OPINION

BRAUER, J.

On April 25, 1980, respondent Stephen Akers, while working as a loading dock supervisor at a freight terminal, sustained permanent brain damage when an adjustable dockboard flew apart and a portion of its mechanism struck him in the head. Through his conservator Akers subsequently brought suit for personal injuries against the manufacturer of the dockboard, appellant Kelley Company, Inc. (hereinafter Kelley). The action was founded on theories of strict product liability and negligence. After a lengthy trial, a jury determined by special verdict (a) that the dockboard was "defective in design and/or manufacture," and that the defect or defects proximately caused the injury; (b) that Kelley had been "negligent," and that such negligence proximately caused the injury; and (c) that Akers himself was without fault. The jury awarded Akers the sum of $3.5 million as compensatory damages. That sum was reduced by the amount of workers' compensation benefits provided by Akers' employer, and the net award amounted to $3,221,976.18. Judgment was entered accordingly, and thereafter Kelley's motions for new trial and for judgment notwithstanding the verdict were denied.

Kelley appeals from the judgment, and contends (1) that a manufacturer cannot be held liable for personal injuries when its product has been "broken" by a third party prior to its use by the injured party; (2) that the trial court committed reversible error in giving certain instructions and rejecting others; and (3) that members of the jury engaged in prejudicial misconduct. We find no error warranting a reversal, and we therefore affirm the judgment.

I. HISTORY

A. Preliminary Notes

The apparatus which caused the injury to respondent Akers is difficult to describe in words alone, and so we have attached as appendices to this opinion excerpts from a Kelley owner's manual for the model 625 dockboard, which manual was admitted in evidence as plaintiff's exhibit 2.

No witness actually observed the impact which was the immediate cause of Akers' injury. Testimony as to how that impact occurred took the form of opinions expressed by defense accident reconstruction experts. Akers produced no contrary evidence. Consequently, while we are obliged to view the evidence in the light most favorable to the prevailing party ( Jessup Farms v. Baldwin (1983) 33 Cal.3d 639, 660 [ 190 Cal.Rptr. 355, 660 P.2d 813]), our recital of the facts relies in part upon a defense theory, because that theory is the only one available.

B. The Parties

At the time of the accident, Stephen Akers was employed as a loading dock supervisor at the Santa Clara freight terminal of Delta Lines, Inc. He was responsible for coordinating freight loading operations at night. On the date he was injured, Akers had been an employee of Delta for approximately six months.

Kelley, the manufacturer of the dockboard which injured Akers, is a Wisconsin corporation having its principal offices in Milwaukee. Kelley first began manufacturing its model 625 dockboard in 1966. Several such dockboards were installed at Delta's Santa Clara terminal in 1967.

C. The Dockboard

The Kelley model 625 dockboard is essentially an adjustable platform designed to bridge the gap between a stationary loading dock and the rear end of a freight-carrying truck or van. The apparatus enables forklift trucks to move from a stationary loading dock, across the dockboard plate, into the rear end of a van, and back again, to load or to unload pallets of freight (see appen. A). The "board" of the dockboard is actually a reinforced steel plate, approximately 6 feet wide and 54 inches long, weighing perhaps 700 pounds. A hinged 16-inch steel "lip" is attached to the front (or truckside) edge of the plate; in operation this "lip" rests on the floor of the van (see appens. A and B).

Below the steel plate, running its full width, is a torque tube assembly containing a torsion bar. A double-beam lifting arm is attached to the torque tube assembly and rides against the underside of the steel plate (see appen. B). The torsion bar and the lifting arm provide the upward force which enables the front edge of the steel plate to be raised.

Several other forces tend to hold the steel plate in place and keep it from flying upward. We focus on two of those forces: (1) a "hold-down and float assembly," actuated by pulling a release cable, which controls the movement of the front (truckside) edge of the plate (see appen. A, figure 4, and appens. B and C); and (2) a "rear hold-down spring," designed to keep the rear (dockside) edge of the plate flush with the level of a stationary loading dock (see appens. B and C).

Welded to the underside of the steel plate's rear edge are two U-shaped hinge loops, which allow one or both rear corners of the steel plate to rise upward. The hinge loops allow the plate to tilt from side to side, to accommodate uneven or unlevel truck beds. As we shall see hereinafter, the hinge loops on the dockboard which injured Akers were broken and deformed before the accident happened.

D. Theory and Practice

The intended operating sequence of the Kelley model 625 dockboard is illustrated in appendix A. In theory, the rear hold-down spring would keep the rear edge of the steel plate flush with the level of the stationary loading dock. In practice, however, on some dockboards the rear edge tended periodically to "float" upward. The "float" phenomenon was "common enough so it wasn't unusual" at the freight terminal where Akers worked. When the rear edge of a dockboard plate rose above dock level, the plate was susceptible to being hit by forklift blades. Since safety regulations required forklift blades to be carried as close to the floor as possible, the result was that forklift blades collided with dockboard plates with some frequency. Several witnesses testified that before the accident such collisions had occurred at Delta's Santa Clara terminal, where Akers worked.

