Defendant appealed his conviction of driving while impaired in violation of N.C.G.S. § 20-138.1. He contended that the trial court erred by admitting into evidence the results of a horizontal gaze nystagmus (HGN) test without the establishment of a proper foundation. Defendant contended that the HGN test is a scientific test requiring expert testimony as to its reliability. The Court of Appeals agreed that the State failed to lay a proper foundation at trial for admission of the HGN test results. Nevertheless, the panel concluded that the error was harmless and upheld defendant’s conviction. We agree with the Court of Appeals on the admissibility of the HGN test results but reverse on the issue of harmless error.
The Court of Appeals held that Monroe Public Safety Officer E.P. Bradley’s testimony regarding the HGN test results was inadmissible and declined to take judicial notice of the validity of the test. Though it concluded that the admission of the HGN test results into evidence was improper, the court determined that the remaining testimony at trial overwhelmingly established defendant’s guilt of driving while impaired. Thus, it held the error was harmless.
Nystagmus has been defined as a physiological condition that involves
“an involuntary rapid movement of the eyeball, which may be horizontal, vertical, or rotary. An inability of the eyes to maintain visual fixation as they are turned from side to side (in other words, jerking or bouncing) is known as horizontal gaze nystagmus, or HGN.”
People v. Leahy, 8 Cal. 4th 587, 592, 882 P.2d 321, 323, 34 Cal. Rptr. 2d 663, 665 (1994) (quoting People v. Ojeda, 225 Cal. App. 3d 404, 406, 275 Cal. Rptr. 472, 472 (1990)) (citations omitted in original). In administering the HGN test,
the subject is asked to cover one eye and then use the remaining eye to track the lateral progress of an object (usually a pen) as *580the officer moves the object at eye-level across the subject’s field of vision. As the moving object travels toward the outside of the subject’s vision, the officer watches the subject’s eye for “nystagmus” — an involuntary jerking movement of the eyeball. If the person’s eyeball exhibits nystagmus, and especially if the nystagmus occurs before the moving object has traveled 45 degrees from the center of the person’s vision, this is taken as an indication that the person is intoxicated.
Ballard v. State, 955 P.2d 931, 933 (Alaska Ct. App. 1998).
This Court has not previously addressed the admissibility of HGN evidence. In now doing so, we look first to other jurisdictions which have addressed the issue. Some courts have held that the results of HGN tests are admissible without evidentiary foundation. They reason that the HGN test is simply another field sobriety test, such as the finger-to-nose, sway, and walk-and-turn tests, admitted as evidence of intoxication. Whitson v. State, 314 Ark. 458, 863 S.W.2d 794 (1993); State v. Murphy, 451 N.W.2d 154 (Iowa 1990); City of Fargo v. McLaughlin, 512 N.W.2d 700 (N.D. 1994); State v. Nagel, 30 Ohio App. 3d 80, 506 N.E.2d 285 (1986); State v. Sullivan, 310 S.C. 311, 426 S.E.2d 766 (1993). The Ohio Court of Appeals, for example, noted that
[t]he gaze nystagmus test, as do the other commonly used field sobriety tests, requires only the personal observation of the officer administering it. It is objective in nature and does not require expert inteipretation. ...
It should be remembered that the [HGN] test was one of a number of field sobriety tests administered by the officer to assist him in assessing [defendant’s] physical condition. Taken together, they were strongly suggestive of intoxication. It does not require an expert to make such objective determinations.
A majority of those jurisdictions addressing the admissibility of HGN evidence, however, have concluded the HGN test is a scientific test requiring a proper foundation to be admissible. See, e.g., Ballard v. State, 955 P.2d 931 (Alaska Ct. App.); State v. Superior Court, 149 Ariz. 269, 718 P.2d 171 (1986); State v. Meador, 674 So. 2d 826 (Fla. Dist. Ct. App.), disc. rev. denied, 686 So. 2d 580 (Fla. 1996); Commonwealth v. Sands, 424 Mass. 184, 675 N.E.2d 370 (1997); Emerson v. State, 880 S.W.2d 759 (Tex. Crim. App.), cert, denied, 513 *581U.S. 931, 130 L. Ed. 2d 284 (1994). The courts which hold that HGN tests are scientific tests note that the HGN test is based on an underlying scientific assumption that a strong correlation exists between intoxication and nystagmus. Because that assumption is not within the common experience of jurors, these courts hold that before HGN evidence may be heard by a jury there must be testimony as to the techniques used by the police officer and the officer’s qualifications to administer and interpret the test.
