928 So.2d 90 (2005)
Donnie FRANKLIN
v.
Tiffany FRANKLIN.
No. 2005 CU 1814.
Court of Appeal of Louisiana, First Circuit.
December 22, 2005.
Writ Denied February 17, 2006.
Rebecca E. May-Ricks, J. Arthur Smith, III, Baton Rouge, Counsel for Plaintiff/Appellee Donnie Franklin.
H. Michael Aaron, Baton Rouge, Counsel for Defendant/Appellant Tiffany Franklin.
Before: WHIPPLE, McCLENDON, and WELCH, JJ.
McCLENDON, J.
In this child custody case, the trial court maintained joint custody, but changed the domiciliary parent to the father, Donnie Franklin,[1] and awarded the mother, Tiffany Franklin, visitation. Mrs. Franklin appealed. We reverse and remand for a new trial.
PROCEDURAL BACKGROUND
The parties were granted a divorce by a judgment signed on July 29, 2003. On September 11, 2003, Mrs. Franklin filed a petition for change of custody. She alleged that a prior stipulated judgment of joint custody, with the mother designated as the domiciliary parent, was never signed, and that one of the two children had been abused. In the prayer, Mrs. Franklin sought an ex parte order awarding her sole custody of the parties' two children, or a provisional award of sole custody pending a hearing. She also asked for a hearing where Mr. Franklin would show cause why Mrs. Franklin should not be awarded sole custody or, alternatively, designated as the domiciliary parent in a new plan of joint custody. According to the court minutes of September 15, 2003, the court ordered a show cause hearing, and awarded Mrs. Franklin temporary custody, with supervised visitation for Mr. Franklin. On September 19, 2003, Mr. Franklin filed an opposing motion asking for sole custody based on Mrs. Franklin's marriage to a "convicted drug dealer," and her accusations of child abuse against Mr. Franklin.
The hearing was held on August 25, 2004. In written reasons issued after the hearing, the trial court ordered the parties to prepare a joint custody plan, designating Mr. Franklin as the domiciliary parent, with frequent visitation for Mrs. Franklin. By judgment signed on May 13, 2005, the trial court decreed that the "filed ... Joint Custody Implementation Plan be and hereby is accepted by this Court, made a part of the record, adopted by and given the same force and effect as a Judgment of this Court."[2]
Mrs. Franklin appealed and assigned errors to the May, 2005 judgment. One of her assignments challenged the admission of a polygraph examination. Finding merit in the claim, we pretermit consideration of her other assignments of error.
ADMISSIBILITY OF POLYGRAPH UNDER DAUBERT TEST
At trial, Mr. Franklin's counsel sought to have Mr. Don Zuelke qualified as an expert who would testify to the results of a polygraph examination given to Mr. Franklin. However, before the evidence could be admitted, the trial court ordered Mr. Franklin to show not only that Mr. Zuelke was an expert polygraphist, but that the field of polygraph examinations met the requirements of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993).
To be generally admissible, evidence must be relevant and not unduly prejudicial. LSA-C.E. arts. 104 & 401-403. "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." LSA-C.E. art. 702.
Additionally, the Louisiana Supreme Court has adopted the reasoning and observations set forth in Daubert, which specifically rejected the "general acceptance" test and set forth guidelines for determining the reliability and admissibility of expert scientific testimony. See State v. Foret, 628 So.2d 1116, 1122-23 (La.1993). In Daubert, the United States Supreme Court "stated that an inference or assertion of scientific knowledge must be derived by the scientific method. Proposed testimony must be supported by appropriate validation, i.e., `good grounds,' based on what is known." State v. Manning, 03-1982, p. 44 (La.10/19/04), 885 So.2d 1044, 1086-1087, cert. denied, 544 U.S. 967, 125 S.Ct. 1745, 161 L.Ed.2d 612 (2005), quoting from Daubert, 509 U.S. at 590, 113 S.Ct. at 2795. "In short, evidentiary reliability will be based on scientific validity." State v. Manning, 03-1982 at p. 44, 885 So.2d at 1087, citing, Daubert, 509 U.S. at 590 n. 9, 113 S.Ct. at 2795 n. 9.