About a month before the accident one Troy Owens, a truck driver employed by Delta, noticed that the rear edge of dockboard No. 87 at Delta's Santa Clara terminal intermittently rose above dock level. Owens reported the condition to one of Delta's dispatchers, but did not inform Akers. Dockboard number 87 continued to be used thereafter; the record does not disclose that it was repaired or adjusted before the accident.

E. Events of April 25, 1980

Shortly before 7 p.m. on Friday, April 25, 1980, Ted Simmons, a forklift truck driver employed by Delta, was engaged in unloading pallets of freight from a truck trailer parked at door number 87 at the Santa Clara terminal. While unloading he drove his forklift truck across dockboard number 87 several times. Then, heading into the trailer to pick up another pallet of freight, a blade of his forklift struck the rear edge of the dockboard's steel plate, and his forklift truck stopped. Simmons backed up his truck, got off of it, and examined the dockboard; he testified that the "whole plate" was raised up "[a]pproximately a couple of inches or so." He felt that the dockboard was "obviously abnormal" and decided that it could not be used for further unloading. Simmons reported the situation to his immediate supervisor, Allen Selby; then he "punched out" and went home.

When Selby examined the dockboard a few minutes later, he saw that the plate was "cockeyed," i.e., the left rear corner was down an inch or two, and the right rear corner was raised a corresponding distance. In an effort to level the plate, Selby stepped onto it and flexed his knees. At that time Akers was standing two or three feet away, watching Selby. Selby's efforts produced no results. Selby decided "not to use that board anymore," and told Akers to have the trailer moved to another door and to finish the unloading there. Shortly thereafter Selby went home.

Here we pause to note that experts on both sides agreed that the forklift impact broke the welds on the rear hinge loops of dockboard number 87, and deformed the hinge loops from a "U" shape to a lopsided "C" shape. Defense experts went even further; they testified that the forklift impact also dislodged the rear hinge loops from their hinge pins, leaving the rear edge of the steel plate completely unattached. But those defects were not apparent to the casual observer. Simmons testified that in examining the dockboard he could not tell whether the steel plate was attached or detached. Selby testified that he could not see underneath the plate.

About 8 p.m. Frank Klinger, the terminal manager, returned to the terminal to check on the progress of freight movement. He knew nothing about the earlier forklift impact. He and Akers toured the terminal to determine how much work remained to be done. The trailer was still parked in front of door number 87. Klinger and Akers both walked across dockboard number 87 into the rear of the trailer, to inspect its remaining freight. As Klinger stepped on the right rear corner of the steel plate, that corner sank two or three inches and the left corner of the plate's front edge rose a corresponding distance. Klinger, who weighed 235 pounds, jumped on the left front corner, without result. He decided that the dockboard could not be used further because the steel plate was not level with the bed of the trailer. He told Akers to have the trailer moved to an adjacent door, "[a]nd that he was to secure the area and we would look into the necessary repairs or problem on Monday." Akers did not mention the earlier forklift impact. Klinger did not believe that the dockboard was actually "broken"; he thought it was "momentarily out of adjustment." Sometime later Klinger left the terminal and went home.

At Akers' instruction, the trailer was pulled away from door number 87. When this was done, the front edge of the steel plate remained in a raised position (see appen. A, figure 7), and the plate itself remained tilted. As a result the overhead door could not be completely closed and locked. It was one of Akers' duties as loading dock supervisor to close down and secure the terminal for the weekend.

The accident occurred somewhere between 10:30 and 11 p.m. Although no one actually saw him do so, it was the opinion of defense accident reconstruction experts that Akers, in an effort to lower or to level the steel plate, pulled the release cable attached to the front edge hold-down assembly. This action transformed the dockboard into a catapult. As the torsion bar and lifting arm exerted an upward force, the steel plate flew upward and forward, and came to rest upside down in the truck yard some 10 or 15 feet away from the loading dock. As the plate flew off, the double-beam lifting arm slammed back against the loading dock and struck Akers in the head. Moments later Akers was found lying face down, unconscious, with his head hanging over the edge of the loading dock. He was bleeding and, according to a witness, "[t]here was a big ball of hair right in the small of his back."

F. Revelations After the Accident

With the steel plate lying upside down in the truck yard, the broken welds and the deformed hinge loops were plainly visible. Photographs of the plate, the loops, and the welds, taken shortly after the accident, were admitted in evidence at trial.

It was also discovered that the rear hold-down spring had been stretched beyond its elastic limit, and that the eyebolt meant to secure the lower end of the spring had been deformed. The actual spring and eyebolt were admitted in evidence.

II. LIABILITY

(1a) Kelley's first contention is that it cannot be held liable for personal injuries caused by one of its dockboards where the dockboard in question had been "broken" by a third party prior to the accident. Implicit in this argument are the notions (a) that prior to the forklift impact the dockboard was not defective in either design or manufacture, and (b) that any "defect" in the dockboard which proximately caused injury to Akers was the product of the forklift impact itself. Akers, on the other hand, contends that the welds on the rear hinge loops and the rear hold-down spring were defective in both design and construction, and that forklift impacts, such as the one that occurred in this case, were foreseeable and could have been prevented by proper design and manufacture.

