Opinion for the court filed by Circuit Judge BRYSON. Dissenting opinion filed by Circuit Judge PLAGER.
BRYSON, Circuit Judge.Michael and Carolyn Lampe appeal from the decision of the Court of Federal Claims denying their request for compensation under the National Childhood Vaccine Injury Act of 1986, 42 U.S.C. §§ 300aa-l to 300aa-34 (Vaccine Act), for injuries suffered by their child Rachael.
Early in her life, at about the time she received the standard series of Diphtheria-Pertussis-Tetanus (DPT) vaccinations, Rachael began to experience seizures. By the time she was five years old, she suffered from frequent seizures and was mentally retarded. The Lampes petitioned for compensation under the Vaccine Act. In the proceedings, the Chief Special Master of the Court of Federal Claims conducted two hearings. At the first hearing, the special master investigated whether the onset of Rachael’s seizures had occurred within the time period established by the Vaccine Injury Table, 42 U.S.C. § 300aa-14(a). Following the hearing, the special master determined that the onset of her seizures had not occurred within that period. At the second hearing, the special master heard evidence to determine whether the DPT vaccine had caused Rachael’s disabilities. Following that hearing, the special master concluded that the Lampes had failed to prove that any of Rachael’s DPT vaccinations had caused or significantly aggravated her condition. Accordingly, the special master dismissed the Lampes’ petition. The Court of Federal Claims reviewed the special master’s ruling and sustained it. We affirm.
I
Rachael Lampe was born on September 15, 1975. She received DPT vaccinations on November 10, 1975, December 18, 1975, and January 19, 1976. During the course of the immunization series, Rachael had episodes of rhythmic jerking of her right leg, which were later described in Rachael’s medical records as “bicycle-pedaling movements.” Rachael’s parents did not report the incidents of leg jerking to her pediatrician at the time, and her parents later were unable to establish exactly when those incidents began.
On January 26, 1976, seven days after her third DPT vaccination, Rachael suffered a seizure during which she became stiff, lost consciousness, and exhibited jerking motions on the right side of her body. She was taken to a hospital, where her parents said that the seizure had lasted five minutes, followed by a 30-minute period of limpness and unresponsiveness. A month later, Rachael suffered another seizure, and in the ensuing years she suffered numerous additional seizures, one of which occurred the day following her DPT booster shot in January 1977. Although she developed normally for some time, Rachael began to show signs of developmental impairment by the time she was 27 months old, and by age five her development had deteriorated substantially. She now suffers from a residual seizure disorder and profound mental retardation.
Rachael’s parents petitioned for compensation under the Vaccine Act. The petition alleged that the three DPT vaccinations *1359she received in 1975 and 1976 were followed “[w]ithin hours of each administration” by “injection-site leg stiffening and jerking, bicycle-type pedaling motions, and screaming episodes,” which ultimately culminated in her seizure disorder and subsequent mental retardation. Their petition was assigned to the Chief Special Master of the Court of Federal Claims. In the proceedings before the special master, the Lampes offered evidence that Rachael had experienced a jerking episode on the same day that she had her first DPT vaccination. Based on their evidence that Rachael’s symptoms developed within three days of her vaccination, the Lampes argued that their claim was governed by the Vaccine Injury Table, which presumes a causal relation between a DPT vaccination and a residual seizure disorder if the onset of the condition occurs within three days after the vaccination. See 42 U.S.C. §§ 300aa-ll(c)(l)(C)(i),.300aa-14(a).
After the first hearing in this case, the special master rejected as not credible the Lampes’ evidence that Rachael had experienced a seizure and other strong reactions within three days of her first DPT vaccination. The special master found that Rachael began to exhibit the bicycle-pedaling movements sometime between her second and third DPT shot. In light of the vagueness of the references to the dates of the bicycle-pedaling movements, however, the special master could not pinpoint the onset of those movements and therefore could not find by a preponderance of the evidence that those movements began within three days of any of Rachael’s DPT vaccinations. The special master noted that it was undisputed that the seizure that resulted in her hospitalization had occurred more than three days after her third DPT vaccination. Accordingly, the special master ruled that the Lampes had failed to prove that any of Rachael’s seizure activity had occurred within three days of any of the three DPT vaccinations. Because the Lampes failed to show that any of Rachael’s seizure activity occurred within ■ the three-day period specified in the Vaccine Injury Table, they were not accorded the benefit of the statutory presumption of causation applicable to so-called “Table injuries.” The special master therefore ruled that in order to establish their entitlement to compensation under the Vaccine Act, the Lampes would have to prove by a preponderance of the evidence that one or more of the DPT vaccinations caused Rachael’s condition: See 42 U.S.C. § 300aa-ll(c)(l)(C)(ii).
