphasis supplied).
Courts have used the phrase “age-appropriate sexual terminology,” and similar phrases, to refer to a child’s use of words to describe human genitalia and sexual functions that would be consistent with words used by children of a similar age. See, e.g., People v. Diefenderfer, 784 P.2d 741, 744, 748 (Colo.1989); District Court of El Paso County, 776 P.2d at 1084-85, 1089; In the Interest of O.E.P., 654 P.2d 312, 314, 318 (Colo.1982); accord, e.g., United States v. Nick, 604 F.2d 1199, 1201, 1204 (9th Cir.1979); State v. Robinson, 153 Ariz. 191, 204, 735 P.2d 801, 814 (1987); State v. Bellotti, 383 N.W.2d 308, 310-11, 313 (Minn.Ct.App.1986), review denied (1986). The rationale underlying scrutiny of a child’s “sexual terminology” is that if a child’s sexual terminology were more sophisticated than the terminology of children of similar age, the child’s statements may have been the product of “coaching” or other improper influence, and therefore unreliable. See, e.g., District Court of El Paso County, 776 P.2d at 1089; cf. State v. Sorenson, 143 Wis.2d 226, 246, 421 N.W.2d 77, 85 (1988). On the other hand, a child’s use of sexual terminology appropriate for her age to describe the sexual abuse perpetrated on her is considered among the [955]*955factors indicating the statement’s reliability. See, e.g., District Court of El Paso County, 776 P.2d at 1089.
Age-appropriate sexual terminology plainly cannot serve to corroborate in this case because the particular words the children used for human genitalia or sexual functions have no probative value in determining whether or not a sexually abusive act occurred. J.L.’s use of the words “my bottom,” “poop,” and “peed” are entirely non-probative on the issue of the existence of sexual abuse. Similarly, E.E.’s use of the words “peanut,” “peed,” “body,” and “sticky and icky” also are non-probative on the issue of the existence of sexual abuse. No amount of scrutiny of the children’s words themselves can make the existence of sexual abuse any more or less likely.7
We hold that the age-appropriate sexual words the children used in this case cannot supply corroborative evidence of a sexually abusive act, and the court of appeals erred to the extent it held otherwise.
The defendant argues that the dolls evidence is inherently unreliable and that the dolls evidence should not have been used as corroborative evidence under section 13-25—129(1)(b)(II).8 The answer to the defendant’s argument is that section 13-25-129’s corroboration requirement is not concerned with reliability except to the extent that the rules of evidence governing admissibility demand it — that is, corroborative evidence must bear some indicia of reliability or it would be irrelevant as having no probative value. We cannot conclude that anatomically correct dolls evidence is inherently unreliable. See State v. Fletcher, 322 N.C. 415, 421, 368 S.E.2d 633, 636 (1988).
Nothing in the record suggests that any of the children’s therapists or Detective Piluras, all of whom observed the children manipulating the dolls, acted in such a way as to render the dolls evidence unreliable. In particular, the therapists’ and Detective Piluras’s testimony indicated that they asked no leading questions, and allowed the children to freely use the dolls. The children identified on their own the dolls as themselves or the defendant, and the children on their own demonstrated with the dolls the sexual acts about which they had narrated to various individuals. Moreover, the children’s demonstrations with the dolls accurately reflected their statements concerning the sexual abuse that were given prior to their demonstrations. We conclude that the dolls evidence contained sufficient indicia of reliability to satisfy the child hearsay statute.
The defendant next argues that a child’s behavioral changes are inconclusive on the issue of whether the child has suffered sexual abuse. As we have stated, a corroborating fact need not conclusively establish the fact of sexual abuse; it need only tend to establish that the sexual abuse described by the child occurred.
We of course recognize that not all suspicious behavioral changes of a child are attributable to sexual abuse. Many symptoms attributed to sexual abuse may actually be caused by other events, such as intra-family conflict or school problems. See generally J. Conte, A Look at Sexual Abuse 24-26 (1986); Note, The Unreliability of Expert Testimony on the Typical Characteristics of Sexual Abuse Victims [hereinafter “Expert Testimony ”], 74 Geo. L.Rev. 429, 439-43. We also recognize that social science is an inexact science that [956]*956attempts to understand highly complex behavioral patterns. McCleskey v. Kemp, 753 F.2d 877, 887 (11th Cir.1985), aff'd on other grounds, 481 U.S. 279, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987). Social scientists are still collecting and evaluating data concerning symptoms exhibited by sexual abuse victims, and only recently have social scientists attempted to organize their observations of the psychological impact of sexual abuse on children. C. Walker, B. Bonner & K. Kaufman, The Physically and Sexually Abused Child: Evaluation and Treatment 115, 142 (1988).
Nonetheless, social scientists have identified certain behavioral changes in children that frequently are associated with sexual abuse. See generally Sgroi, Handbook of Clinical Intervention in Child Sexual Abuse 39-79 (1982); Walker, Bonner & Kaufman, supra, at 120-23 (from infancy to age four, symptoms of sexually abused child include loss of toilet training, sleep disturbances, fear of men, excessive clinging, and "sex talk”); Conte, supra, at 24-25; Expert Testimony, supra, at 442 n. 94. In particular, psychologists have identified a narrow range of behavior — including sexual knowledge that is new or atypical for a given child, physical trauma to the genitals, and inappropriate sexual behavior — that is rarely caused by any event other than sexual abuse. Conte, supra, at 24-26; see Sgroi, supra, at 39-79.
The inability of the social sciences to establish an exclusive cause-effect relationship between sexual abuse and certain behavioral changes in children does not require the prohibition of evidence concerning behavioral changes. It is enough in the context of section 13-25-129 that psychologists generally have concluded — and courts have adopted their conclusions9 — that sexually abused children frequently exhibit certain behavior following their sexual abuse. The inconclusiveness of these symptoms on the question of sexual abuse merely emphasizes the importance of the trial court’s examination of the evidence as a whole.
In this case, J.L. and E.E. experienced similar, abrupt behavioral changes that included bedwetting and other urination — which was inappropriate because both J.L. and E.E. had been toilet-trained— and temper tantrums. J.L. and E.E. both experienced nightmares, expressed fears of going to and using the bathroom, and expressed a fear of men. While they are not conclusive on the issue of whether sexual abuse occurred, these behavioral changes constitute corroborative evidence of the existence of the sexual abuse of which the children complained.10
IV.
The trial court did not abuse its discretion in admitting into evidence the children’s hearsay statements based on the [957]*957corroborative evidence of the children’s behavioral changes. Moreover, we find no abuse of discretion in the trial court’s conclusion that the corroborative evidence adduced was sufficient to satisfy section 13-25-129’s corroboration requirement.
Judgment affirmed.
ERICKSON and KIRSHBAUM, JJ., join in the result only.
LOHR, J., specially concurs, and QUINN, J., joins in the special concurrence.