dissenting.
I believe that the jury had evidence before it that was sufficient to support its finding. The threshold required to sustain the jury verdict is very low, see Bonjorno v. Kaiser Aluminum & Chemical Corp., 752 F.2d 802, 811 (3d Cir.1984), cert. denied, — U.S. -, 106 S.Ct. 3284, 91 L.Ed.2d 572 (1986), and in my view was easily surpassed in this case. I would therefore reverse the district court’s grant of judgment n.o.v. and reinstate the jury verdict.
In this failure to warn case, as in all such cases, the plaintiff had to establish (1) that the allegedly required warning contemplated the situation that caused the injury to the plaintiff; and (2) that the warning would have been sufficient to change the plaintiff’s conduct. The sufficiency of the record in these two facets of the case are the subject of the majority’s analysis, and I take them up in turn.
I. WAS THE HARMFUL SITUATION CONTEMPLATED BY THE OMITTED WARNING THAT WHICH GAVE RISE TO THE INJURY?
The plaintiff advanced expert testimony that if the tire changer paused or moved in a jerky fashion for any reason, the situation was dangerous and could reasonably be foreseen to lead to injury. More specifically, the condition that the expert testimony identified as necessary to be warned against was a jerk or pause in the motion of the main tool, including the condition resulting from the demount tool snagging on the bead of the tire.1 In my view there was evidence before the jury that the machine jerked as the result of such a snag that occurred when Mr. Overpeck reversed the movement of the machine’s pedal.
The testimony before the jury reflected, inter alia, the following concerning the manner in which the tire changer operated. The Powerman 990 is a pneumatic tire changer designed to assist an operator in *758mounting and demounting automobile tires to and from the metal wheel. It is constructed in the form of a metal stand with a horizontal platform upon which a tire wheel is placed. Because of the design of the machine, the operator must lean over it, placing his head and shoulders directly above it.
Rising vertically from this platform is a threaded steel shaft around which is set the center aperture of the wheel and tire to be repaired or replaced. Also positioned in the center of the machine is a solid cast vertical cylindrical rotating shaft on the tongue (also called the “key”) on which is placed the mounting tool: a slotted eighteen-inch cast bar that, when properly positioned, sits not horizontally but at a forty-five degree angle. No mechanism secures the mounting tool to the center post key. Rather, Coats relies on the frictional forces exerted by the rotating center post to keep the mounting tool attached while the opposite end of the tool moves around the circumference of the tire bead. See Supp. App. at 105b-114b.
The situation we face concerns an interference with this frictional force at the point of operation that the operating instructions designate a “bonus feature,” which is an unusual step that sets the Coats Powerman 990 apart from other tire changers. As the instruction manual describes this unusual step:
Bonus feature: If by chance the tire is not completely mounted or demounted upon full rotation of the centerpost, all you need do is let up on foot pedal and tilt mount-demount tool free of the cen-terpost. After centerpost returns half a revolution, simply re-engage tool. You now may depress foot pedal for another full power stroke.
Nowhere do the instructions warn against letting up on the foot pedal without disengaging the mounting tool. If the tool remains on the centerpost key when the operator takes his foot off the pedal, however, it will not remain stationary, but will automatically attempt to return along the bead to its starting position.
The plaintiffs liability expert, Dr. James A. Currie, testified that there was a potential hazard with the operation of the air pedal because “the mount-demount tool when used in the mounting mode, has to be confined to its motion by a squeeze or a pinch produced between the rim and the bead and being twisted at the other end of that line by the center post.” Supp. App. at 28b-29b. The dangerous situation exists, he continued, when there is “any unevenness, jerkiness, out of adjustment, [or] shift of the pressures in such a way that the tool will change its ... contact points, [which] appears to be sufficient to allow that tool to then come free of its caught position and, indeed, to be able to fly off.” App. at 14-15; Supp. App. at 28b-29b. Thus, Dr. Currie was concerned with any pause or jerky motion, regardless of its cause.
Dr. Currie specifically drew the jury’s attention to the possibility that this pause or jerk would result from the demount tools nagging on the bead of a tire. “In the case of something catching or snagging,” he testified, “the forces that are there maintaining the bar[ ], can most easily ... become overcome and the bar can become free. If two hands are occupied, squeezing the tire toward the rim, that bar can go — fly off and go just about anywhere.” Supp. App. at 27b-28b.