At this point we pause to summarize the views of the various engineering experts who appeared at trial. In a nutshell, their views were these:

Akers' experts testified (1) that the rear hold-down spring was of inadequate size, and of poor design in that it had open hooks at either end; (2) that the spring was stretched beyond its elastic limit in the course of normal use of the dockboard; (3) that when so stretched, the hook end of the spring became disconnected from the eyebolt; (4) that the eyebolt which anchored the lower hook of the spring was itself not properly secured by double nuts; (5) that the use of a different kind of spring would have prevented the rear edge of the steel plate from rising randomly upward; (6) that the welds of the rear hinge loops were defective in both design and fabrication; and (7) since the dockboard was meant to operate in proximity to forklift trucks, collisions between forklift blades and steel plates were entirely foreseeable. This testimony was buttressed to a certain extent by Robert Kuhns, Kelley's president, who initially testified that Kelley had anticipated that the rear edge of the steel plate would be ". . . smacked with . . . a fork lift blade in normal operation." Kuhns also testified that after the event Kelley engineers simulated the accident with a dockboard at the Kelley factory.

On the other hand defense experts testified (1) that the rear hold-down spring was properly designed and manufactured for its application; (2) that the spring could not have been stretched beyond its elastic limit in normal use of the dockboard; (3) that just before the forklift impact, the rear edge of the steel plate was flush with the dock level, and the forklift blade struck downward and lifted the plate; (4) that the forklift impact stretched the rear hold-down spring beyond its elastic limit and deformed the eyebolt; (5) that no weld could have withstood the forklift impact; and (6) that the force of the forklift impact that occurred in this case was entirely unforeseeable.

(2) Jurors may give an expert opinion only that weight to which they deem it entitled, and they may reject any such opinion if in their judgment the reasons given for it are unsound. ( Kastner v. Los Angeles Metropolitan Transit Authority (1965) 63 Cal.2d 52, 58 [ 45 Cal.Rptr. 129, 403 P.2d 385]; Hyatt v. Sierra Boat Co. (1978) 79 Cal.App.3d 325, 339-340 [ 145 Cal.Rptr. 47].) (3) Moreover, under ordinary circumstances foreseeability is a question of fact for the jury, and it may be decided as a question of law only if "`under the undisputed facts there is no room for a reasonable difference of opinion.'" ( Bigbee v. Pacific Tel. Tel. Co. (1983) 34 Cal.3d 49, 56 [ 192 Cal.Rptr. 857, 665 P.2d 947].) (1b) In this case the jury was faced with conflicting testimony. The jurors evidently accepted the opinions of Akers' experts, and concluded that the forklift impact was foreseeable. From our review of the record we cannot say that they were incorrect in doing so.

Kelley points out that dockboard number 87 was installed in 1967, and that the accident occurred in 1980, some 13 years later. But the mere passage of time is not in itself sufficient to break the chain of causation. ( Hale v. Depaoli (1948) 33 Cal.2d 228, 231-232 [ 201 P.2d 1, 13 A.L.R.2d 183]; Reynolds v. Natural Gas Equipment, Inc. (1960) 184 Cal.App.2d 724, 738 [ 7 Cal.Rptr. 879].)

(4a) Kelley further complains that Akers "was clearly negligent in attempting to operate the broken dockboard," and that the jury plainly ignored evidence of his fault. But as we have noted, the jurors were not bound to accept the expert opinion that Akers did indeed pull on the release cable. Furthermore, given the circumstances (a) that the dockboard was elevated and unlevel, (b) that the overhead door could not be completely closed and locked, and (c) that Akers had the responsibility of securing the terminal for the weekend, the jury could well have concluded that his pulling of the release cable (if such he did) was completely reasonable. Either way, the evidence does not disclose that Akers was "clearly negligent." When contributory negligence and assumption of risk were legal defenses, their presence or absence were questions of fact for the jury to decide. ( Ewing v. Cloverleaf Bowl (1978) 20 Cal.3d 389, 399 [ 143 Cal.Rptr. 13, 572 P.2d 1155].) (5) More recently, the existence of "comparative fault" has likewise been held to be a question of fact, to be decided by means of a special verdict form. ( Daly v. General Motors Corp. (1978) 20 Cal.3d 725, 743 [ 144 Cal.Rptr. 380, 575 P.2d 1162]; BAJI No. 9.06.) (6) When two or more inferences can reasonably be drawn from the facts, a reviewing court is without power to substitute its own deductions for those of the finders of fact. ( Crawford v. Southern Pacific Co. (1935) 3 Cal.2d 427, 429 [ 45 P.2d 183].)

(4b) Kelley also points out that Akers was warned by his superiors not to "use" the dockboard. But nothing in the record suggests that Akers or anyone else suspected that the dockboard had become a potential catapult. Both Selby and Klinger recognized that the damaged dockboard posed a hazard to forklift trucks and to merchandise, but neither realized that the dockboard might fly apart. The jury could well have determined that although Akers knew the dockboard was damaged, he did not and could not have been required to appreciate fully the magnitude of the risk. (See Buccery v. General Motors Corp. (1976) 60 Cal.App.3d 533, 550 [ 132 Cal.Rptr. 605].)

