The Workers’ Compensation Court of Appeals, by majority decision, affirmed a compensation judge’s denial of benefits for permanent partial disability. We affirm.
The facts on which the decision below rests, briefly stated, are these:
The employee had a rectocele and an enterocele which were surgically repaired in September 1982. The employee alleged that while she was lifting at work on October 4, 1983, she suffered an injury in the nature of an enterocele which was visible to her upon self-examination. The employee underwent repair of the enterocele in January 1984. In March 1984, the employee had another non-work injury in the nature of a cystocele.
A rectocele is a bulging of the rectum into the vagina; an enterocele is a bulging of the rectosigmoid colon into the vagina; and a cystocele is a bulging of the bladder or the “cystovaginal wall” into the vagina. These events occur as a result of a thinning of the fascia, and unlike the “common hernia,” no real hole in the fascia develops. While in extreme cases a rectocele and a cystocele can be seen through the opening of the vagina, rarely does an enterocele bulge all the way down so that it is visible at the introitus.
The compensation judge concluded that the employee had failed to prove that her enterocele injury was due to a lifting incident at work because her testimony at the hearing was impeached by prior inconsistent statements concerning (a) the time of day that it occurred, (b) the location in the store where it occurred, (c) and the type of stock she was lifting. In addition, the medical records made shortly after the alleged lifting incident do not contain any indication that the incident had occurred at work. Finally, the employee’s testimony that she found evidence of the enterocele upon self-examination was contradicted by the medical records of her treating gynecologist.1 The decision was, in essence, a credibility evaluation which was affirmed by the Workers’ Compensation Court of Appeals, albeit by majority decision. By writ of certiorari, the employee seeks reversal of the causation determination.
On numerous occasions, this court has stated that assessment of the credibility of a witness is the unique function of the trier of fact. E.g., Tolzmann v. McCombs-Knutson Assocs., 447 N.W.2d 196, 198 (Minn.1989); Even v. Kraft, Inc., 445 N.W.2d 831, 835 (Minn.1989); Tews v. Geo. A. Hormel & Co., 430 N.W.2d 178, 180 (Minn.1988); Brennan v. Joseph G. Brennan, M.D., P.A., 425 N.W.2d 837, 839-40 (Minn.1988).2 In this case, the impeach*681ment of the employee’s testimony concerning how her injury occurred justified the compensation judge’s determination that there was a lack of evidence to sustain the employee’s claim that the injury occurred at work. See Kulenkamp v. Timesavers, Inc., 420 N.W.2d 891, 894 (Minn.1988) (compensation judge’s finding that a worker did not sustain a traumatic hernia reinstated by this court where contemporaneous medical records did not support the claim); Zellmer v. Univac, 290 Minn. 271, 275, 187 N.W.2d 280, 282 (1971) (compensation judge not obligated to accept impeached testimony as true).
Moreover, the compensation judge’s determination had sufficient medical support. Although the employee’s treating gynecologist did state in a letter dated April 16, 1984, that her enterocele repair four months earlier was “related to an incident occurring at her job,” his medical records of her visit shortly after the alleged incident make no reference to a work-related injury, and experienced counsel did not depose the doctor to explore the basis for his opinion. Instead, the employee elected to rely on the testimony of an industrial medicine specialist who is not a gynecologist, and who does not regularly perform surgery outside of necessary “emergency medicine.” The industrial medicine specialist opined that the alleged lifting incident at work created “the symptoms that she had acutely and that those symptoms are related to the development of further pelvic floor relaxation” and the need for surgery. That opinion was based essentially on the employee’s report of the onset of symptoms at work, the continuation of her symptoms, and her ultimate need for surgery in light of his general knowledge of hernias. The doctor had never surgically repaired an enterocele and was unaware of any medical literature linking the development of an enterocele with lifting. The difficulty with both opinions is, of course, that the compensation judge did not believe the employee’s history of a work-related lifting incident on which these doctors’ opinion of causation rested.
The consulting gynecologist retained by the employer believed the alleged work-related injury was not a substantial contributing cause of the employee’s disability. He was of the opinion the employee’s entero-cele was the result of a weak fascial support, that it had developed over a period of time, and that even though lifting of any nature “would not help things,” it was not the cause. The employer’s other gynecological consultant testified that in his 27 years of practice he had never seen a case of an enterocele resulting from a sudden strain such as heavy lifting. Generally, the compensation judge’s choice between experts whose testimony conflicts is upheld when the facts assumed by the chosen expert in forming his opinion are supported by the evidence. Nord, 360 N.W.2d at 342-43. Accordingly, even if the compensation judge’s credibility assessment were suspect, there was sufficient medical evidence to support her ultimate conclusion that the work incident was not a substantial contributing factor in the need for surgical repair some three months later.
The burden rests on the employee to prove by a fair preponderance of the evidence that her injury is compensable. Where two opposing inferences can be drawn with equal justification from the same circumstantial evidence, it cannot be said that one preponderates over the other, and in that event the party having the burden of proof must lose. If different inferences can justifiably be drawn from the evidence in the ease, the inference .drawn by the factfinder will not be disturbed on appeal. Gerhardt v. Welch, 267 Minn. 206, 210, 125 N.W.2d 721, 724 (1964). It is not the function of this court to weigh the evidence to find facts to meet our preference. Hengemuhle v. Long Prairie Jay*682cees, 358 N.W.2d 54, 60 (Minn.1984). The firmly established rule is that “this court will intrude only if, viewing the facts in the light most favorable to the findings, it appears that the findings are manifestly contrary to the evidence or that it is clear reasonable minds would adopt a contrary conclusion.” Id. at 61 (footnote omitted). That this court might have found the facts differently is not the basis for reversal in a case such as this where the facts, viewed in the light most favorable to the findings, amply sustain the determination of the compensation judge and the Workers’ Compensation Court of Appeals.3
Affirmed.
. The employee's doctor found no evidence of an enterocele two days after the alleged work incident. One of the gynecologists retained by the employer testified that an enterocele seldom will be so severe as to protrude as described by the employee. The dissent attempts to explain the absence of evidence of an enterocele or bulge two days after the alleged injury by reference to medical records made weeks after the alleged injury when the enterocele became manifest and was diagnosed by the employee’s treating doctor, speculating that the bulge must have been present as alleged on the date of injury and receded after the employee had rested before her first appointment with her doctor in October 1983. Not even the employee’s principal medical expert advanced such a theory.
. The dissent cites Nord v. City of Cook, 360 N.W.2d 337, 342 (Minn.1985) and Klapperich v. Agape Halfway House, Inc., 281 N.W.2d 675, 680 (Minn.1979) as supporting the proposition that a compensation judge’s credibility evaluation need not be accorded any more deference than any other factual finding. Neither Nord nor Klapperich addressed the evaluation of the cred*681ibility or believability of a witness testifying under oath. Rather Nord and Klapperich both involved the reliability of a medical expert’s opinion premised on factual assumptions that were not supported by evidence adduced at a subsequent evidentiary hearing. The adequacy of the foundation for expert opinion testimony has nothing to do with the credibility of a witness’ recitation of facts on which a claim is based; and neither Nord nor Klapperich apply in the context of credibility assessments as that issue is presented here.
. The dissent apparently believes that the employee’s first surgery in 1982 followed a work-related disability and was therefore, compensa-ble. That the 1982 disability and surgery was not compensable was never in dispute. While the employee’s primary medical expert, the industrial medicine specialist, ' at one time "thought there might be a case for a Gillette type injury prior to 1982,” there was no other support in the record for that proposition; and the employee has never alleged that the 1982 disability and need for surgery arose out of the employment with the employer.