Dr. Currie also called attention to the defendant’s failure to warn about this obviously dangerous situation, although it is true that he characterized it at one point as “the failure to call attention to the need for adjustment.” App. at 17. One of his concerns, it seems, was that “this machine can over time come out of adjustment and contribute to an uneven, jerky kind of rotation or movement of that tool, which would then contribute to its possibly becoming un-gripped from the bead and the center post and tire.” App. at 16. This concern, however, only reinforces the notion that any pause or jerkiness would produce a dangerous situation. The fact that one cause of such a dangerous pause could be ill adjustment of the machine does not suggest that ill adjustment is the only source. As demonstrated above, Dr. Currie was concerned *759with any pause, including the possibility that it would result from a snag.
Indeed, one of the defendant’s liability experts, Mr. Jimmy Holladay, reiterated Dr. Currie’s concerns, testifying on cross-examination that any jerky motion would be a problem:
Q: [Wjouldn’t I be correct to say that it’s important to Coats that the machinery move smoothly; isn’t that correct?
A: Yes.
Q: You don’t want any jerky movements?
A: That’s correct.
Q: Now, jerky movements for whatever reason can be dangerous, isn’t that correct?
A: It's possible.
App. at 18. Another of defendant’s liability experts, Robert Houston, confirmed on cross-examination that if the motion is not smooth for any reason, it could be dangerous. Supp. App. at 225b.
As I read the record, the jury could have found that, during the unusual “bonus step,” the mounting tool snagged on the bead of the tire that Mr. Overpeck was mounting and that this caused the jerk in the machine that led to the accident that so gravely injured his eye. Because the debilitating accident happened so quickly, as all the witnesses to the incident testified, it was difficult to prove by direct evidence that operation of the machine was not smooth or fluid. Nevertheless, there was testimony that the demount tool flew off when Mr. Overpeck reversed the motion of the machine’s pedal. The video tape deposition of Dr. Currie, which was played at trial, provided the jury with additional (albeit circumstantial) evidence on this point, for Dr. Currie drew the connection between a reversal of the pedal and the fateful pause. Dr. Currie describes what happened as follows: “He was on the mounting end of it, and he got to a point in his operation where he did something with the pedal, probably let off on the pedal and at the same instant that he did that, the.... bar or tool came up and hit him.” Supp. App. at 7b.
Further evidence that a reversal in the pedal movement would lead to a pause or jerk in the motion of the machine was provided in the cross-examination testimony of defendant’s own liability expert, Mr. Holladay. He testified that the tool could have been snagged between the rim and the tire upon reversal of the pedal, and that such a snag would cause a pause that could jerk the tool off when the machine’s pressure resumed.2 Holladay’s cross-examination further revealed that an intra-office memorandum had previously been circulated by Coats’ Chief Engineer that when a tire would temporarily bind the tool and then break lose, the tool would tend to be thrown off the rim. Supp. App. at 176(b)-177(b). In my view, the foregoing constitutes sufficient evidence for the jury to have determined that the reversal of the pedal movement caused a snag which in turn caused a jerk which threw off the tool into Overpeck’s eye.
The majority’s characterization of the facts rests on little more than semantics concerning what constitutes “any unevenness, jerkiness ... [or] shift in pressures,” Supp. App. at 28b, which would cause the machine tool to be projected upwards towards the machine operator. Examining the testimony of the diverse witnesses, the *760majority finds that they “never suggested that the mounting tool operated in an uneven or jerky manner at the time of the accident.” This account conveniently fails to regard the snagging of the mounting tool between the bead and the rim as a source of the fateful jerk. No one disputes the fact that the tool flew off the machine and into Overpeck’s eye: reconstruing the circumstances of this event, there was expert testimony that this was caused by the tool binding or snagging between the bead and the rim as Overpeck released the pedal and the machine continued with its “bonus step.” Rather than countering this testimony, the defendant’s experts testified consistently with it. True, no experts invoked the word “jerk.” But it was certainly permissible in light of all the evidence, including the reconstruction of experts, for the jury to infer that the machine jerked as the result of a snag that occurred as Mr. Over-peck reversed the movement of the machine’s pedal, and that as the result the demount tool was propelled into his eye. Therefore, there was ample evidence from which the jury could have concluded that the accident happened in the very way contemplated by the warning advanced by the plaintiff as necessary to make the product safe.