The Daubert/Foret guidelines are used as an aid in interpreting article 702 and ensure that scientific and technical expert testimony meets minimal standards of reliability and relevance. See State v. Foret, 628 So.2d at 1123. The Daubert/Foret guidelines require that expert opinions be grounded in approved methods and procedures of science, rather than subjective belief or unsupported speculation. The trial court must also ensure that the scientific "evidence admitted is not only relevant, but reliable." Daubert, 509 U.S. at 589, 113 S.Ct at 2795; see LSA-C.E. 104 & 401-403. Before the expert opinion can be admitted, the trial court must make "a preliminary assessment that the reasoning or methodology underlying the testimony is scientifically valid" and of whether that reasoning or methodology properly can be applied to the facts at issue. Vardaman v. Baker Center, Inc., 96-2611, p. 6 n. 6 (La.App. 1 Cir. 3/13/98), 711 So.2d 727, 731 n. 6. The court must also determine whether the "probative value" of the expert testimony or opinion would be "substantially outweighed by the danger of" confusion or an undue prejudicial effect on the fact finder. LSA-C.E. art. 403; see Foret, 628 So.2d at 1127; Fussell v. Roadrunner Towing and Recovery, Inc., 99-0194, p. 3 (La.App. 1 Cir. 3/31/00), 765 So.2d 373, 376, writ denied, 00-1264 (La.6/23/00), 765 So.2d 1042; State v. Brooks, 98-1151, p. 19 (La.App. 1 Cir. 4/15/99), 734 So.2d 1232, 1242, writ denied, 99-1462 (La.11/12/99), 749 So.2d 651. To fulfill the trial court's gatekeeper function for proposed expert evidence, various factors may be considered by the trial judge: (1) whether the technique has been subjected to peer review or publication, (2) the "known or potential rate of error," (3) a "reliability assessment," in which the "degree of acceptance" within a scientific community may be determined and reviewed, and (4) the "testability" of the technique. Wingfield v. State, Department of Transportation and Development, 01-2668, pp. 9-10 (La.App. 1 Cir. 11/8/02), 835 So.2d 785, 796, writs denied, 03-0313 (La.5/30/03), 845 So.2d 1059 and 03-0339 & 03-0349 (La.5/30/03), 845 So.2d 1060, cert. denied, 540 U.S. 950, 124 S.Ct. 419, 157 L.Ed.2d 282 (2003), quoting from Daubert, 509 U.S. at 593-94, 113 S.Ct. at 2796-97. However, the approach is a flexible one, and the list of factors is neither exclusive nor mandatory. Daubert, 509 U.S. at 593-94, 113 S.Ct. at 2796-2797; Wingfield, 01-2668 at p. 9, 835 So.2d at 796. The admission of evidence, expert or otherwise, is subject to the trial court's discretion. Wingfield, 01-2668 at pp. 9-10, 835 So.2d at 796; Fussell, 99-0194 at p. 3, 765 So.2d at 375; see State v. Catanese, 368 So.2d 975, 983 (La.1979).
The Louisiana Supreme Court has ruled that polygraphs are not admissible in criminal trials, but are admissible under certain conditions in an administrative hearing.[3]See Evans v. DeRidder Municipal Fire, 01-2466, pp. 7 & 13 (La.4/3/02), 815 So.2d 61, 67 & 71, cert. denied, 537 U.S. 1108, 123 S.Ct. 884, 154 L.Ed.2d 779 (2003). However, the court has not ruled on whether polygraphs are admissible in a civil trial. Evans, 01-2466 at p. 6, 815 So.2d at 66.
During the Daubert hearing, Mr. Zuelke testified that polygraphs are widely used in the law enforcement field, that polygraph examiners, or polygraphists, are licensed in Louisiana, that his examinations had a ninety-eight percent rate of correctness, and that the American Polygraph Association would give the overall rate of correctness for polygraphs as ninety percent. After Mr. Franklin's counsel asked that the expert testimony of Mr. Zuelke be admitted, Mrs. Franklin's counsel objected to the acceptance of polygraphs based on the absence of empirical scientific research supporting the reliability of polygraph examinations and the lack of evidence on whether the examination was accepted by the scientific community outside the examiners themselves. Based on the initial testimony, the trial court ruled that Mr. Franklin had not yet met the Daubert/Foret requirements or test for reliability.
In response, Mr. Zuelke further testified that he was aware of studies in the field conducted by polygraph examiners. Although Mr. Zuelke was generally aware of those studies, he was unable to give any specifics of the studies beyond the finding of most that the overall correctness rate was ninety percent. Nor did he relate the names of the studies, where published, or how they were conducted.
On cross-examination, Mr. Zuelke admitted that he had never been qualified as an expert by a federal court or by a district court in Louisiana, and that to his knowledge, the field of polygraph examinations had not successfully passed the Daubert/Foret test in Louisiana. He also admitted that he had read a study by the National Academy of Science and was aware of a study done by the federal government, which noted the lack of research validating screening polygraphs. When asked about the results of the studies, Mr. Zuelke distinguished both studies because they were concerned with screening examinations involving multiple issues, and not the single issue testing Mr. Zuelke used. However, Mr. Zuelke was not aware of any outside scientific studies validating single issue polygraphs. For the three studies conducted by polygraphists mentioned by Mr. Zuelke, he testified that all but one found a correctness rate in the high ninety percentile. He gave no information on the publication of the studies, the method of testing employed, or the statistical basis underlying the correctness rate. During the cross examination, Mr. Zuelke also testified that the "biggest problem of studying polygraph[s] is that it's not like other sciences where you can create a laboratory setting" and that creates "a problem in determining accuracy." Further, he agreed that the alleged ninety percent overall rate of correctness could not be validated using the scientific method of testing.
