OPINION
FOLEY, Judge.Allen D. Jorschumb appeals from a December 9, 1985 judgment awarding respondent Jackie L. Jorschumb permanent custody of the couple’s only child and from the denial of his motion for a new trial. He contends on appeal that the custody award constitutes an abuse of discretion under Pikula v. Pikula, 374 N.W.2d 705 (Minn.1985), and that respondent’s failure to comply with a discovery request for psychological results rendered the testing psychologist’s trial testimony inadmissible. We affirm.
FACTS
This case arises out of the parties five-year marriage and centers on the custody of their son, J.A., who was four years old at the time of trial. For the duration of the marriage, the couple lived on a farm with respondent’s 12-year-old son from a former marriage. On August 10, 1984, a hearing for temporary relief was held.
Both parties testified at the temporary hearing that they were better suited to provide for J.A.’s daily needs and functioned as the child’s primary caretaker. Respondent’s parents stated that both parties were loving parents but believed that appellant would provide a more stable environment for J.A. The trial court awarded the couple joint temporary custody with each receiving custody Zlh days per week. The court further ordered preparation of a child custody study in the event review of the custody situation was necessary.
A final custody hearing took place on September 26, 1985. During this intervening period, both parties received family counseling from Dr. John Husted, a licensed psychologist. After 15-20 sessions, appellant ceased contact with Dr. Husted. Respondent remained in Dr. Husted’s care at the time of trial. A custody report prepared by the court-appointed social worker, Beth Palm, was filed on January 2, 1985. Palm was unable to make a specific custody recommendation without the benefit of psychological reports which both parties withheld. Palm did note however, that based on the information provided by the parties, appellant would likely provide the most stable home environment.
Dr. Husted performed MMPI tests on both parties. During discovery, appellant sought access to respondent’s medical records, including the MMPI results. The request was denied on the grounds that part of the material was privileged. Appellant’s motion to compel discovery was deferred for consideration by the trial judge assigned to the case. Appellant did not renew his motion.
At the final custody hearing on September 26, 1985, respondent testified that she was regularly left alone with the children while appellant attended to farming duties and that she assumed responsibility for bathing, toilet-training, feeding and playing with J.A. She further claimed that her *809attempts to discipline J.A. were often undermined by appellant.
Appellant claimed that he was primarily responsible for grooming, disciplining and toilet-training J.A., and that respondent often acted irresponsibly by refusing to get out of bed when J.A. awoke in the morning and by frequently leaving him in the care of grandparents or his 12-year-old stepbrother. He also testified that respondent’s inability to control her temper and her need to seek psychological counseling provided an unstable environment for J.A.
Dr. Husted testified for respondent based on 40-50 counseling sessions with her individually and six with the children present. Both children were observed to exhibit a warm, open relationship with respondent and with each other. In Dr. Husted’s opinion, respondent satisfied the emotional needs of her children and had an excellent prognosis for meeting these needs in the future. When asked about respondent’s MMPI results, Dr. Husted responded, without objection, that respondent was in excellent mental health and had made great strides in becoming emotionally self-sufficient. Both parties’ MMPI results were admitted into evidence without objection. Dr. Husted ultimately concluded:
I think at this point that the boy’s best interests would be to stay — to have the primary relationship being with his mother at this time. It seems to me that she has shown sufficient stability and sufficient capacity to nurture that, at least for the foreseeable future this would probably be the most effective relationship for him and for her.
Margaret Ostergaard, an intern therapist, testified for appellant. Her testimony was based in part on the results of a neighborhood survey from a list of names provided by appellant. Some of those who responded testified at trial. In general, they agreed that J.A. had a “special relationship” with appellant, a point respondent agreed was accurate. According to Oster-gaard, the survey indicated that the older child was often observed babysitting for J.A. for long periods of time. Ostergaard further stated that her interviews with appellant strongly indicated his desire and ability to meet J.A.’s daily needs. Oster-gaard concluded:
I need the court to understand that I have not interviewed Mrs. Jorschumb, that my evidence would be on the survey and reading of the transcripts, and I undoubtedly would state that Mr. Jor-schumb would be the more fit parent * *.
In awarding permanent custody of [J.A.] to respondent, the court found:
7. Split custody has been tried for the past year, but because of the parents’ inability to relate in an amicable manner, this is no longer a feasible alternative.