(1c), (4c) We conclude, therefore, that the jury's findings of (a) liability on the part of Kelley, and (b) absence of comparative fault on the part of Akers, are both supported by substantial evidence and ought not to be disturbed.

III. INSTRUCTIONS

Kelley next complains that the court below committed reversible error in giving certain instructions and in rejecting others. We examine the rejections first.

A. Instructions Not Given

(7) Kelley offered an instruction which read: "A manufacturer is not liable for injuries resulting from the plaintiff's use of a broken product after the plaintiff became aware of the fact that the product is in a broken and/or damaged state." This instruction was refused. Kelley claims error, arguing that the refusal to give the instruction prevented the jury from considering the "broken" condition of the dockboard. It relies on cases dealing with products "altered" after leaving the manufacturer's possession.

The vice of the proffered instruction is, of course, that it assumes that the dockboard in question had no inbuilt defects when it left the Kelley factory. But as we have seen, there was substantial evidence suggesting that the dockboard was initially defective in both design and manufacture. The jury could have determined from that evidence that the defect or defects caused both the forklift impact and Akers' injury. Consequently it would have been reversible error to instruct the jury that under the circumstances of this case Kelley could not, as a matter of law, be held liable for the injury. (8) Questions of causation, foreseeability, and reasonableness are questions of fact to be decided by the jury. ( Ewing v. Cloverleaf Bowl, supra, 20 Cal.3d at p. 399.)

(9a) Kelley next contends that the trial court should have instructed the jury to apply the "risk-benefit" test set forth in Barker v. Lull Engineering Co. (1978) 20 Cal.3d 413, 435 [ 143 Cal.Rptr. 225, 573 P.2d 443, 96 A.L.R.3d 1]. Kelley requested an instruction which incorporated both the "consumer expectation" and the "risk-benefit" tests; that instruction, the full text of which is set forth in the margin, was refused. Using a form of BAJI No. 9.00.5, the court instructed the jury to apply the "consumer expectation" test, and crossed out that portion of the form instruction relating to the "risk-benefit" test. According to Kelley, the refusal to apply the "risk-benefit" test prevented the jury from considering the "broken" condition of the dockboard. (10a) In addition Kelley argues that there can be no reasonable "consumer expectation" with regard to the performance of a "broken" dockboard.

There appears to be a difference of opinion in the Courts of Appeal as to whether the "consumer expectation" test is appropriate for the jury's consideration in a case where the product in question is one not readily available on a supermarket shelf. For example, Lunghi v. Clark Equipment Co. (1984) 153 Cal.App.3d 485 [ 200 Cal.Rptr. 387], was an action for wrongful death which arose when a front-end loader crushed the decedent. The trial court refused to instruct on the "consumer expectation" test, and instead instructed the jury to consider the "risk-benefit" test. The jury returned a verdict for the manufacturer. On appeal the plaintiffs claimed error in (among other things) the trial court's refusal to apply the "consumer expectation" test. On this point the First Appellate District, Division Three, found no error. It said: "[¶] Appellants themselves identify the subject of inquiry in the instant case as `whether the user of a loader would expect that boom arms and bucket could descend with fatal crushing force when the loader is at rest with the engine off.' This does not seem to be the kind of question that twelve ordinary people drawn from the community at large could evaluate based on their own `common knowledge.' The `user of a loader' is distinguishable from `ordinary consumers.' [Citation.] Frankly, we think that an ordinary consumer would not know what to expect from a piece of heavy machinery like the Bobcat loader, even if the engine was turned off." ( Id., at p. 496.) Similar language was used by the First Appellate District, Division Two, in Bates v. John Deere Co. (1983) 148 Cal.App.3d 40, 52 [ 195 Cal.Rptr. 637] where the product was a commercial cotton picking machine. On the other hand the Second Appellate District, Division Two, upheld the giving of an instruction which incorporated both the "consumer expectation" and "risk-benefit" tests in Fierro v. International Harvester Co., supra, 127 Cal.App.3d 862, 867 [ 179 Cal.Rptr. 923], where the product was a truck fuel tank; and the Fourth Appellate District, Division Two, upheld the giving of an instruction mentioning only the "consumer expectation" test in Grimshaw v. Ford Motor Co. (1981) 119 Cal.App.3d 757, 801-803 [ 174 Cal.Rptr. 348], where the product was an automobile with a badly placed fuel tank.