II. CAUSATION: THE EFFECT OF THE WARNING
In granting judgment n.o.v., the district court found that “there was not sufficient proof of causation to allow the case to go to the jury.” 634 F.Supp. at 640. For the reasons that follow, I believe that the district court placed too high a burden on Overpeck, misstated Pennsylvania law concerning causation in a failure to warn case, and thereby gave too little respect to the role of the jury as fact finder. Moreover, I find sufficient evidence in the record for the jury to have found as it did on causation.
In Webb v. Zern, 422 Pa. 424 (1966), Pennsylvania adopted Section 402A of the Restatement (Second) of Torts as its formulation of strict tort liability. Comment j to that section, also specifically adopted by Pennsylvania, see Incollingo v. Ewing, 444 Pa. 263, 287 (1971); see also Fravel v. Suzuki Motor Co., Ltd., 486 A.2d 498, 501 (Pa.Super.1984) (relying on comment j), addresses the failure to warn situation and provides that, “[wjhere warning is given, the seller may reasonably assume that it will be read and heeded.” The implications of comment j’s presumption is clear. “Such a presumption works in favor of the manufacturer when an adequate warning is present. Where there is no warning, as in this case, however, the presumption that the user would have read an adequate warning works in favor of the plaintiff user.” Technical Chemical Co. v. Jacobs, 480 S.W.2d 602, 606 (Texas 1972) (construing Restatement (Second) of Torts § 402A, comment j).3
The reason for this presumption is clear. Proof concerning the hypothetical future state of mind of the user, which requires evidence concerning the situation had the omitted warning been given, is all but impossible to adduce in the first instance.4 Comment j, presuming causation in order to avoid hypothetical debate, is thus one of the “instances in the law where rules of right are founded upon the inherent and inevitable difficult or impossibility of proof.” Walker v. Great N. Ry. of Ire., 28 L.R. Ir. 69, 81 (Q.B.1891) (O’Brien, J., concurring). The comment j presumption therefore sensibly obviates the need for courtroom hypothesizing as a substitute for proof.
This presumption is not conclusive and may be rebutted by the presentation of evidence to the contrary. “When the opponent of the presumption has met the burden of production thus imposed ... the *761office of the presumption has been performed; the presumption is of no further effect and drops from the case.” Commonwealth v. Vogel, 440 Pa. 1, 17, 268 A.2d 89 (1970). (quoted in Philadelphia to use of DePaul v. Philadelphia Authority for Industrial Development, 230 Pa.Super. 226, 230-31, 326 A.2d 502 (1974)). As Pennsylvania has formulated the effect of this rebuttable presumption in failure to warn cases, “the question of causation is generally for the jury, [unless] the relevant facts are not in dispute and the remoteness of the causal connection between the defendant’s negligence and the plaintiffs injury clearly appears.” Liney v. Chestnut Motors, Inc., 421 Pa. 26, 29, 218 A.2d 336 (1966) quoted in Greiner v. Volkswagenwerk Aktiengesellschaft, 429 F.Supp. 495, 497 (E.D.Pa.1977) (Lord, C.J.); see also Powell v. J.T. Posey Co., 766 F.2d 131, 134 (3d Cir.1985) (citations omitted). Only upon the defendant’s undisputed demonstration of sufficiently remote connection does the question become one of law for the court to determine without the jury. At that point, the jury could no longer draw a reasonable inference of causation, but would be indulging in pure conjecture. See Greiner, at 498.
The Pennsylvania cases, on which the district court and appellee rely, involve the situations where the absence of any possible causal connection has been proven and a judgment n.o.v. would be appropriate. For example, an independent intervening causal agent was indisputably proven in Powell v. J.T. Posey Co., 766 F.2d 131 (3d Cir.1985). In Powell, the plaintiff complained of injuries sustained after a patient escaped from a restraining vest manufactured by the defendant. She predicated her theory of liability upon the manufacturer’s failure to warn about possible escapes. In defense, the manufacturer argued that it was the hospital’s policy to use the vest on all such patients; because the plaintiff would thus not have been able to change her actions if a warning had been supplied, the lack of a warning could not be a cause of the harm. Because this hospital policy was not disputed by the plaintiff, the evidence was undisputed and demonstrated a supervening causal agent. Cf. Liney, 421 Pa. at 29, 218 A.2d 336. Explicitly finding that the plaintiff “has represented no evidence to the contrary,” Powell, 766 F.2d at 134, the court held “that the manufacturer is entitled to judgment in its favor as a matter of law because the plaintiff has failed to prove that the existence of any warning would have caused the plaintiff to act any differently than she did.” Id.