8. The evidence is mixed as to whether Mr. Jorschumb or Mrs. Jorschumb has been the child’s primary caretaker, but taking all evidence together, including the expert testimony of Dr. John Husted, a licensed consulting psychologist who has counselled both parties, the greater weight of credible evidence favors a finding that Mrs. Jorschumb has demonstrated the normal indicia of primary parenthood and that the best interest of [J.A.] would indicate that sole custody should be granted to her for the foreseeable future. (Emphasis supplied.)
In response to appellant’s post-trial motion for reconsideration under Pikula v. Pikula, 374 N.W.2d 705 (Minn.1985), the trial court amplified its earlier findings to reflect that a “unique co-parenting” arrangement existed at the time of the temporary hearing and that neither party functioned solely as primary parent. This appeal followed.
ISSUES
1. Did the trial court properly exercise its discretion when it awarded respondent permanent custody of the couples’ only child?
2. Did the trial court err in allowing a psychologist’s testimony when it was based in part on an undisclosed psychological test performed on respondent?
*810ANALYSIS
Appellate review of a custody award is limited to whether the trial court abused its discretion by making findings unsupported by the evidence or by improperly applying the law. Pikula v. Pikula, 374 N.W.2d 705, 710 (Minn.1985); Ozenna v. Parmelee, 377 N.W.2d 483, 488 (Minn.Ct.App.1985). A trial court’s findings will be upheld unless clearly erroneous. Minn.R.Civ.P. 52.-01.
The Primary Parent Doctrine
The principle underlying all child custody determinations is the best interests of the child. Minn.Stat. § 518.17, subd. 3 (1984). Relevant factors to be considered by a trial court in assessing the “best interests of a child” are set out in Minn.Stat. § 518.17, subd. 1 (1984).1
In Pikula, the supreme court held that consideration of the “best interest” factors in Minn.Stat. § 518.17, subd. 1, requires:
[W]hen both parents seek custody of a child too young to express a preference for a particular parent and one parent has been the primary caretaker, custody be awarded to the primary parent absent a showing that that parent is unfit to be the custodian.
"The best interests of the child" means all relevant factors to be considered and evaluated by the court including:
(a) The wishes of the child’s parent or parents as to his custody;
(b) The reasonable preference of the child, if the court deems the child to be of sufficient age to express preference;
(c) The interaction and interrelationship of the child with his parent or parents, his siblings, and any other person who may significantly affect the child’s best interests;
(d) The child's adjustment to his home, school and community;
(e) The length of time the child has lived in a stable, satisfactory environment and the desirability of maintaining continuity;
(f) The permanence, as a family unit, of the existing or proposed custodial home;
(g) The mental and physical health of all individuals involved;
(h) The capacity and disposition of the parties to give the child love, affection, and guidance, and to continue educating and raising the child in his culture and religion or creed, if any; and
(i) The child’s cultural background.
Pikula, 374 N.W.2d at 713.2
The Pikula court ultimately remanded the matter for a determination “of which, if either, parent was the primary caretaker of the children at the time the dissolution proceeding was commenced.”3 Id. at 714 (emphasis supplied). Here, the trial court expressly found that neither party functioned solely as the child’s primary caretaker at the time of separation:
2. Evidence presented at the initial hearing revealed that each parent had a loving, close relationship with [J.A.] (then three years old). There was evidence that [respondent] disciplined [J.A.], bought food and clothes for him, took him to physicians for medical care, took him on outings, and spent about 12 hours a day with him daily. There was also evidence that [appellant] spent 2 to 3 *811hours a day with him, played with him, etc. The evidence demonstrated that neither parent could clearly be called the primary parent, but that both functioned in that capacity. (Emphasis supplied.)
Appellant contends and the dissent agrees that the evidence overwhelmingly supports that he alone functioned as the child’s primary caretaker at the time of separation and that this point was conceded by Dr. Husted on cross-examination:
Q: Dr. Husted, if evidence was introduced demonstrating that Mr. Jor-schumb fed the children on a regular basis before the separation, * * * that he was the primary keeper of the children, he would wake them, he would change the diapers, he would feed them, and he would put them to sleep and take care of them, * * * would you frame a different — form a different opinion?
A: No, that would be consistent with my perception of the marriage and family as I first knew them. (Emphasis supplied.)
We do not interpret Dr. Husted’s response as his opinion that appellant was the primary parent in the child’s life at the time of separation. The question was phrased hypothetically and did nothing more than frame the central issues. The evidence at trial established that the parties often performed different nurturing functions. Simply because appellant ordinarily assumed primary responsibility for the child’s toilet training, bedtime preparation and discipline, does not imply that he alone functioned as the primary parent. The indicia of parenthood factors adopted in Pikula are not prioritized and are not exclusive. See Pikula, 374 N.W.2d at 713.