During the testimony, the trial court asked questions, and seemed focused on how the examination was conducted and on certain questions and Mr. Franklin's answers. At the end of the Daubert hearing, the trial court qualified Mr. Zuelke as an expert, and allowed him to testify about the polygraph test given to Mr. Franklin, including Mr. Zuelke's opinion that Mr. Franklin was truthful in answering that he had not abused his daughter.
After reviewing the totality of the record on this issue, we find that the trial court abused its discretion in ruling that the polygraph examination was admissible. It did not meet the Daubert analysis.[4] Therefore, we find that the improper admission of the polygraph impermissibly interdicted the fact finding process.
STANDARD FOR APPELLATE REVIEW AND REMAND
Generally, when a legal error interdicts the fact finding process, the manifest error standard no longer applies. If the record is otherwise complete, the reviewing court should conduct a de novo review. Landry v. Bellanger, 02-1443, p. 15 (La.5/20/03), 851 So.2d 943, 954; Ragas v. Argonaut Southwest Insurance Co., 388 So.2d 707, 708 (La.1980); Savin v. Allstate Insurance Company, 579 So.2d 453, 457 (La.App. 1 Cir.1991); Moore v. Clark, 517 So.2d 293, 302 (La.App. 1 Cir.1987). However, "[t]here are cases where the weight of the evidence is so nearly equal that a first-hand view of witnesses is essential to a fair resolution of the issues." Ragas, 388 So.2d at 708; see Landry, 02-1443 at p. 15, 851 So.2d at 954. Where such a need arises, "the case should be remanded for a new trial." Landry, 02-1443 at p. 15, 851 So.2d at 954; Ragas, 388 So.2d at 708. It is the duty of the appellate court to determine when "the court can fairly find a preponderance of the evidence from the cold record," or whether the case should be remanded. Id.
As in Savin, the evidence here is in substantial conflict. The trial court's firsthand view of the witnesses is necessary for a fair determination of credibility, especially in this custody case with equivocal allegations of child abuse. With such substantial conflicts and the need for numerous credibility determinations, a preponderance of the evidence cannot be fairly determined from this cold record. Therefore, we are convinced that the interests of justice will be best served by remanding this case for a new trial. See Savin, 579 So.2d at 457-58; LSA-C.C. P. art. 2164.
CONCLUSION
For these reasons, we reverse the judgment and remand for a new trial, to be heard as expeditiously as possible.[5] However, based on the conflicts and ambiguity in the record on the best arrangement for the children, and children's overall need for stability and continuity, the present custody is maintained. See Bergeron v. Bergeron, 492 So.2d 1193, 1199 (La.1986). Of course, upon remand, the parties have at their disposal all the legal avenues for requesting emergency changes of custody, if that is deemed necessary. The costs of the appeal are assessed to the appellee, Donnie Franklin.
REVERSED AND REMANDED.
WELCH, J., concurs with reasons.
WELCH, J., Concurring.
I respectfully concur. I agree with the majority's finding that the trial court abused its discretion in ruling the testimony of Mr. Don Zuelke, the polygraphist, and the results of the polygraph he performed on Donnie Franklin were admissible under the Daubert/Foret standards. This erroneous admission of evidence constituted legal error that impermissibly interdicted the fact finding process. Therefore, this matter should be remanded for a new trial. However, because the inadmissible polygraph evidence has already tainted the fact finder in this matter, I believe that it is inappropriate to remand this matter back to the same fact finder. Therefore, not only should this matter be remanded for a new trial, but I also believe that it should be re-allotted to another division of the Family Court.[1]
NOTES
[1] In the record, Mr. Franklin is sometimes referred to as Donald W. Franklin.
[2] Although the implementation plan was not included in the record before us on appeal, it is undisputed that the joint custody was maintained and the father was designated the domiciliary parent.
[3] Courts outside Louisiana have made similar rulings limiting the use of polygraphs. See Cassamassima v. Florida, 657 So.2d 906 (Fla. App. 5th DCA 1995), and cases cited therein. The Florida appellate court in Cassamassima also noted that the finder of fact may give "disproportionate weight" to the polygraph as a means of determining credibility. Cassamassima, 657 So.2d at 908.
[4] We do not address the admissibility of polygraphs upon a proper showing of the Daubert/Foret requirements for reliability.
[5] Although we pretermitted the other assignments of error, we note that any possible errors may be cured by the order of a new trial, and the trial court may be aided by the admission of additional evidence.
[1] While I have great respect for the trial judge in this matter, from a litigant's perspective, they would believe that it would be impossible for any trial judge to "erase" from his mind the polygraph evidence that he has already heard and considered. Therefore, any decision reached would appear to be tainted by the prior polygraph evidence.