The difference of opinion seems to stem from our Supreme Court's opinion in Campbell v. General Motors Corp. (1982) 32 Cal.3d 112, 126-127 [ 184 Cal.Rptr. 891, 649 P.2d 224, 35 A.L.R.4th 1036], which opinion was filed after Grimshaw v. Fierro, but before Lunghi and Bates, supra. In Campbell, a 62-year-old female bus passenger was injured when a bus manufactured by the defendant turned a corner sharply, and she was propelled from her seat. In a subsequent product liability action, the plaintiff claimed that the bus had been defective in design because it lacked handrails or guardrails which would have prevented her fall. In her case-in-chief the plaintiff herself testified as to what had happened, and she presented photographs of the bus. At the close of her case-in-chief the trial court granted the manufacturer's motion for judgment of nonsuit. The Supreme Court reversed, holding that the plaintiff had presented sufficient evidence, under either the "risk-benefit" or "consumer expectation" test, to withstand a motion for nonsuit. Under the "consumer expectation" test, it was not necessary for the plaintiff to present expert testimony as to design defects. The court said: "[¶] Here, plaintiff presented sufficient evidence to have the case submitted to the jury on this theory as well. Not only did she testify about the accident (her use of the product), but she also introduced photographic evidence of the design features of the bus. This evidence was sufficient to establish the objective conditions of the product. The other essential aspect of this test involves the jurors' own sense of whether the product meets ordinary expectations as to its safety under the circumstances presented by the evidence. Since public transportation is a matter of common experience, no expert testimony was required to enable the jury to reach a decision on this part of the Barker inquiry." ( Id., at p. 126; fn. omitted.) The court noted that "it is difficult to conceive what testimony an `expert' could provide" on the issue of "the safety expectations of the general public as represented by the ordinary consumer, . . ." ( Id., at pp. 126-127.) The court concluded thus: "[¶] The quantum of proof necessary to establish a prima facie case of design defect under the first prong [i.e., the `consumer expectation' test] of Barker cannot be reduced to an easy formula. However, if the product is one within the common experience of ordinary consumers, it is generally sufficient if the plaintiff provides evidence concerning (1) his or her use of the product; (2) the circumstances surrounding the injury; and (3) the objective features of the product which are relevant to an evaluation of its safety. That evidence was provided in this case." ( Id., at p. 127.)

Kelley and our brethren in the First Appellate District read the foregoing quoted language to say that the "consumer expectation" test cannot be applied where the product is beyond the ken of the ordinary person on the street. We do not so read it. Instead, we read it to say simply that the plaintiff in that case was entitled to establish, and did establish, a prima facie case of failure to meet "consumer expectations" without the benefit of expert testimony, because public transportation is a matter of common experience. Our view is reinforced by Barker itself, where it was suggested that the "consumer expectation" test could be applied to a 23-foot-long 4-wheel-drive loader, weighing 17,500 pounds and equipped with forks similar to those of a forklift. ( Barker v. Lull Engineering Co., supra, 20 Cal.3d at pp. 419, 435.) The "consumer expectation" and "risk-benefit" tests are "alternative tests." ( Id., at p. 432; Curtis v. State of California ex rel. Dept. of Transportation (1982) 128 Cal.App.3d 668, 687 [ 180 Cal.Rptr. 843]; Grimshaw v. Ford Motor Co., supra, 119 Cal.App.3d at pp. 802-803.)

In our view the "consumer expectation" test is entirely appropriate in a case such as this one. There are certain kinds of accidents — even where fairly complex machinery is involved — which are so bizarre that the average juror, upon hearing the particulars, might reasonably think: "Whatever the user may have expected from that contraption, it certainly wasn't that." Here, a dockboard flew apart and injured Akers. A reasonable juror with no previous experience of dockboards could conclude that the dockboard in question failed to meet "consumer expectations" as to its safety. Of course, proximate cause was in issue; Akers' injury had to be shown to have been the proximate result of defective design or manufacture. For that reason Akers produced experts who testified that the dockboard had inbuilt defects at the time it left the hands of the manufacturer, and that forklift impacts should have been foreseen by the designer. This evidence warranted an application of the "consumer expectation" test.

(9b) The evidence did not support the giving of a "risk-benefit" instruction, because there was no showing that the benefits of the dockboard outweighed the risk of injury. Kelley produced evidence to the effect that the design of the model 625 dockboard posed no danger in normal use, and that the "use" in this case was entirely unforeseeable. But Kelley did not present evidence of other balancing factors in the "risk-benefit" test, such as evidence relating to "the mechanical feasibility of a safer alternative design, the financial cost of an improved design, and the adverse consequences to the product and to the consumer that would result from an alternative design." ( Barker v. Lull Engineering Co., supra, 20 Cal.3d at p. 431.) Instead, (1) Robert Kuhns, Kelley's president, testified that from 1969 onward Kelley eliminated rear hold-down springs in dockboards produced subsequent to the model 625, and (2) Akers' experts testified that the additional cost of a different design of spring and eyebolt, which would have prevented the accident, was "minimal," or "in the range of a dollar, or two, additional cost if it had been engineered that way at the time of the original construction." That evidence, which was not contradicted, indicated that a safer alternative design was not only feasible but economical. Consequently, even if the jury had been instructed to apply the "risk-benefit" test, it would have had little "benefit" evidence to balance against the risk of danger it found inherent in the dockboard's design.

Furthermore, the proposed "risk-benefit" instruction offered by Kelley (fn. 9, ante) included as relevant considerations (1) "whether the hazard was recognized by the industry"; (2) "whether the design was relatively more dangerous than that used on other dockboards"; and (3) "whether or not the danger posed by [the] design was one which would have been known to the designer of the dock board." The quoted portions would have injected into the case notions of industry custom and usage, and of the reasonableness of the manufacturer's conduct. Neither notion is relevant when applying the "risk-benefit" test. ( Barker v. Lull Engineering Co., supra, 20 Cal.3d at p. 434; Grimshaw v. Ford Motor Co., supra, 119 Cal.App.3d at p. 805.) In addition, the proffered instruction omitted any mention of the manufacturer's burden of proof under the "risk-benefit" test, and Kelley offered no other instruction on that subject. The proffered instruction therefore was inaccurate and misleading ( ibid.), and the trial court had no duty to give it. (11) In a civil action, a court has no duty to modify or to correct instructions offered by a party. ( Fierro v. International Harvester Co., supra, 127 Cal.App.3d at p. 869.)

