In this personal injury case arising out of a car-truck collision, plaintiff appeals from a judgment on a jury verdict for defendants.1 She assigns error to several of the trial court’s evidentiary rulings and to its jury instructions. We reverse and remand.
On December 19,1988, plaintiff was driving her car southbound on a two-lane highway in Douglas County. Defendant Price was travelling northbound on the same road in a truck with a full-size trailer and a ‘ ‘pup, ” or short, trailer. It had rained, and the road was wet. The two vehicles passed each other at a bend in the road. The bend was just south of a narrow bridge and was to plaintiffs left and to Price’s right. Price testified that, as the cab of his truck passed plaintiffs car, each vehicle was in its proper lane. Price testified that his truck popped out of gear and, as he was putting it back in gear, he looked in his side-view mirror and saw plaintiffs car in a wreck on her side of the road. His pup trailer had lost its réar wheels and was dragging on the ground. Plaintiff was severely injured and does not remember the collision. There were no other witnesses. Plaintiffs theory of the collision is that the pup trailer swept into her lane as Price’s truck went through the bend in the road. Defendants’ theory is that plaintiff cut the corner and crossed into the northbound lane and into the pup trailer. The jury returned a general verdict for defendants.2
Plaintiffs first two assignments of error are related, and we analyze them together. She argues that the trial court *323erred in allowing defendants’ expert witness, Weaver, to testify that drivers generally tend to cut corners and that drivers generally tend to cut the particular corner where the collision occurred. Specifically, plaintiff argues that the general tendency testimony is irrelevant to prove what she did at the time of the collision.3 Defendants counter that the general tendency testimony is relevant to causation, because it concerns the effect of highway design on the tendency of drivers to cut corners and supports their theory that plaintiff crossed the center line. We review rulings on relevance for errors of law. OEC 401. If we find error, we may reverse only if we find that the error affected a substantial right of the party. OEC 103(1). That test is satisfied if the result of the trial might have been different. Pearson v. Galvin, 253 Or 331, 340, 454 P2d 638 (1969); Hass v. Port of Portland, 112 Or App 308, 314, 829 P2d 1008, rev den 314 Or 391 (1992).
Defendants’ argument that the testimony concerned the effects of highway design on the general tendencies of drivers mischaracterizes the evidence. Although couched in terms of highway design, Weaver’s testimony concerned the general tendency of all drivers to cut any corner on any road, and the general tendency of all drivers to cut the specific corner where the collision occurred. Weaver did not discuss any particular aspect of highway design except that the road curved.
General tendency evidence is in the same vein as evidence of prior conduct or events, which is generally inadmissible to prove negligence or lack of negligence. See, e.g., Warner v. Maus, 209 Or 529, 304 P2d 423 (1956). Here, the case against admissibility is even stronger. Defendants used the general tendencies of drivers as a group to show that plaintiff was negligent and caused the collision. Although *324prior conduct or events maybe admissible to prove causation, danger, knowledge or intent,4 the line between causation and plaintiffs possible negligence is blurred beyond distinction in this case. By trying to establish that plaintiff caused the collision, defendants are trying to establish that plaintiff crossed the center line and was negligent per se through evidence of the general tendencies of other people.
Defendants rely on Carlson v. Piper Aircraft Corp., 57 Or App 695, 646 P2d 43, rev den 293 Or 801 (1982), and Chance v. Ringling Bros., 257 Or 319,478 P2d 613 (1970), to support their argument that the general tendency evidence is relevant. Those cases are of little help to defendants.
At issue in Carlson was the cause of the mid-air break-up of the decedent’s small plane. We upheld the admissibility of expert testimony concerning spatial disorientation, a phenomenon that occurs when inexperienced pilots lose the visual horizon, as when flying in clouds, and become so disoriented that they disregard their instruments and abruptly adjust the aircraft, “with possibly disastrous consequences.” 57 Or App at 699. The testimony on spatial disorientation concerned an involuntary reaction to particular and unfamiliar circumstances. The general tendency testimony at issue here concerned a voluntary decision by drivers in general to violate the law and cross the center line when driving through a bend in the road, a common and familiar circumstance. The testimony did not concern any reactions that might have been triggered by the particular conditions present at the time of the collision, such as the bank or angle of the corner, the weather conditions or, most notably, the presence of a tractor-trailer in the on-coming lane.