A subset of those courts which hold that HGN tests are scientific in nature also hold that expert testimony is required to establish that the scientific principles upon which the HGN test is based are generally accepted by the scientific community. According to these cases, unless a police officer has special training or adequate knowledge qualifying him as an expert to explain the correlation between intoxication and nystagmus, his testimony is not adequate foundation for the admission of HGN test results. People v. Leahy, 8 Cal. 4th 587, 882 P.2d 321, 34 Cal. Rptr. 663; State v. Ruthardt, 680 A.2d 349 (Del. Super. Ct. 1996); Schultz v. State, 106 Md. App. 145, 664 A.2d 60 (1995); Hulse v. State,-Mont.-,-P.2d-, 1998 WL 239615 (May 5, 1998) (No.96-541); Commonwealth v. Miller, 367 Pa. Super. 359, 532 A.2d 1186 (1987); State v. Cissne, 72 Wash. App. 677, 865 P.2d 564, disc. rev. denied, 124 Wash. 2d 1006, 877 P.2d 1288 (1994).
 In the instant case, the Court of Appeals held, in accord with the majority view, that the HGN test does not measure behavior a lay person would commonly associate with intoxication, but rather represents specialized knowledge that must be presented to the jury by a qualified expert. We agree. Once the expert testifies as to the relationship between HGN test results and intoxication, he or she is then subject to cross-examination to test the validity and reliability of the HGN test. Appropriate questions on cross-examination might be whether eye twitching or nystagmus could also be caused by nervousness, certain diseases, lack of sleep, or certain medications rather than alcohol intoxication. See Schultz, 106 Md. App. at 180-81, 664 A.2d at 77 (listing thirty-eight causes of nystagmus other than alcohol intoxication).
 Under the North Carolina Rules of Evidence, “new scientific methodjs] of proof [are] admissible at trial if the method is sufficiently reliable.” State v. Pennington, 327 N.C. 89, 98, 393 S.E.2d 847, 852 (1990); State v. Goode, 341 N.C. 513, 461 S.E.2d 631 (1995) (special agent’s testimony on bloodstain pattern interpretation admissible *582after voir dire testimony showing reliability). This Court has stated that, “ ‘[i]n general, when no specific precedent exists, scientifically accepted reliability justifies admission of the testimony of qualified witnesses, and such reliability may be found either by judicial notice or from the testimony of scientists who are expert in the subject matter, or by a combination of the two.’ ” State v. Bullard, 312 N.C. 129, 148, 322 S.E.2d 370, 381 (1984) (quoting 1 Henry Brandis, Jr., Brandis on North Carolina Evidence § 86, at 323 (2d ed. 1982)). We find nothing in the record of the case before us to indicate that the trial court took judicial notice of the reliability of the HGN test. Further, while Officer Bradley testified that he had taken a forty-hour training course in the use of the HGN test, the State presented no evidence and the court conducted no inquiry regarding reliability of the HGN test. Until there is sufficient scientifically reliable evidence as to the correlation between intoxication and nystagmus, it is improper to permit a lay person to testify as to the meaning of HGN test results. Accordingly, in this case the admission of Bradley’s testimony regarding the results of the HGN test administered to defendant was error.
Notwithstanding its conclusion that the admission of the HGN test results was improper,, the Court of Appeals held that receipt of the evidence constituted harmless error because the remaining evidence presented at trial overwhelmingly established defendant’s guilt of the crime of driving while impaired. The remaining evidence against defendant presented by the State is as follows: (1) Bradley testified that defendant’s car weaved from its lane of travel and struck the right curb with its right front tire before coming back into its lane; (2) after Bradley activated his blue light, defendant’s car made a wide right turn onto a side street, veering into the opposite lane before stopping; (3) Bradley noticed a strong odor of alcohol coming from the car when defendant rolled down the driver’s side window; (4) defendant was unsteady on his feet; (5) defendant’s eyes were bloodshot and his clothes were disheveled; (6) there was an odor of alcohol coming from defendant; (7) defendant’s speech was “mumbled;” and (8) defendant failed other field sobriety tests. The following additional evidence was elicited on cross-examination: (1) Bradley admitted that the layout of Hill Street requires a wide turn and that no dividing line exists on the street; (2) Bradley admitted that alcohol itself has no odor, but the flavorings of the beverage cause it to smell like an alcoholic beverage; (3) there was no evidence in the record that defendant had been drinking an alcoholic beverage and not a nonalcoholic beverage with similar flavorings; (4) there are *583many different reasons which could cause a person’s eyes to be red other than the use of alcohol; and (5) Bradley could not say for sure that defendant’s speech was abnormal on the night in question because he had never heard defendant speak before.
The evidence presented at trial was clearly sufficient to send the case to the jury and to support a jury finding of guilty of driving while impaired. However, that is not the question before us. The question is not one of sufficiency of the evidence to support the jury verdict. In order to establish prejudicial error in the erroneous admission of the HGN evidence, defendant must show only that had the error in question not been committed, a reasonable possibility exists that a different result would have been reached at trial. N.C.G.S. § 15A-1443(a) (1997). We conclude that, in light of the heightened credence juries tend to give scientific evidence, there is a reasonable possibility that had evidence of the HGN test results not been erroneously admitted a different outcome would have been reached at trial.
Accordingly, and for the reasons stated herein, we reverse the Court of Appeals on the issue of prejudicial error and remand to that court for further remand to the trial court for a new trial.
REVERSED AND REMANDED.