(9c), (10b) For the foregoing reasons we conclude that the court below committed no error in instructing the jury in terms of the "consumer expectation" test rather than the "risk-benefit" test.

(12) Kelley's next contention is difficult to bring into focus. It appears that Kelley offered, and the trial court refused, an instruction which read: "[¶] If you find that the defects alleged to exist in the dock board did in fact exist but that the resulting injuries would have occurred even if the defects did not exist, the defects were not a substantial factor in bringing about the injuries and therefore were not contributing factors to the resulting injuries." This instruction evidently was based upon one offered by the defendant in Self v. General Motors Corp. (1974) 42 Cal.App.3d 1, 10 [ 116 Cal.Rptr. 575]. In his opening brief, Kelley's counsel characterizes the instruction as one dealing with "substantial factor" causation; but in his closing brief, counsel refers to it as one dealing with "superseding" causation. In any event, counsel claims that the trial court committed prejudicial error in rejecting the instruction.

In Self v. General Motors Corp., supra, a Chevrolet station wagon was struck in the rear by another car, and its fuel tank ruptured and caught fire. The plaintiff, a passenger in the right front seat, sustained severe burn injuries. In a subsequent action based on theories of negligence and strict product liability, one of the plaintiff's theories was that the fuel tank had been improperly placed in the left rear fender section of the station wagon. The theory of the defendant manufacturer was that no matter where the fuel tank might have been located, the station wagon would have caught fire in a vehicular impact of the magnitude that occurred, and the plaintiff's injuries would have been the same. The manufacturer offered, and the trial court refused, the following instruction: "`If you find that the gasoline tank in the 1962 Chevrolet station wagon was improperly located, but that the fire would have occurred even if the tank had been properly located, its location was not a substantial factor in bringing about the fire and was, therefore, not a contributing cause thereof.'" A jury returned a verdict in favor of the plaintiff and against the manufacturer. The Court of Appeal reversed, holding (among other things) that the trial court had committed reversible error in rejecting the manufacturer's proposed instruction. The manufacturer was entitled to have the jury properly instructed on its theory of the case. ( Self v. General Motors Corp., supra, 42 Cal.App.3d at pp. 10-11.)

Self is factually distinguishable from this case. Here Kelley's experts testified that the forklift impact stretched the rear hold-down spring beyond its elastic limit, and that no weld could have withstood that impact. That testimony was controverted by Akers' experts. But there was undisputed evidence to the effect that the dockboard did not fly apart until almost four hours after the forklift impact. Therefore that impact could not have been a superseding cause of Akers' injury; it was at best a concurrent cause of the accident, and the jury was properly instructed on the theory of concurrent causation. (See Hasson v. Ford Motor Co. (1982) 32 Cal.3d 388, 405 [ 185 Cal.Rptr. 654, 650 P.2d 1171], cert. dism. (1983) in 459 U.S. 1190 [75 L.Ed.2d 422, 103 S.Ct. 1167].) We perceive no error in the rejection of Kelley's proposed causation instruction.

B. Instructions Given

(13) Kelley next complains that the trial court failed to instruct the jury not to consider Akers' family in determining the amount of any award. We pause to examine pertinent events that occurred during jury deliberations.

The jurors began deliberating on December 22, 1981, the 23d day of trial. Later that same day they were given copies of all of the instructions. The court carefully admonished the jurors not to consider or to speculate about any deletions, and to give equal weight to both the handwritten and typewritten portions of the instructions. At the close of the third day of deliberations, December 24, the jury had not reached a verdict. Just before being excused for the day, one of the jurors asked the court the following question: "Judge, you said we were not to consider the family. Does that apply in this particular case?" The court responded: "I think I can answer the question this way. [¶] You have instructions including the instructions on damages. You should consider the elements of damage referred to in the instructions and they completely define the parameters of an award. [¶] That is all I can really tell you. Does anyone else need any further guidance or help in this regard?" Another juror asked: "You said that was in the instructions under damages?" The court replied: "No. You should confine yourself to the elements of damage recited in the instructions. [¶] In other words, the instructions that are in the jury room define the compensible elements of damage that are applicable in this case. [¶] That is all I can tell you."

It is Kelley's position that the quoted comments of the trial court amounted to an incomplete and inadequate admonition. We disagree. The court quite plainly said, "You should confine yourself to the elements of damage recited in the instructions," and further, that the instructions "completely define the parameters of an award." We have reviewed the damage instructions given, and we find nothing in them which remotely suggests that Akers' family should be considered in determining the amount of any award. Moreover, in reviewing the testimony of an economist witness called by Akers, we note that the actual award of $3.5 million was less than the combined total of Akers' pretrial and estimated future medical expenses. Consequently we cannot presume that the jury disregarded the trial court's admonition.