Defendants’ reliance on Chance barely merits discussion. In that case, the plaintiff was injured when the defendant’s boxer dog lunged at her. In support of her argument that the defendant knew of the vicious propensities of the dog, the plaintiff was allowed to present evidence that boxers tend to be protective and also tend to jump on people. The evidence was admitted to demonstrate the defendant’s knowledge and did not concern the general tendencies of dog owners or some other class to which the defendant belonged.
*325The general tendency of drivers to cut corners, even this corner, is too remote to be relevant. What other drivers do at different times, locations and under different traffic and weather conditions is not probative of what plaintiff might have done at the time and place in question. See Carter v. Moberly, 263 Or 193, 198, 501 P2d 1276 (1972); Savage Am’x v. Palmer, 204 Or 257, 280 P2d 982 (1955); Southern Pacific Co. v. Consolidated Freightways, Inc., 203 Or 657, 281 P2d 693 (1955).5 The trial court erred in allowing the general tendency testimony.
Because we cannot tell whether the jury might have based its general verdict for defendants on the improperly admitted general tendency evidence, and because “the result of the trial might have been different” had the evidence been excluded, we reverse and remand. Pearson v. Galvin, supra, 253 Or at 331; Hass v. Port of Portland, supra, 112 Or App at 314. We discuss plaintiffs remaining assignments of error, because the issues will likely arise on remand.
In her third assignment, plaintiff argues that the trial court erred in allowing into evidence an aerial photograph of the scene of the collision, because Weaver used the photograph to illustrate his testimony that drivers tend to cut that particular corner. A photograph may be excluded if its probative value is outweighed by the danger of unfair prejudice, confusion of the issues or misleading the jury. OEC 403. Where the evidence is ruled relevant and the trial court has ruled on other admissibility issues, such as those presented by OEC 403 or OEC 702, we review the latter rulings for abuse of discretion. If the record supports either admission or exclusion, we must affirm the trial court’s ruling. Carter v. Moberly, supra, 263 Or at 200.
Defendants argue that the photograph was properly admitted because a proper foundation had been laid, it accurately depicted the scene of the collision and was helpful to the trier of fact. We agree that defendants laid a proper foundation and that any changes in the area as depicted in the *326photograph were explained to the jury. See DeMaris v. Whittier, 280 Or 25, 569 P2d 605 (1977). The photograph could have helped the trier of fact by illustrating the scene of the accident, the relative distances involved and the landmarks discussed by various witnesses. OEC 901. However, the photograph also depicted two vehicles at or near the corner in question, at least one of which was over the center line as it entered the corner. That particular aspect of the photograph raised the risk of unfair prejudice or misleading the jury. However, because the trial court erroneously allowed testimony concerning the general tendencies of drivers to cut corners, it did not rule on whether the risk of unfair prejudice or misleading the jury raised by that particular aspect of the photograph, whether commented on or not, outweighed its probative value. We cannot review a ruling that was not made.
In her fourth assignment of error, plaintiff argues that the trial court erred in allowing Weaver to testify that, in his opinion, based on the general tendencies of drivers, plaintiff must have cut the corner and collided with defendants’ pup trailer in the northbound lane. Plaintiff does not argue that Weaver was not qualified to give such an opinion, but rather that the facts or data on which he relied are not a proper basis for that opinion.
The proponent of an expert opinion has the burden of providing a proper foundation. Kingsbury v. Hickey, 56 Or App 492, 496, 642 P2d 339, rev den 293 Or 146 (1982). The facts or data relied on by an expert witness need not be admissible evidence but, to provide a proper foundation, they must be of “a type reasonably relied on by experts in the particular field in forming opinions.” OEC 703. Whether that test is met is a preliminary matter for the court, which we review for abuse of discretion. OEC 104; Kirkpatrick, Oregon Evidence 459 (2d ed 1989), Legislative Commentary to Rule 703; see State v. Baker, 87 Or App 285, 291, 742 P2d 633, rev den 304 Or 405 (1987). The trial court abused its discretion in allowing expert testimony if the opinion lacked a sufficient foundation, because such an opinion is too speculative. Urbanski v. Johnson, 283 Or 169, 581 P2d 948 (1978).