In support of their theory of actual causation, the Lampes submitted evidence from two experts who concluded that the DPT vaccinations she received as an infant had caused her seizures. The government submitted evidence from an expert who concluded that it was not possible to determine that the DPT vaccinations had caused Rachael’s seizure disorder or had significantly aggravated a pre-existing seizure disorder. Following the second hearing in this case, the special master filed a lengthy opinion in which he concluded that the Lampes had not met their burden of proving that it is more likely than not that Rachael’s seizure disorder and mental retardation were caused or significantly aggravated by the DPT vaccinations she received.
The Lampes sought review of the special master’s decision in the Court of Federal Claims. After reviewing the record, the court concluded that the special master had properly considered the relevant evidence and had not made a clear error of judgment in the case. Accordingly, the court affirmed the special master’s decision denying compensation to the petitioners.
II
The Vaccine Act provides two means for establishing eligibility for compensation. See, e.g., Munn v. Secretary of Health & Human Servs., 970 F.2d 863, 865-66 (Fed.Cir.1992) (describing in detail the statutory framework of the Act). This court has characterized the two routes to establishing eligibility as follows:
*1360One route is easy, as far as evidentiary proof goes. Bring the case within the timetable and specifications of a Table Injury and the statute does the heavy lifting — causation is conclusively presumed. Failing that, the heavy lifting must be done by the petitioner, and it is heavy indeed.
Hodges v. Secretary of Health & Human Servs., 9 F.3d 958, 961 (Fed.Cir.1993). The Lampes do not challenge the special master’s conclusion that they did not establish by a preponderance of evidence that the onset or aggravation of Rachael’s condition occurred within the three-day period set by the Table for presumptive causation. They do contend, however, that they met their burden of proving actual causation.
The Lampes challenge the special master’s decision in several respects, but the basic theme of their appeal is that the special master erred in concluding that they failed to prove that the DPT vaccinations caused Rachael’s seizure disorder. The special master’s resolution of the proof question turned largely on his decision to credit the evidence given by the government’s expert, Dr. Snyder, rather than the evidence given by the Lampes’ experts, Dr. Conkling and Dr. Lewis. On that issue, judicial review of the special master’s decision is very limited. The statutory standard of review applicable to the factual findings of a special master in a Vaccine Act case requires the Court of Federal Claims to uphold a special master’s findings unless the court concludes that those findings are arbitrary or capricious. See 42 U.S.C. § 300aa-12(e)(2)(B); Saunders v. Secretary of Health & Human Servs., 25 F.3d 1031, 1033 (Fed.Cir.1994); Munn, 970 F.2d at 870 & n. 10 (noting that the arbitrary and capricious standard is “well understood to be the most deferential possible”). In reviewing a ruling by the Court of Federal Claims that a special master’s findings of fact were not arbitrary and capricious, this court exercises de novo review. See, e.g., Bradley v. Secretary of Health & Human Servs., 991 F.2d 1570, 1574 (Fed.Cir.1993); Hines v. Secretary of Health & Human Servs., 940 F.2d 1518, 1523-24 (Fed.Cir.1991). In effect, this court performs the same task as the Court of Federal Claims and determines anew whether the special master’s findings were arbitrary or capricious.
The arbitrary and capricious standard of review is difficult for an appellant to satisfy with respect to any issue, but particularly with respect to an issue that turns on the weighing of evidence by the trier of fact. In general, reversible error is “extremely difficult to demonstrate” if the special master “has considered the relevant evidence of record, drawn plausible inferences and articulated a rational basis for the decision.” Hines, 940 F.2d at 1528. As this court stated in Munn, 970 F.2d at 871:
it is not ... the role of this court to reweigh the factual evidence, or to assess whether the special master correctly evaluated the evidence. And of course we do not examine the probative value of the evidence or the credibility of the witnesses. These are all matters within the purview of the fact finder.
