Appellant Bessie Landis brought suit for personal injuries allegedly sustained on May 28, 1979, when a vehicle she was driving was struck by a truck driven by appellee Rodger Hastings, fifteen year old son of appellee Bobby Hastings. The jury returned a verdict for the defendants, notwithstanding Rodger Hastings’ admission the collision was his fault and testimony on behalf of Mrs. Landis that she sustained permanent injuries as a result of the collision. Her motion for new trial was denied and she has appealed.
Four points are argued: the j ury disregarded undisputed evidence in reaching a contrary result; the verdict was not supported by substantial evidence; the trial judge’s refusal to grant a new trial was an abuse of discretion and it was error to allow the jury to consider a verdict for the defendant. We find the arguments to be without merit and we affirm.
The Landis proof was that Mrs. Landis, driving a borrowed vehicle, entered a protected intersection in Judsonia and was struck on the right rear side by the Hastings truck. The impact spun her around and into a deep ditch, knocked off her shoes, and caused damage to her vehicle in excess of $1,200. Not thinking she was injured, she drove another vehicle to Little Rock to meet her niece at the airport but by the time she arrived she was in pain and unable to drive home.
*137The next day Mrs. Landis consulted Dr. Sidney Tate for pain and tenderness in her neck and shoulders. Dr. Tate found motion limitation and muscle spasms; he made a diagnosis of cervical sprain and possible ligament damage. He prescribed conservative treatment of rest, muscle relaxors and later ultra sound treatments, physical therapy and use of a cervical collar. Mrs. Landis was still under Dr. Tate’s care at trial two years after the accident, and he estimated pain in some degree would continue the remainder of her life. In addition to Dr. Tate, Mrs. Landis was treated by Dr. L. S. Tensuan during a visit to Pennsylvania and examined by Dr. William Steel. Her x-rays on May 29 and July 6, 1979, were read by Dr. Robert Elliott, a radiologist. Her medical bills, admitted without objection, totaled $973 as of the trial on May 12, 1981.
In defense, Rodger Hastings testified the collision was his fault. He estimated his speed at less than 10 miles per hour at impact. He said his truck moved only a few feet after striking Mrs. Landis’s vehicle and sustained no damage at all. He went directly to Mrs. Landis and found her on her seat behind the wheel putting on high heel slippers. She told him she was not hurt and she walked around while waiting on the police. Hastings heard Mrs. Landis tell the police and by-standers that she was not hurt in any way. He described Mrs. Landis as upset but said he observed nothing to cause him to believe she suffered any injury. He told of having seen Mrs. Landis dancing at the Electric Cowboy some months prior to the trial. A passenger in the Hastings truck corroborated testimony that Mrs. Landis did not appear to have suffered any injury and told several people she was not hurt. Both witnesses said the area Mrs. Landis described as a "deep ditch” was a depression of about three inches.
In acting on a motion for a new trial the test to be applied by the trial court is whether the verdict is against the preponderance of the evidence. ARCP Rule 59 (a). But on review the test depends on whether the motion was granted, in which case we will affirm absent a manifest or clear abuse of discretion (General Motors Corp. v. Tate, 257 Ark. 347, 516 S.W.2d 602, 1974), whereas if the motion is denied, as here, we look only to see if the verdict is supported by *138substantial evidence and, if so, we do not disturb the trial court’s action. Ferrell v. Whittington, 271 Ark. 750, 610 S.W.2d 572 (1981); Brady v. City of Springdale, 246 Ark. 1103, 441 S.W.2d 81 (1969). The rule is founded on the superior position of the trial court to hear and weigh the evidence. Ferrell, supra-
Appellant submits that the trial judge was confused as to the test to be applied in passing on her motion, arguing that an abuse of discretion resulted. But while there may have been some initial uncertainty as to the proper test, it is clear the opposing arguments at a hearing on the motion put the issue in proper perspective and as a result the trial court advised counsel he would reread the cases and review his “detailed notes on the case.” We believe his decision in the end was in accordance with the preponderance of the evidence standard. From his comments he evidently believed the evidence was such the jury could have reasonably concluded that Mrs. Landis did not sustain injuries as a result of the collision.