Weaver testified on direct and cross-examination that his opinion that plaintiff crossed the center line and *327caused the collision was based on the general tendencies of drivers to shy away from solid objects on their right and to cut corners.6 Defendants argue that Weaver’s opinion was premised on several other factors as well, such as his study of the photographs of plaintiffs wrecked car and the accident reports. The record does not substantiate that representation.7 In State v. Baker, supra, we upheld the exclusion of expert testimony about the speed of the defendant’s motorcycle at the time that it collided with an automobile. We agreed with the trial court that the testimony was too speculative, because
“the witness relied solely on police reports and photographs of the vehicles involved in the accident. He did not inspect the *328vehicles personally. He was unaware of the condition of the roadway at the time and place of the accident. He did not hear the witnesses’ description of the accident. He assumed that defendant’s motorcycle brakes were functioning fully but had no personal knowledge as to their effectiveness. He did not consider that defendant’s motorcycle had collided with another motorcycle before the collision with the automobile.” 87 Or App at 291.
If reliance on police reports and photographs of the vehicles involved in an accident is an inadequate basis for an expert opinion as to point of impact, then reliance on the general tendencies of other drivers must also be an inadequate basis. Weaver’s testimony lacked a sufficient foundation based on the facts and physical evidence of this case. It was too speculative and therefore the trial court abused its discretion in admitting it. State v. Baker, supra; see also Urbanski v. Johnson, supra; Kingsbury v. Hickey, supra.
Plaintiff next assigns error to the trial court’s refusal to give her requested instruction that evidence of the general tendencies of other drivers is not evidence of plaintiffs conduct at the time of the collision. Plaintiff is correct for the reasons discussed in our analysis of her first and second assignments of error.
In her sixth assignment, plaintiff argues that the trial court erred in excluding a videotaped demonstration of the phenomenon of trailer sweep. The videotape was offered on three separate occasions, on direct, redirect and rebuttal examination8 of plaintiff’s expert witness, Limpert, who made the videotape. Each time it was offered, defendants objected on the ground that its probative value was outweighed by the risk of unfair prejudice, due to the many dissimilarities between the conditions at the time and place of the collision and at the time and place of the demonstration. The trial court agreed and excluded the videotape because it found that the risk of prejudice outweighed its probative value. OEC 401; OEC 403.
*329The test for the admissibility of such demonstrative evidence is whether the differences in conditions of the demonstration and the event at issue are so great that they risk misleading or confusing the jury. Loibl v. Miemi, 214 Or 172, 327 P2d 786 (1958). The trial court has broad discretion in making that determination. Tuite v. Union Pacific Stages, Inc., 204 Or 565, 284 P2d 333 (1955). Here, the trial court said that, “[i]n balancing whether or not [the demonstration] is sufficiently similar [to the collision] against the prejudice it would create, I find it is too dissimilar and too prejudicial.” We cannot say that the trial court abused its discretion by excluding the videotape on that basis. We note, however, that the trial court allowed plaintiffs expert to testify at length about trailer sweep in general and the demonstration in particular. Also, over defendants’ objections, the court allowed two still photographs of the trailer-sweep demonstration into evidence. Allowing some evidence of the demonstration is inconsistent with the trial court’s ruling that the demonstration was not sufficiently similar to the accident scene to be admissible. We decline to speculate whether the circumstances on retrial will justify introduction of the videotape.
Plaintiffs remaining assignments require only brief discussion. She argues that the trial court erred in failing to order defendants to produce the entire contents of the file of one of their expert witnesses, which the witness had with him on the stand. Defendants claimed the attorney-client privilege. The witness testified that the particular file contained only correspondence with defendants’ attorney and that the correspondence took place before he was asked to testify for defendants. That kind of correspondence is within the attorney-client privilege and need not be disclosed. OEC 503(2)(b). The trial court did not err.