A similar case of supervening causal agency was presented in Greiner v. Volkswagenwerk Aktiengesellschaft, 429 F.Supp. 495 (E.D.Pa.1977). In Greiner, the plaintiff’s automobile overturned when the plaintiff swerved to avoid a collision in an emergency. The court rejected the plaintiff’s failure to warn contention because “a serious accident was inevitable, warning or no warning,” and the plaintiff would not have risked the peril of an accident to follow a warning. Id. at 497. Although the court summarized four cases where recovery was allowed despite the absence of specific causal evidence, it found them all distinguishable. In the cases presuming causation, “it was reasonable to infer from the evidence that the simply, easily understood directives and limitations could and would be followed had they been given ... No such inference can be made here because under the circumstances of this case, a warning, even if read, could not have been heeded.”
A judgment notwithstanding the verdict may also be appropriate when the omitted warning would have conveyed information of which the plaintiff was previously and indisputably aware. In Conti v. Ford Motor Co., 743 F.2d 195 (3d Cir.1984), the plaintiff complained that the manufacturer of his standard transmission automobile did not warn him that starting the car without disengaging the clutch would cause the car to lurch. Such a lurch had injured his wife. Because the plaintiff testified to his knowledge that in “driving a standard transmission you would have to depress the clutch,” the court held that “the jury’s conclusion was based on the mere speculation that if Mr. Conti’s eyes had caught a sticker warning him to depress the clutch pedal *762as he was talking to his wife, he may have remembered to either shift the car to neutral or to disengage the clutch prior to starting the car.” Id. at 198-99. Because of his own testimony, it was undisputable that the plaintiff knew of the danger that came to pass, but for which no warning was given.
Similarly, Sherk v. Daisy-Heddon, 498 Pa. 594, 450 A.2d 615 (1982), involved undisputed prior knowledge. The administra-trix of an estate sued the manufacturer of a B-B gun after a child was killed by an accidentally inflicted shot to the head. She argued that the manufacturer should have warned of the lethal danger involved with the gun. The Supreme Court of Pennsylvania found that uncontroverted evidence established that the user fully knew the dangers involved. “Where, as here, the lethal propensity of a gun was known ... to the user, liability cannot be imposed upon the manufacturer merely because the manufacturer allegedly has failed to warn of that propensity.” Id at 618, 498 Pa. 594; see also id. at 617-18, 498 Pa. 594 (direct testimony establishing knowledge).
The district court in this case did not find that the manufacturer presented comment j evidence that so conclusively rebutted the presumption that the issue should have been taken from the jury. Rather, it found that Mr. Overpeck had an initial burden to demonstrate causation that he had not shouldered. “[Wjithout evidence that these warnings would have prevented the harm,” wrote the court, “to permit the jury to draw such an inference allows a jury impermissibly to base its verdict on mere speculation.” 634 F.Supp. at 641. This was an incorrect interpretation of the requirements of 402A and constitutes a basis for reversal. First, the presumption itself acted as a surrogate for the initial burden. Second, apart from the presumption supplied by comment j and contrary to the district court’s conclusion, a review of the record reveals that sufficient evidence was presented to the jury to support the conclusion it rendered concerning causation.5
Mr. Overpeck had operated tire changers on many occasions and without prior incident, see Supp. App. at 61B, 74B, thus demonstrating his ability to understand and conform his actions to those instructions communicated to him. The fact that Mr. Overpeck was not troubled when he did not receive an owner’s manual with the tire changer is not conclusive,6 for this fact does not demonstrate that Mr. Overpeck generally does not follow directions that are given to him, or that he would not have heeded had an adequate warning on the machine had it been given to him. Indeed, because the evidence demonstrates that Mr. Overpeck properly operated the tire changer without referring to the manual, the jury had sufficient evidence to conclude that Mr. Overpeck was sensitive to information generally available to the community of tire changer users. Hence, the jury properly reasoned that, had any warning been provided, Mr. Overpeck would have followed it, just as he followed the other instructions appellee had provided concerning the use of the tire changer. In sum, the majority is wrong to suggest that the mere fact that Mr. Overpeck was generally knowledgeable about the operation of tire changers imports that he necessarily would have ignored warnings. The jury’s common sense told it that Overpeck’s common sense would not have let him ignore them.