The special master analyzed this case as follows: Based on an assessment of the evidence presented at the first and second hearings, the special master found that the bicycle-pedaling movements that Rachael displayed between her second and third DPT shots constituted the onset of a seizure disorder. The special master concluded, however, that the Lampes did not prove that the bicycle-pedaling movements were caused by Rachael’s second DPT vaccination. The special master further found that the seizure she suffered on January 26, 1976, was a continuation of her earlier seizure disorder; for that reason, the special master concluded, the third DPT vaccination cannot be viewed as the cause of her injury. Based on those findings, the special master concluded that the only remaining theory on which liability could be premised was that the third DPT shot aggravated a pre-existing seizure condition *1361that was not caused by the DPT vaccinations. After analyzing all the evidence pertaining to that question, the 'special master concluded that the Lampes had failed to prove causation on that theory as well, and he therefore denied compensation.
As is often true in Vaccine Act cases based on a theory of actual causation, the expert medical testimony was important in this case. Following his rejection of the Lampes’ Table-injury claim, the special master requested additional reports from the expert witnesses regarding the relationship between the bicycle-pedaling movements and the seizure disorder. In particular, he requested that the doctors address whether Rachael’s bicycle-pedaling movements constituted seizure activity, the significance of an affirmative answer to that question to the doctors’ opinion of causation, and any reasons supporting a conclusion that the prior seizure activity was not significant.
The Lampes’ experts reported and testified that in their view the January 26, 1976, event marked the onset of Rachael’s seizure activity, and that her seizure disorder and mental retardation were medically related to the DPT immunizations she received. They gave three reasons for concluding that the DPT vaccinations caused Rachael’s seizures: (1) other possible causes had been eliminated through comprehensive testing; (2) Rachael was a highly allergic child with a family history of allergies, and the pertussis component of the DPT vaccine that she was given had compounds in it that could cause a strong reaction in a highly allergic person; and (3) the January 26, 1976, seizure occurred within seven days of the third DPT shot.
The force of the testimony given by the Lampes’ experts was undermined in several ways. First, as the special master noted, the absence of alternative causes for a condition does not alone suffice to ascribe causation to the vaccine. See, e.g., Grant v. Secretary of Health & Human Servs., 956 F.2d 1144, 1149 (Fed.Cir.1992). Moreover, to support their increasing-allergic-reaction theory, both doctors relied heavily on the Lampes’ representation that the leg-jerking movements occurred shortly after each administration of the DPT vaccine, even though the special master specifically rejected that evidence after the first hearing. Finally, the doctors supported their theory of causation based on temporal association by citing a major epidemiological study, R. Alderslade et al., The National Childhood Encephalopathy Study, in Whooping Cough: Reports from the Committee on Safety of Medicines and the Joint Committee on Vaccination and Immunisation 79 (Department of Health & Social Security ed., 1981) [hereinafter NCES ]. The special master, however, found the NCES inapplicable to the facts of this case.
Dr. Snyder, on the other hand, specifically addressed the special master’s concerns regarding the bicycle-pedaling movements. He testified that the bicycle-pedaling movements that Rachael displayed between her second and third DPT shots constituted seizure activity and that the seizure she suffered on January 26, 1976, was a continuation of that prior seizure activity. His conclusions were based on the contemporaneous medical records of the neurologist who attended Rachael during her hospitalization following the January 26 seizure- — -records the special master had previously found to be credible. Dr. Snyder further testified that he did not regard it as likely that the January 26 seizure was caused by the DPT vaccination Rachael received seven days earlier because (1) she had experienced previous seizure activity, (2) she did not have the seizure for a significant period of time following the vaccination, and (3) she did not suffer another seizure for about a month. In addition, because Rachael did not suffer any decline in development until much later, Dr. Snyder concluded that the third DPT shot did not significantly aggravate Rachael’s condition.
Summarizing his assessment of the experts’ testimony on the issue of causation, *1362the special master credited Dr. Snyder’s evidence, but found the testimony given by Dr. Conkling and Dr. Lewis “unpersuasive.” Those findings, which are at the core of the special master’s decision in this case, are largely based on his assessments of the credibility of the witnesses and the relative persuasiveness of the competing medical theories of the case. As such, they are virtually unchallengeable on appeal. As we have previously stated,
Congress assigned to a group of specialists, the Special Masters within the Court of Federal Claims, the unenviable job of sorting through these painful cases and, based upon their accumulated expertise in the field, judging the merits of the individual claims. The statute makes clear that, on review, the Court of Federal Claims is not to second guess the Special Masters’ fact-intensive conclusions; the standard of review is uniquely deferential for what is essentially a judicial process. ... That level of deference is especially apt in a case in which the medical evidence of causation is in dispute.