Finally, plaintiff assigns error to the trial court’s instruction to the jury on the basic speed rule.9 That instruction was not error. It was a correct statement of the *330law, and the speed at which the parties were travelling was at issue.
Reversed and remanded.
Defendants are Price, the driver of the truck and R. E. Christiansen Trucking, Inc., the corporate owner of the truck and Price’s employer.
The verdict stated:
“We, the jury, find:
“1. Were the defendants negligent in one or more of the respects claimed in the plaintiffs complaint that caused damage to plaintiff?
“ANSWER: NO”
We comment with some frustration on the common practice of using general verdict forms in complex cases such as this one. It may be that the jury accepted plaintiffs theory of the collision but concluded that, although the pup trailer swept into plaintiffs lane, defendants were not negligent. The use of special interrogatories would clarify what the jury concluded, as well as who, if anyone, was at fault. That information could save time and money for plaintiffs, defendants and the courts.
At trial, plaintiff objected:
“[Wlhat traffic does at other times and places is irrelevant and immaterial and has no probative value in this instance. * * * If other people at other times cut the comer, it doesn’t mean she would cut the corner just because of her personality, her make up, her driving habits and practices. * * * What other people do at other times in other circumstances is not relevant or material. It has no probative value as to what the two drivers did in this case.”
Plaintiff did not object that the probative value of Weaver’s testimony was outweighed by its prejudicial effect. Her objection was simply that the testimony had no probative value and therefore was not relevant and not admissible.
See Kirkpatrick, Oregon Evidence 111 (2d ed 1989).
In Carter v. Moberly, supra, the Supreme Court said:
“[E]vidence of negligence at some other time and place is inadmissible to prove negligence on the occasion in question * * * [unless there is] sufficient evidence to indicate that the [negligent conduct] continued v. to the time of the accident.” 263 Or at 198. (Citations omitted.)
Weaver testified, in part:
“DIRECT EXAMINATION
“ * * * *
“Q: [Do] you have an opinion regarding what happened in this accident?
“A: Yes, I do.
“Q: What is that opinion?
“A: My opinion is that the vehicle southbound at some point impacted the back trailer axles of the - - what was called the pup of the northbound truck.
“ * * * *
“I think it is more likely than not that it was on the northbound portion of the road.
“Q: Can you explain to the jury the basis for your opinion?
‘ ‘A: Part of it is when people are driving down a road they’re less perceptive of where the right side of their vehicles are, * * *. So when there is something on the right side and in this case there was a bridge there, there is more of a tendency for people to move away from that and in highway design we allow for this, and as a matter of fact it’s called a shy distance. It’s an unwillingness of people to test those limits and so I think it is more likely that when confronted with a solid object on one side and the center of the road on the other, that a person is likely to move away from that solid object.
“ * * * *
“RE-CROSS EXAMINATION
“Q: Let me clarify. Then your opinion is based not on the physical evidence of the damage to the vehicles * * * but it’s based on what you understand tc be the tendency of drivers at that location?
“A: To the tendencies of drivers at locations similar to this.”
The witness also testified that he had reviewed photographs of plaintiffs damaged vehicle and accident reports, but could not recall whether he had seen any photos of the tractor-trailer or the pup trailer or the maintenance records for the truck and pup trailer. He testified that neither the damage to plaintiffs car nor the damage to the tractor-trailer and pup trailer would affect his opinion that the collision occurred in the northbound lane.
See n 6, supra.
The videotape was offered as rebuttal evidence against defendants’ expert witnesses, who testified that trailer sweep could not occur under the conditions present at the time and place of the collision; that, if trailer sweep did occur, the driver would always feel it; and that, if trailer sweep were possible under the conditions at the time of the collision, the result would have been a ‘ ‘catastrophic loss of control” of the entire tractor-trailer rig.
The instruction was:
“In connection with the basic speed rule, I instruct you that the statutes of this state also provide for certain designated speeds. Speed in excess of the designated speed is a violation of the basic speed rule unless you are satisfied that such speed was reasonable under the existing circumstances.
“In this case, the applicable designated speed is fifty-five miles per hour in areas not otherwise described.”