(14) On the subject of damages the trial court gave an instruction which combined the concept of "pain and suffering" with the concept of "loss of enjoyment of life." Akers' counsel submitted the instruction; we have set forth its full text in the margin. Citing Huff v. Tracy (1976) 57 Cal.App.3d 939, 942-944 [ 129 Cal.Rptr. 551], Kelley protests that the trial court committed prejudicial error in giving the instruction.

We agree that there was error. Huff v. Tracy, supra, holds: "[¶] The real question is . . . whether a court may instruct on loss of enjoyment of life in addition to or distinct from general damages. We conclude that it may not." The vice of coupling "pain and suffering" with "loss of enjoyment of life" is that the door is thereby opened to double compensation. ( Id., at p. 943.)

However, we are not persuaded that there was prejudicial error. The record discloses (1) that Akers sustained permanent brain damage; (2) that at the time of trial he was in a condition known to the medical profession as "coma vigil"; (3) that he was, and perhaps still is, virtually bedridden; (4) that he must breathe and be fed through tubes; (5) that he must be manipulated and exercised by others in order to prevent muscle atrophy and bedsores; (6) that he can neither talk nor walk; (7) that because of a permanent tracheotomy he is susceptible to various kinds of bronchial infections; and (8) that periodically he suffered seizures of unknown origin. Before the trial began, Akers had incurred medical expenses amounting to $265,550. According to an economist witness, at the time of trial the present value of Akers' future medical costs was $3,343,969, based upon a 20 percent reduction in normal life expectancy. Adding past medical expenses ($265,550) to projected future medical expense ($3,343,969) produces a total of $3,609,519, and that figure does not include any allowance for pain, discomfort, fears, anxiety, or other distress. In view of the injury, the jury's award of $3.5 million was well within reasonable bounds. Whatever may have been the erroneous instruction's potential for inciting double compensation, it does not appear from the record that the award would have been reduced if the phrase "loss of enjoyment of life" had been omitted from the instruction. (Cal. Const., art. VI, § 13; People v. Watson (1956) 46 Cal.2d 818, 836 [ 299 P.2d 243].)

IV. ALLEGED JUROR MISCONDUCT

In connection with its motions for new trial and for judgment notwithstanding the verdict, Kelley presented to the trial court a declaration signed by one of the trial jurors. Kelley's claims of juror misconduct are predicated in part upon that declaration. Because of its length, we set forth the full text of the declaration in the margin.

Kelley asserts that the trial court erred in refusing to consider the declaration. The record simply does not support the assertion. The trial judge not only said, "I have read the declaration over;" he also said, "I think the declaration has to be analyzed line by line, really, to determine whether it is violative of [section] 1150 of the Evidence Code;" and then he read portions of the declaration into the record. The trial judge must of necessity have considered the contents of the declaration in order to make the quoted comments.

(15a) Kelley next contends that given the contents of the declaration, the trial court erred in denying Kelley's motions for new trial and for judgment notwithstanding the verdict. In evaluating this contention, several principles come to mind. (16) First, not every insignificant infringement of the rules by a juror calls for either a mistrial or a new trial. ( Deward v. Clough (1966) 245 Cal.App.2d 439, 444 [ 54 Cal.Rptr. 68], hg. den. (1966).) (17) Second, the "statements made, or conduct, conditions or events relied upon as constituting misconduct must be objectively ascertainable and subject to corroboration." ( Andrews v. County of Orange (1982) 130 Cal.App.3d 944, 953 [ 182 Cal.Rptr. 176]; Rogers v. County of Los Angeles (1974) 39 Cal.App.3d 857, 863-864 [ 114 Cal.Rptr. 540].) (18) Third, a juror's declaration which purports to relate what he or she "felt" is incompetent evidence to impeach a verdict. ( People v. Sutter (1982) 134 Cal.App.3d 806, 819 [ 184 Cal.Rptr. 829].) Fourth, a declaration by one juror as to what other jurors "felt" is likewise incompetent evidence to impeach a verdict. ( People v. Flores (1979) 92 Cal.App.3d 461, 468-469 [ 154 Cal.Rptr. 851]; Clemens v. Regents of University of California (1971) 20 Cal.App.3d 356, 360 [ 97 Cal.Rptr. 589].) Fifth, a declaration which relates that not all of the evidence was reviewed, that some jurors were confused, and that not all jurors agreed with the ultimate verdict, has been held to be incompetent evidence to impeach a verdict. (See Silverhart v. Mount Zion Hospital (1971) 20 Cal.App.3d 1022, 1028-1031 [ 98 Cal.Rptr. 187, 54 A.L.R.3d 250], and Continental Dairy Equip. Co. v. Lawrence (1971) 17 Cal.App.3d 378, 385-387 [ 94 Cal.Rptr. 887].) (19) Sixth, while we are obliged to make an independent review of the record ( People v. Harris (1981) 28 Cal.3d 935, 948-949 [ 171 Cal.Rptr. 679, 623 P.2d 240], cert. den. (1981) in 454 U.S. 882 [70 L.Ed.2d 192, 102 S.Ct. 365]); Hilliard v. A.H. Robins Co. (1983) 148 Cal.App.3d 374, 409 [ 196 Cal.Rptr. 117]), we nevertheless "should accord great deference to [the] trial judge's evaluation of the prejudicial effect of jury misconduct." ( Andrews v. County of Orange, supra, 130 Cal.App.3d at pp. 954-955.)