The dissent focuses much of its attention on the testimony of respondent’s parents in arriving at its conclusion that appellant was the primary parent at the time dissolution proceedings were commenced. While damaging to respondent’s custody claim, the Hasbargens’ testimony was undeniably an issue of credibility for the trial court. See Minn.R.Civ.P. 52.01. At the temporary hearing, Chester Hasbar-gen did state that Allen would be the better parent. When asked to explain why this was so, Mr. Hasbargen stated:
Well, on a farm is a better place to bring up and raise a child — children. They have more, ah, place to play with their stuff, and if [Jackie] is going to be in an apartment, and she has — I hear that she has an upstairs apartment, well there’s no place — the children will have to be mostly in the house all the time, and if there out, then they’ll be on the streets.
Chester Hasbargen has been a farmer all his adult life. While it is easy to understand why he believes the environment is advantageous to raising children, his testimony also reflects a personal bias. Kennedy v. Kennedy, 376 N.W.2d 702 (Minn.Ct.App.1985), cautions that an award of property should not be determinative of custody. The same basic principles apply here.
Lois Hasbargen’s testimony was based on her admission that she observed Jackie, Allen and the child together 1-2 times per month plus holidays.
Q. Now, just a minute. I’m just asking in what situations do you see them? A. Well, mostly on—
Q. You told me—
A. Well, I have driven out there.
Q. But how often in a month? A normal month?
A. Oh, ah, once or twice. I don’t go out there very often.
It is also apparent from the record that the relationship between Jackie and her mother was often marked with tension. The trial court could properly account for all these factors in reaching its decision.
When viewed as a whole, the parents’ testimony merely reinforces the trial court’s finding that the parties assumed different parental functions. The dissent also places heavy emphasis on a portion of Lois Hasbargen’s testimony at the temporary custody hearing reflecting her daughter’s instability and immaturity. However, there is nothing in the record to suggest *812that this in any way detrimentally affected the relationship between mother and child.
The dissent also asserts throughout its discussion that the trial court failed to focus on the proper period of time in making the primary parent determination. This is plainly resolved in the trial court’s post-trial memorandum:
[T]here is no need for a new trial to determine who was the primary caretaker at the time of the initial proceeding because the facts in this case showed that there was a unique co-parenting situation at that time. (Emphasis supplied.)
The close relationship between father and son is to be applauded but it should not be emphasized to the total exclusion of the nuturing activities and love provided by respondent. If the trial court had determined that respondent functioned as the primary parent under the facts, it would be evident that the law under Pikula was misapplied. A similar result would hold true if appellant was declared the sole primary parent.
The trial court was in a better position to assess the credibility of the witnesses and this court will not substitute its judgment. See Minn.R.Civ.P. 52.01. Its finding that neither party functioned exclusively as the primary parent at the time of the temporary hearing is not clearly erroneous. Id4 See Pekarek v. Pekarek, 384 N.W.2d 493 (Minn.Ct.App.1986); Sydnes v. Sydnes, 388 N.W.2d 3 (Minn.Ct.App.1986).
Since neither party functioned solely as primary parent at the time of separation, the trial court was obliged to consider “other indicia of parental fitness” in reaching its custody decision. As stated in Pikula:
When the facts demonstrate that responsibility for and performance of child care was shared by both parents in an entirely equal way, then no preference arises and the court must limit its inquiry to other indicia of parental fitness.
Pikula, 374 N.W.2d at 713-14 (emphasis supplied).
“Best Interest” Factors
In custody matters, it is well established that a trial court must make written findings that reflect proper consideration of the factors set forth in Minn.Stat. § 518.17, subd. 1. Rosenfeld v. Rosenfeld, 311 Minn. 76, 82, 249 N.W.2d 168, 171 (1976). The trial court’s findings reflect proper consideration of the statutory criteria:
4. J.A. and [his step-brother] have a close sibling relationship * * *. [See Minn.Stat. § 518.17, subd. 1(c) and (e)].
5. Both Mr. Jorschumb, a full-time farmer, and Mrs. Jorschumb, a part-time waitress, are fit parents and are now in good mental and physical health. [See Minn.Stat. § 518.17, subd. 1(g) and (h) ].
6. [J.A.] has developed a close, warm, open, physical, and loving relationship with each parent. [See Minn.Stat. § 518.17, subd. 1(c), (e) and (h) ].
Judgment, October 3, 1985.