The majority finds that this case falls under the rule of Conti and Sherk, that “[a] warning that advised Overpeck of the possibility of the mounting tool’s flying off during uneven or jerky operation would have provided him with no new information *763and thus would not logically have affected his behavior.” Maj.Op., typescript at 755. Under the facts of this case, I must disagree. In the first instance, the record presents contradictory evidence as to whether Overpeck actually kept his hand on the mounting tool; indeed, closer to the time of the accident than his testimony, Overpeck told the man from whom he bought the machine that “he didn’t have his hand on the bar.” Supp. App. at 63 (testimony of Gerald Tittle). This conflict is properly for the jury to resolve. Moreover, Overpeck’s statements at trial make it clear that he did not feel it necessary to keep his hand on the bar because he thought it would fly up into his face. Although he did at one point express concern about taking his hand off the tool, the testimony makes clear what he thought would happen:
Q: Suppose you let go with your hand?
A: You don’t let go. You don’t let go with your hand.
Q: Suppose you just put it between the end, between the rim and the tire and the slot on the center post, what would hold it in place or would it fall on the floor?
A: It would probably fall forward.
Q: Even though the end is between the rim and tire?
A: I don’t know. I never tried it.
Supp. App. at 48.
Finally, there is no evidence that Over-peck was specifically aware of the dangers of a jerky movement. In contrast to the record in either Conti or Sherk, there was no direct, indisputable evidence that the plaintiff knew of the defective condition that resulted in his injury. While Overpeck may have taken steps to keep the unsecured mounting tool from falling forward from the pinch of the tire bead, no testimony established that he was aware of the hazards of jerky motion during the mounting procedure.
In my view the jury could easily have concluded that Overpeck was unaware of the dangers of jerky movement and that, had he been adequately warned, he would have been on guard against a jerky movement such as occurred. Overpeck could have taken steps to secure the tool onto the machine, positioned himself over the machine in a way to avoid the path of a flying tool,7 or stood ready to react instantly in order to avoid it. In sum, it would have been eminently reasonable for the jury to have concluded that Overpeck would have been on his guard had the machine explained the hazards of jerky movement.
Reasonable minds may certainly differ over the conclusion to be reached by the evidence presented in this case. One might believe that, because of Mr. Overpeck’s long experience with tire changers, he would have disregarded any warning that contradicted his experience with such machines; one might also believe that failure to obtain an owner’s manual demonstrates a disrespect for all instructions in general. Similarly, the evidence might support the majority’s inference that Overpeck kept his hand on the machine out of fear that the mounting tool would fly up into his face. However, there is also evidence to the contrary, and the fact that reasonable minds can differ as to the proper interpretation of the evidence only reinforces the contention that the issue is one for the jury to decide with reference to its collective common sense and insight into human nature, born of experience and in light of its personal observations of Mr. Overpeck. “Whatever the complexity, if enough evidence is adduced to support a reasonable inference that defendant’s conduct played a part in the result, the issue is for the jury.” Green, The Casual Relation Issue, 60 *764Mich.L.Rev. 543, 561 (1962). In my view there was such evidence here, and the jury’s solemn and considered verdict was much too lightly set aside by the district court. I respectfully dissent.8
. Because the plaintiff never looked at the instructions (although he did proceed according to the manufacturer’s directions), the question arises whether the warning would necessarily have to have been on the machine itself. However, I need not address the ramifications of this question, or even discuss the caselaw that holds that a manufacturer may not escape liability if it places a warning within an owner's manual, because the district court assumed that the warning would have to have been on the machine and instructed the jury in the charge to make that assumption. See App. at 5. Moreover,, the jury had evidence before it concerning the necessity of the placement of the warning on the machine itself. Dr. Currie, the plaintiffs liability expert, testified that a warning needed to be placed right on the machine. See Supp. App. at 30b. Even Dr. Robert Houston, one of the defendant’s liability experts, testified on cross-examination that a warning should have been placed on the machine itself. See App. at 31-32. Therefore, under the facts of this case, no weight can be given to the argument that the instructions might have been hidden in the manual to which Overpeck did not refer.