Hodges, 9 F.3d at 961. Given the exacting “arbitrary and capricious” standard of review that applies to factual findings in Vaccine Act cases, we accept the special master’s weighing of the evidence in this case. We now turn to the Lampes’ contention that the special master made legal errors that materially affected his decision.
A
The Lampes first argue that the special master erred by limiting the grounds on which they were permitted to prove their entitlement to compensation. Because the special master found that the bicycle-pedaling movements commencing between the second and third DPT vaccinations marked the onset of Rachael’s condition, he ruled that the Lampes had two potential avenues of recovery: (1) to prove that the second DPT vaccination caused the bicycle-pedaling movements; or (2) to prove that the third DPT vaccination significantly aggravated her pre-existing seizure condition. That characterization of the avenues of proof was inconsistent, the Lampes argue, with the statutory provision that allows proof of an actual-causation case if the victim “sustained, or had significantly aggravated, any illness, disability, injury, or condition set forth in the Vaccine Injury Table the first symptom or manifestation of the onset or significant aggravation of which did not occur within the time period set forth in the Table but which was caused” by her DPT vaccinations. 42 U.S.C. § 3OOaa — 11(c)(1)(C)(ii)(II).
The special master’s characterization of the permissible avenues of proof open to the Lampes was based on his previous finding that “the bike pedaling movements Rachael experienced between her second and third vaccinations constituted the onset of a seizure disorder which included the January 26, 1976 seizure and subsequent seizures.” That finding is incompatible with the Lampes’ theory that the onset of Rachael’s seizure condition and associated mental retardation was caused by the third vaccination, since the seizure condition, in the special master’s view, predated the third vaccination. See Shalala v. Whitecotton, 514 U.S. 268, 274, 115 S.Ct. 1477, 131 L.Ed.2d 374 (1995) (holding that the onset of a condition cannot be established for a Table injury if “evidence of the injury appeared before the vaccination”).
In light of the special master’s finding that Rachael’s seizure condition was not caused by her first or second DPT vaccinations and had its onset before the third DPT vaccination, the only theory on which a finding of liability could be based was that the third vaccination aggravated a seizure condition that had a separate and unknown etiology. That is, because the special master permissibly found that Rachael’s seizure condition began before the January 26 incident and that the seizures she suffered before that incident were not caused by the vaccinations, it was proper for the special master to rule that in order to recover the Lampes would have to prove that the third vaccination resulted in a significant aggravation of her condition.
*1363The Lampes’ contention that the special master erred in limiting the grounds for recovery thus rests on a challenge to the special master’s factual finding that the onset of Rachael’s seizure condition occurred in the month or so prior to the January 26, 1976, seizure. The Lampes argue that the January 26 seizure marked the onset of Rachael’s seizure condition and that the special master erred in finding that it was a continuation of the prior seizure activity that Rachael had experienced during the previous month. It was contrary to the medical evidence, the Lampes argue, for the special master to conclude that the bicycle-pedaling movements during the month before the January 26 seizure constituted the onset of Rachael’s seizure condition. The Lampes’ own petition, however, stated that the bicycle-pedaling movements “became more noticeable and intense ... following the second and third DPT administrations and ultimately culminated in [the] major motor seizure” on January 26, and it cited the attending neurologist’s conclusion the seizure did not “represent a new acute situation.” Dr. Lewis similarly testified that “I think it’s a high probability that they [the bicycle-pedaling movements] were seizures,” further undermining the Lampes’ position on that issue.
In addition, Dr. Snyder’s testimony supports the special master’s finding on that point, and although the Lampes contend that the more compelling evidence is to the contrary, we do not sit to reweigh the evidence. Since the special master’s conclusion was based on evidence in the record that was not wholly implausible, we are compelled to uphold that finding as not being arbitrary or capricious. And once the special master’s underlying findings are accepted — that Rachael’s bicycle-pedaling movements were seizures and that those movements represented the onset of her seizure condition — there is no legal flaw in the special master’s refusal to treat the third vaccination as the triggering event for Rachael’s seizure condition.