7. The evidence presented at the [final custody hearing] hearing including expert testimony of psychologists, demonstrated to the Court’s satisfaction that the mother had become the primary caretaker of [J.A.], and there were also relevant factors justifying placement with the mother including the close filial relationship [J.A.] had established with his older half-brother who resided with the mother. [See Minn.Stat. § 518.17, subd. 1(c), (d), (e), (f) and (h) ]. (Emphasis supplied.)
Trial Court Memorandum, November 26, 1985.
*813A trial court is not required to address every “best interest” factor, rather, “[i]t is sufficient if the findings as a whole reflect that the trial court has taken the statutory factors into consideration, in so far as they are relevant, in reaching its decision.” Rosenfeld, 311 Minn, at 83, 249 N.W.2d at 171-72.
Although the findings sufficiently incorporate the criteria in Minn.Stat. § 518.17, we must further determine whether the trial court abused its discretion in making findings unsupported by the evidence and ultimately in awarding custody to the mother.
The evidence demonstrated that both parties have a positive influence on the child and are able to provide a stable and loving environment. The child is emotionally well-adjusted, has a particularly close relationship with his step-brother and frequently verbalizes his love for respondent. It was also clearly established, however, that the child has an especially strong bond with his father and that he enjoys spending time on the family farm.
Custody decisions are generally one of the most difficult decisions a trial court must make. While we might have acted differently in this case, we cannot say that the trial court abused its broad discretion in light of the record as a whole. Nothing prevents appellant from seeking a modification in custody under Minn.Stat. § 518.18 should a change in circumstances occur.
We do not approve, however, of the trial court’s temporary award of split custody in order to determine which party would “grow into the role of primary caretaker.” This is inappropriate under Pikula and can only serve to increase competition between parents at their children’s expense.
2. Appellant additionally contends that Dr. Husted’s testimony was improperly considered since it was based in part on an MMPI test of respondent’s that she refused to disclose during discovery. This argument fails on numerous grounds. First, appellant failed to make a timely motion to compel discovery. Although a motion to compel was made on August 13, 1985, the court deferred the matter to the trial judge who would hear the case. Appellant failed to make a subsequent motion. Second, appellant failed to object when Dr. Husted was asked about the MMPI at trial. Third, appellant, not respondent, introduced the MMPI tests into evidence without objection. Fourth, appellant’s own witness, Marie Ostergaard, testified that although she had not interviewed respondent, she had read the MMPI report and “it had a bearing” on her opinion that appellant should receive custody. Even if the MMPI should not have been considered, it does not follow that Dr. Husted’s testimony was inadmissible in its entirety.
DECISION
The trial court did not abuse its discretion in awarding custody of the child to respondent.
Affirmed.
. Minn.Stat. § 518.17, subd. 1 (1984) provides:
. To aid trial courts in determining which parent, if either, has taken primary responsibility for the child’s nurturing and care, the court adopted the indicia of parenthood set forth in Garska v. McCoy, 278 S.E.2d 357 (W.Va.1981). Important factors include, inter alia:
(1) preparing and planning of meals; (2) bathing, grooming and dressing; (3) purchasing, cleaning, and care of clothes; (4) medical care, including nursing and trips to physicians; (5) arranging for social interaction among peers after school, i.e. transporting to friends’ houses or, for example, to girl or boy scout meetings; (6) arranging alternative care, i.e. babysitting, day-care, etc.; (7) putting child to bed at night, attending to child in the middle of the night, waking child in the morning; (8) disciplining, i.e. teaching general manners and toilet training; (9) educating, i.e. religious, cultural, social, etc.; and, (10) teaching elementary skills, i.e., reading, writing and arithmetic.
Pikula v. Pikula, 374 N.W.2d 705, 713-14 (Minn.1985) (quoting Garska, 278 S.E.2d at 363).
.Expounding on the phrase "at the time the dissolution proceeding was commenced,” the Pi-kula court stated:
[This phrase] is used to indicate the point in time at which the family relationships were physically disrupted by events leading to the dissolution of the marriage, e.g., at the time of the parties' separation or the interruption of the functioning full family unit.
Pikula, 374 N.W.2d at 714, n. 3 (emphasis supplied).
. The mailed "survey" approach utilized by appellant and partially relied upon by Ostergaard as a basis for her opinion does not find favor with this court. Significantly, many of those responding to the questionnaire had ties with appellant’s family. None of those who responded to the survey had observed respondent and the child together on an ongoing basis. The prejudicial impact and unreliability inherent in such a device is obvious and its future use is discouraged.