Next, appellant cites us to the dicta of a number of cases stating that testimony which is not substantially contradicted by other testimony, and does not seem unreasonable or improbable, may not be arbitrarily or capriciously disregarded by the jury. General Motors Corp. v. Tate, supra; Missouri-Pacific Rd. Co. v. Ross, 194 Ark. 877, 109 S.W.2d 1246 (1937); St. Louis-San Francisco Ry. Co. v. Pace, 193 Ark. 484, 101 S.W.2d 447 (1937); St. Louis-San Francisco Ry. Co. v. Williams, 180 Ark. 413, 21 S.W.2d 611 (1929). These cases have a common feature, in each a verdict for the plaintiff was reversed and dismissed because of a lack of evidence to sustain it. The cases present a somewhat different evidentiary problem from the case before us and it is in that context that the cited dicta must be examined. In the Ross and Pace cases, wrongful death claims were brought against railroad companies resulting in verdicts for the plaintiff where plaintiffs’ decedents were found on or near the railroad track with no eyewitness accounts of how the accidents occurred. The plaintiffs’ theories of recovery were founded essentially on conjecture whereas testimony from witnesses for the railroad was uncontradicted. In Williams *139the plaintiff was injured while a trespasser on railroad property, and the court noted an absence of any fact or circumstance in evidence substantially contradicting the testimony of the engineer that because of cars on a siding he could not see the plaintiff until nearly upon her. Thus, it was said in rationalizing a basis for reversal that testimony of witnesses, if plausible, may not be arbitrarily disregarded. Of Tate it can be said that here, too, the court found an absence of substantial evidence of negligence in support of the jury’s verdict. The difference is to be found we think in the fact that in the cited cases the juries’ disregard of evidence was arbitrary, whereas here there was a genuine dispute.
Turning to the issue of substantial evidence, it should be noted the testimony of Mrs. Landis that her injuries were caused by the collision is, of course, controverted as a matter of law. Bittle v. Smith, 254 Ark. 123, 491 S.W.2d 815 (1973). And while her claim was generally supported by the testimony of the medical witnesses we do not find their testimony to have been so certain that the jury was required to attribute her complaints to the accident. From their testimony it is clear that Mrs. Landis demonstrated evidence of traumatic injury, but whether caused by the episode on May 28, 1979, they were not able to say. Dr. Elliott admitted the trauma could have been years earlier and acknowledged he could not say the conditions he observed were caused by the accident.
It is undisputed Mrs. Landis, who is 63, had had a previous injury and chronic, long-standing degenerative disc disease with spurring and lipping of the cervical vertebrae, a condition sufficiently severe of itself to cause periodic pain. The jury may have decided her complaints were attributable to pre-existing conditions rather than to the accident. There was testimony that she stated repeatedly she was not hurt and made no complaint during the 30 minutes or so she remained at the scene. Too, the jury may have been influenced by contradictions in her testimony and the opposing evidence: Mrs. Landis denied having danced at the Electric Cowboy in contrast to Rodger Hastings’s testimony he had seen her there dancing; she said her vehicle ended up in a “deep ditch,” in contrast to testimony the *140yards and roadway were level at the scene except for a swale no deeper than three inches and, finally, her denial that she was having pain or taking medication or tranquilizers prior to the accident was in direct conflict with the medical records of an earlier physician. Under the evidence the jury could have found for the defendant.
Appellant’s remaining argument is the court should not have submitted a verdict form for the defendant to the jury, because it was admitted by Rodger Hastings that the collision was his fault. But that would have been tantamount to a directed verdict for the plaintiff and for reasons already discussed we think whether Mrs. Landis sustained an injury as a result of the collision was a disputed issue and properly submitted to the jury.
The order denying the motion for a new trial is, accordingly, affirmed.