The dissent contends that the special master committed legal error by requiring that Rachael’s injury be linked to one particular injection rather than focusing on the cumulative effect of the series of injections. That error, the dissent argues, prejudiced the Lampes because it prevented the special master from focusing on, and accepting, their theory that her seizure disorder was the result of a severe progressive allergic reaction to the entire series of DPT vaccinations.
The special master’s opinion makes clear that he understood the “progressive allergic reaction” theory articulated by the Lampes’ experts, which he set forth accurately and in detail; he simply rejected that theory for failure of proof. Significantly, the Lampes’ experts both based their opinions on reactions by Rachael that they believed occurred within a short time after each administration of the DPT vaccine. The special master’s rejection of the Lampes’ testimony that Rachael exhibited reactions to the first DPT vaccination and his conclusion that the Lampes failed to show that the bicycle-pedaling movements in December 1975 occurred within seven days of the second vaccination significantly undermined the Lampes’ theory of causation.
The Lampes’ experts also expressed the view that, consistent with their theory of causation, the bicycle-pedaling movements were a reaction to the vaccine and therefore represented a step in the course of reactions leading to the severe seizure that followed the third vaccination. But the special master, after considering all the evidence, concluded that the bicycle-pedaling movements were not shown to have been caused by her second DPT vaccination. Even more significantly, the special master found that the January 1976 seizure was a product of the seizure condition that had its onset with the December 1975 bicycle-pedaling movements. That finding is flatly inconsistent with the Lampes’ “progressive allergic reaction” theory, which sought to explain the seizures that Rachael suffered in December 1975 and January 1976 as progressive reactions to *1364the DPT vaccine that ultimately led to her seizure condition and mental retardation. It is those factual findings by the special master, not any error of law, that explains why the special master rejected the “progressive allergic reaction” theory of causation.
The dissent seeks to avoid the effect of those findings by arguing that the special master improperly focused on the effect of each vaccination rather than on the overall impact of the DPT vaccine on Rachael’s system. That, however, is a false dichotomy. If the vaccine caused Rachael’s injuries, it did so by causing her to have increasingly severe reactions to each vaccination, which in turn led to her development of a chronic seizure condition. Even under the dissent’s “overall impact” theory, at least one of the shots (the third shot in the dissent’s theory) has to have caused a seizure that led to Rachael’s injuries.
The Lampes’ experts traced her injuries to the seizures she suffered, beginning with the January 1976 seizure, which, they contend, was a reaction to the DPT vaccination seven days earlier. If that seizure was the product of an independently caused, pre-existing seizure condition, as the special master found, the vaccine cannot be said to have caused that reaction and Rachael’s seizure condition. Accordingly, even though the dissent characterizes the special master’s decision as the product of an error of law, the dissent’s quarrel is really with the special master’s evaluation of the evidence, which is a “matter! 1 within the purview of the fact finder.” Munn, 970 F.2d at 871.
B
The Lampes next argue that even if Rachael’s leg-jerking movements in or about December 1975 constituted seizures, it was improper for the special master to characterize her seizure condition at that time as a “pre-existing condition,” because the December 1975 seizures did not occur prior to the first of her DPT vaccinations. According to the Lampes, such a characterization is contrary to the Vaccine Act and our cases.
While it is true that the onset of the seizure condition identified by the special master occurred after Rachael’s first two DPT vaccinations, and thus cannot be regarded as a “pre-existing condition” with respect to those two vaccinations, the special master found that those vaccinations did not cause Rachael’s seizure condition. Although Dr. Lewis thought that the second shot caused the bicycle-pedaling movements, she conceded that she could not fix the time of those movements with respect to the second DPT vaccination. As she put it, “[M]y bet is it [the bicycle-pedaling movements] probably did occur within seven days [of the DPT vaccination], I don’t know, but it’s only a bet.” Dr. Conkling did not address the cause of the bicycle-pedaling movements other than in relation to his increasing-allergic-reaction theory. That theory, however, was based' — -like Dr. Lewis’s similar theory — in large part on the parents’ assertion of the temporal proximity of the bicycle-pedaling movements to the first and second vaccinations, evidence that the special master discredited after the first evidentiary hearing. The special master’s determination that the first and second vaccinations did not cause the bicycle-pedaling movements therefore was not arbitrary and capricious. As a result, the seizure condition was properly regarded as a pre-existing condition in relation to the third vaccination.
