concurring in the result.
I expect that lawyers practicing family law in this state will be lining up to retain Dr. Wilson in child custody cases. Her position that, in the case of equally fit parents, father should be given custody of the boys, and, I think, mother, the girls, has the beneficial potential for doing what, in general, needs to be done in domestic-relations litigation, namely, injecting predictability into custody decisions and removing coerciveness from settlement negotiations. However, it does greater mischief. First, “the data do not provide a sufficient basis for adopting a legal preference for same-sex custody.” Linda Whobrey Rohman et al., The Best Interests of the Child in Custody Disputes, in Psychology and Child Custody Determinations 59, 68 (Lois A. Weithorn ed., 1987) [hereinafter The Best Interests of the Child]. Second, a same-sex preference substitutes one sex-biased rule for another, the tender years doctrine, which the law has abandoned. We ought not subscribe to sex-based rules that are themselves based on sexual stereotypes. But see Moran v. Moran, 200 N.W.2d 263, 269 (N.D.1972) [“Undoubtedly, the court recognizes that as the boys near puberty they will more likely be dependent upon the guidance and discipline of a father rather than a mother.”].
Data suggest that the custody of young children of either sex should be awarded to the parent who has functioned as their primary caretaker. The Best Interests of the Child, supra at 67, 92. I, too, have embraced the idea that the primary caretaker should break the tie between equally fit parents. See, e.g., Gravning v. Gravning, 389 N.W.2d 621, 624-25 (N.D.1986) (Levine, J., dissenting). However, I continue to lose that battle. See, e.g., Dinius v. Dinius, 448 N.W.2d 210 (N.D.1989). Giving the primary caretaker the custodial advantage is based upon the psychological phenomenon of bonding. It is gender neutral, at least in theory, given the fact that either parent may fulfill the role of primary caretaker. Apparently, Dr. Wilson’s opinion is that biology is conclusive.
However unpersuaded I and others are by the merits of Dr. Wilson’s biological determinism, I am wholly persuaded that the trial court did not base its decision on Dr. Wilson’s opinion. While it would have been preferable to remand this case when requested to do so by the appellant, so that the trial judge could have expressly informed us (and reassured me) that she gave the expert opinion the little weight it deserved, I am comfortable with the trial court’s findings of fact in this case. In fact, the trial judge did not find these parents equally fit, making Dr. Wilson’s opinion on biological determinism irrelevant.
Even if there were data that did suggest a particular general- trend, individual cases must be evaluated clinically to determine the applicability of general findings. See The Best Interests of the Child, supra at 84. I find equally troubling the penchant, in the name of “statistical probability,” to blithely slide from the general to the particular without evidence that would either support the general premise or fill in the gap between-the general and the so-called “scientific” conclusion or opinion. Here, we do not know what statistical data underlay “the literature” that Dr. Wilson relied on. In Branson v. Branson, 411 N.W.2d 395, 400 (N.D.1987), the-expert based his custody recommendation on the data that “there is a statistical probability that there could be” a problem someday in the bond between the child and the parent who had been a victim of child abuse. There was, of course, no evidence of a bonding problem in the case or evidence of the “statistical probability” of the major premise. The Branson majority exerted significant energy in minimizing the prejudice resulting from that fallacious and improper expert opinion.
Psychologists testifying in legal proceedings should be advised by counsel and the court to restrict their opinions to matters about which they have requisite knowledge. A psychologist can assist the factfinder by providing it with thorough information. But an expert witness retained by one side must resist the temptation to become a biased advocate rather than an objective expert whose role is to assist the factfinder. As one expert puts it:
“In the determination of appropriate custody, the psychologist is often put into a role *730of evaluating a child in an adversary situation. In this role the psychologist may be requested to only have contact with one of the parents, thus the position of rendering an opinion as to the award of custody to the parent not employing the psychologist has to be suspect.” Douglas Knowlton, Psychology Evaluations of Children: Their Place in the Courtroom, 66 N.D.L.Rev. 673, 676 (1990) (footnote omitted).
Dr. Knowlton cautions that “[p]sychologists should be aware of the impact of their own values” on the opinions they offer and should not exceed their proper function by giving an opinion about the fitness of a parent when the psychologist has had no contact with that parent. Id. “[W]hen attorneys or judges ask the psychologist to provide more information than obtained and push for opinions not directly based on the data available, the appropriateness of the testimony clearly needs to be questioned.” Id. at 685. Here, the specter of impropriety has the appearance of contaminating the trial. That is unfortunate because there is abundant unobjectionable evidence which amply supports the trial court’s findings.
I therefore concur in the result.