(15b) With these principles in mind we examine the declaration. The text is a "mixed bag." It is apparent that it contains many subjective impressions ("I felt," "I was surprised that," etc.), as well as one glaringly speculative conclusion ("I felt that some of the jurors still kept it in the back of their minds," etc.). In addition, the declaration claims that the evidence was not examined closely, and that little of the expert testimony was discussed. These portions of the declaration are incompetent evidence.

The third full paragraph of the declaration (i.e., consisting of more than two lines) purports to relate comments made by individual jurors, apparently at the outset of the deliberations. Those comments, according to Kelley, clearly demonstrate a bias and prejudice in favor of Akers. We disagree. In pursuit of the ideal wholly unbiased jury, the law does not yet demand that jurors be automatons with minds like computers. Diverse points of view are both inevitable and expected; otherwise, there would be no point in selecting a jury in the first place. The law is yet without power to demand that jurors come to court without inbred bias, prejudice, or passion; it can only demand that in reaching a decision, the jurors should not be governed by any such bent. (BAJI No. 1.00.) Jurors are encouraged to discuss the case at hand. (BAJI No. 15.30.) The odd, or perhaps misguided, comment expressed by an individual juror in the early stages of deliberation does not necessarily provide an accurate clue as to how he or she reached a decision. Were we to adopt a practice of focusing upon such comments in order to upset a given verdict, we should thereby seriously undermine, if not completely destroy, the diversity of opinion, the untrammeled repartee, and the individual thought which make juries desirable in the first instance. We decline so to do.

The fourth full paragraph of the declaration reveals that the jurors debated about whether to include provision for Akers' family in the award. But a careful reading of that paragraph discloses that the debate apparently took place before the trial court's comments on the point, which we have described in part III B of this opinion, ante. The declarant's statement that he "felt" that other jurors "still kept it in back of their minds" is, as we have said, incompetent evidence.

The fifth full paragraph discloses that some jurors discussed an award in another case. We do not agree that the fact, if true, necessarily warrants a reversal. In this case the jury deliberated almost four full days. It is difficult to concentrate on the topic of proximate cause for an hour or two, let alone for four full days. We can readily appreciate that at least some jurors, seeking mental refreshment, might have turned temporarily to lesser and relatively uncomplicated topics such as politics, religion, the woeful status of a local baseball team, the weather, or even to a different case. On this point the declaration is more eloquent in what it does not say than in what it says. It does not say that the other case influenced the jury's ultimate decision.

In the sixth full paragraph the declarant avers that two jurors "used their own experience and knowledge gained before they were jurors in the case to say that the nut involved in the case was in their opinion not the correct nut for this application for the dock board." Kelley contends that this statement clearly shows that the two jurors imparted to the others information and knowledge not produced in the evidence. We do not so read the statement. The two jurors expressed opinions which were consistent with some of the expert testimony. The bald assertion that they formed their opinions "on their own experience and knowledge" is a conclusion, possibly based on speculation, and therefore incompetent evidence. (20) Furthermore, "`[i]n determining what is proper and what is improper discussion among jurors, regard must be had for the fact that the jury are supposedly men [and women] of different walks of life, avocations, and necessarily views that would be affected by their past experiences and situations. They could hardly arrive at a solution of their differences without discussion of the facts before them, and each man's discussion would necessarily be tinged or affected by his own viewpoint and experience.'" ( Wagner v. Doulton (1980) 112 Cal.App.3d 945, 950 [ 169 Cal.Rptr. 550], quoting from Frazer v. State (1924) 99 Tex.Crim. 89 [ 268 S.W. 164, 166].)

(21) Kelley further complains that the jury was predisposed to consider Akers' family in making an award, because (1) Akers' two daughters were allowed to attend portions of the trial, although neither testified, and (2) both evidence and argument brought out the fact that Akers had been an affectionate father. Kelley is correct in that ordinarily evidence of a plaintiff's family circumstances is inadmissible to prove the plaintiff's mental suffering. ( Smith v. Atchison etc. Ry. Co. (1919) 179 Cal. 611, 614-616 [ 178 P. 501]; Dillon v. Wallace (1957) 148 Cal.App.2d 447, 453-455 [ 306 P.2d 1044].) But in this case the ability (or lack thereof) of Akers to recognize his own children was relevant to establish the extent of his brain damage. The trial court made a considerable effort to limit the use of "family circumstance" evidence to that issue alone. We have already pointed out in part III B of this opinion, ante, that the jury's verdict awarded Akers less than the combined total of his past and estimated future medical expenses. If error there was, we cannot say that it was prejudicial.

V. DISPOSITION

Judgment affirmed.

Agliano, Acting P.J., and Leach, J., concurred.

A petition for a rehearing was denied November 21, 1985, and appellant's petition for review by the Supreme Court was denied February 20, 1986. Panelli, J., did not participate therein.