Contrary to the Lampes’ assertion, this court’s decision in Whitecotton v. Secretary of Health & Human Services, 81 F.3d 1099 (Fed.Cir.1996), supports the special master’s conclusion. That case concerned Maggie WTiitecotton’s microcephalic condition prior to her third DPT administration. Microcephaly, or small head size, is sometimes linked to encephalopathy, a Table injury. The special master concluded that Maggie was “ ‘at least borderline microce-phalic at birth and ... clearly microce-phalic by the time she received her third DPT shot.’ ” Id. at 1103 (omission in original). The special master also found that Maggie’s microcephaly had indeed caused brain damage before the third vaccination. *1365The court accepted the special master’s findings and agreed that Maggie’s brain damage, which was not shown to be caused by the first two DPT vaccinations, had to be treated as a pre-existing condition with respect to the third vaccination. The court therefore agreed with the special master that the third vaccination did not mark the onset of her disabilities. Significantly, in reaching that conclusion, the court did not determine or even consider whether Maggie’s actual impairment predated her first DPT vaccination.
A parallel analysis applies here. The statutory basis for recovery by presumption of causation under the Vaccine Injury Table essentially mirrors the statutory basis for recovery by proof of actual causation. See 42 U.S.C. § 300aa-ll(c)(l)(C)(i), ll(c)(l)(C)(ii)(II). The special master found that the third DPT vaccination cannot be regarded as the cause of Rachael’s seizure condition, because at the time she received that vaccination she was already suffering from a seizure condition that was not caused by the two prior DPT vaccinations. Accordingly, the only remaining question for the special master was whether the third vaccination significantly aggravated that pre-existing seizure condition. For that reason, the special master did not commit legal error by characterizing the December 1975 seizures as manifestations of a pre-existing seizure condition, even though the December seizures did not occur prior to the first administration of the DPT vaccine.
C
The Lampes next argue that the' special master failed to attach sufficient weight to the NCES, a study of the neurological effects of DPT vaccine conducted during the late 1970s. The Lampes contend that the NCES strongly supports a finding that Rachael’s vaccinations caused her seizure condition and ultimately her mental retardation. The special master, however, adverted to the NCES but found that its conclusions were not especially instructive as applied to this case.
An epidemiological study may be probative medical evidence relevant to a causation determination. See, e.g., Grant, 956 F.2d at 1149. In order for the study to be instructive, however, its conclusions must fit the facts of the case under consideration. See id. (upholding the special master’s rejection of epidemiological studies of the general DPT vaccine, when a different vaccine was administered to the injured child). In this case, the experts did not agree on whether the conclusions of the NCES apply to children with Rachael’s history and symptoms.
For children suffering from a convulsion, the NCES included those cases in which the convulsion lasted more than half an hour. Rachael’s January 26, 1976, seizure was described by one of her doctors as lasting 20 to 25 minutes, but by another as lasting five minutes “followed by a 30 minute period of limpness and unresponsiveness.” Drs. Lewis and Conkling conceded that Rachael’s case might not have been within the scope of the NCES, but nevertheless relied on the conclusions of the NCES to support causation. Dr. Snyder, in contrast, stated flatly that Rachael’s case would not have been included in the NCES and that the study did not suggest that Rachael’s condition was caused by the DPT vaccinations.
Even if Rachael’s January 26, 1976, seizure had qualified her case for inclusion in the NCES, the study does not provide strong evidence of causation on the facts of this case. For a child suffering from repeated convulsions, the NCES researchers first determined whether the convulsions were part of a single pathological process. If so, the date of onset of illness was deemed for purposes of the study to be the date of the first convulsion. If no single pathological process could be identified, the date of onset of illness was deemed for purposes of the study to be the date the child was admitted to a hospital for a major convulsion, and the previous convulsions were considered to be part of the *1366child’s pre-vaccination condition. NCES, supra, at 147.
If Rachael’s bicycle-pedaling movements were, as the special master found, part of a single pathological process that includes the January 26, 1976, seizure, then the NCES would regard the date of onset as the date the bicycle-pedaling movements began. With that onset date indeterminate in Rachael’s case, the NCES cannot support causation because its conclusions are based on strict temporal proximity between the vaccination and the onset of symptoms. On the other hand, if the bicycle-pedaling movements are not related to the January 26 seizure, then the date of onset would fall within the NCES’s outermost temporal limits, but her case would then fall into the category of cases in which the child was previously neurologically abnormal. NCES, supra, at 103, 108. For such children, the NCES was unable “to separate those within this group who suffered further neurological damage during their illness from those who did not.” Id. at 108. Consequently, Rachael’s history does not fit the NCES paradigm closely enough for that study to shed light on the issue of causation in her case.