. Question: If you think along with me for a moment, if it is caught in between the tire and the rim.
Answer: Yes.
Question: I take my foot off the pedal, this key is then driving this took back this way; is that correct?
Answer: No, we’ve taken the — we are taking our foot off the pedal the key we still have is attached to the center post.
Question: Yes.
Answer: It would try to drive it back.
Question: It would try to drive it back?
Answer: Yes, with the spring tension.
Question: The spring tension would try to pull this tool back this way.
Answer: Try to, yes.
Question: If it were caught between the rim and the tire it wouldn't be able to do that, is that correct?
Answer: That’s correct.
Question: And isn’t that consistent with exactly what Mr. Overpeck described. Isn’t it consistent with what he described?
Answer: Yes it is.
. I find hollow the majority’s suggestion that Pennsylvania would not interpret comment j just as Texas did. The interpretation is logical and in the mainstream of Restatement 402A jurisprudence, which Pennsylvania follows. Moreover, the Pennsylvania cases themselves strongly suggest that Pennsylvania would follow it.
. Alternatively, the user’s own testimony is arguably self serving to the point of banality and uselessness.
. To support its conclusion that "[t]he record is devoid of any evidence directly establishing or supporting a reasonable inference” that Over-peck would have acted differently had an adequate warning been provided, 634 F.Supp. at 640, the district court engaged in a weighing of the evidence. See id. at 640-41. However, weighing evidence is an impermissible vehicle for rendering a judgment n.o.v., especially where, as here, the evidence was not viewed in a light most favorable to the party against whom the motion was made. See generally C. Wright & A. Miller, Federal Practice and Procedure, ¶ 2524, at 541-42 (1971).
. Interestingly, no warning was even offered in the owner’s manual.
. Overpeck’s trial testimony demonstrates that he was not positioned in anticipation of a flying tool:
Q: Could you show me exactly how your face was positioned?
A: Standing like this here, facing the machine when I was operating (indicating.)
Q: How far would you estimate that your face is from that combination tool?
A: I’d say a foot and a half.
Supp. App. at 816. Because of this body position, the jury had additional evidence from which to conclude that Overpeck was not expecting the mounting tool to fly upward with substantial force.
. Because of its disposition of the case, the majority did not reach the district court’s alternative grant of a new trial. In view of my proposed disposition, I must reach it. I would reverse on that point as well.
The district court’s statement on the new trial issue was succinct:
The court would, in the alternative, grant a new trial on two grounds:
(1) Highly prejudicial comments of plaintiffs’ counsel may have biased the jury on both liability and damages; and
(2) The jury”s answers to interrogatories No. 2 and No. 4 were against the weight of the evidence.
At trial on recross-examination of defendants’ witness, Jimmy Holladay, plaintiffs’ counsel brought to the attention of the jury a plaintiffs verdict of $200,000.00 in Forsythe v. Coats, where Coats was the only one of several defendants and prevailed on appeal. (Tr. August 22, 1985 at 122). Coats actually paid only $3,500.00 in that case (Tr. August 22, 1985 at 124), but the case was introduced to the jury as if Coats were held liable for $200,-000.00 (Tr. August 22, 1985 at 122). The court attempted to mitigate the prejudice to defendants by these comments (Tr. August 22, 1985 at 122-27), but is convinced that the effort was unsuccessful, particularly in light of the jury’s finding plaintiff Howard Overpeck was damaged in an amount identical to the award in Forsythe.
634 F.Supp. at 641 (emphasis in original).
I have already addressed the court’s second point by explaining why I believe that there was ample evidence to support the verdict. As to the supposed prejudicial material, not only was it not objected to by defense counsel, but the door to its introduction was opened wide by him. The court’s alternative grant of a new trial must have come as a surprise to defense counsel; at all events, it was not warranted under the circumstances. Moreover, in view of the evidence of Overpeck's extremely serious injuries, I see no prejudice; it may indeed have been Overpeck who was prejudiced, for his injuries could have supported a larger verdict.