In any event, a special master’s task is to make a factual determination of causation based on the evidence in a particular case. A study of many individual cases may be useful evidence as to causation, but it does not compel the finder of fact to find causation in a particular case. See, e.g., Hodges, 9 F.3d at 961 n. 4 (noting that the special master’s judgment that the petitioner’s evidence fell short of “proving the case by the standard the law requires is not the same as refusing to consider it”). Accordingly, there is nothing in the special master’s treatment of the NCES that would justify reversal of his decision in this case.
D
The Lampes argue that, even if the special master was correct to conclude that Rachael had a pre-existing seizure condition at the time of her third DPT vaccination, the special master erred in his analysis of whether the third vaccination significantly aggravated that pre-ex-isting seizure condition. According to the Lampes, the special master focused only on selected positive aspects of Rachael’s history following the third vaccination and thus failed to compare her pre-vaccination condition with her current condition of severe retardation and residual seizure disorder.
We do not believe the special master disregarded the difference between Rachael’s pre-vaccination condition and her present condition. Rather, the special master focused on the absence of any immediate deterioration in Rachael’s condition following the third vaccination in an effort to determine whether that vaccination should be regarded as the cause of the deterioration that ultimately occurred.
The experts clearly disagreed about the impact of the third vaccination. Drs. Lewis and Conkling thought that the DPT vaccine caused Rachael’s seizure disorder, which itself ultimately led to her developmental delays and mental retardation. Dr. Snyder, on the other hand, thought that the January 26, 1976, seizure was a natural progression from Rachael’s bicycle-pedaling movements and that it did not represent “exacerbation of a condition.” Moreover, Dr. Snyder found the timing of Rachael’s major developmental delays' — • between ages three and six — difficult to reconcile with an injury traceable to the January 19,1976, vaccination.
In resolving the issue of causation over which the experts disagreed, the special master was justified in taking into account the period of time that elapsed between the third vaccination and the onset of Rachael’s developmental delays. The passage of time between an event and the consequences that are alleged to flow from it is often significant, and in light of Dr. Snyder’s testimony, it was not improper for the special master to attach some significance to the lengthy period of delay between the vaccination and the deterioration in Rachael’s condition.
*1367The special master of course recognized that Rachael’s current condition is far worse than her pre-vaccination condition. The problem is that the Lampes failed to prove to the special master’s satisfaction that it was the vaccine that was responsible for the deterioration in her condition. Nowhere in his opinion did the special master conclude that the third DPT vaccination could not aggravate a pre-existing seizure condition such as Rachael’s. Rather, the special master concluded that the Lampes did not establish that the third DPT vaccination had that effect.
The dissent argues that in analyzing the presentations of the experts, the special master erred (1) by discounting the testimony of the Lampes’ experts because they were unwilling to testify unequivocally that the vaccine caused Rachael’s injuries, and (2) by crediting Dr. Snyder’s testimony even though Dr. Snyder was not able to provide an alternative theory of causation for Rachael’s seizure condition. That, however, is not an accurate characterization of what the special master did.
First, rather than imposing some Procrustean requirement that the experts state their opinions in unequivocal terms, the special master simply noted the points at which the various experts qualified then-answers to questions about the case; it was certainly legitimate, for example, for the special master to point out that Dr. Lewis testified that the bicycle-pedaling movements “could have been” indicative of an allergic reaction to the DPT vaccine, and there was certainly nothing wrong with the special master’s reference to Dr. Conkling’s concession, with respect to the “allergic reaction” theory, that “it is difficult to assess allergic reactions in infants,” and that “it is difficult to pinpoint the vaccination as the causal agent” for Rachael’s injuries.
Second, the special master was not obliged to discredit Dr. Snyder’s testimony simply because Dr. Snyder was unable to identify the agent that caused Rachael’s seizure condition. Dr. Snyder gave detailed testimony about why he did not regard the DPT vaccinations as causally linked to Rachael’s condition, and he did so, as the special master found, “to a reasonable degree of medical certainty.” Dr. Snyder also explained that he would have expected Rachael to react differently if the DPT vaccination had been causally linked to her seizures. In an actual-causation case, as we have noted, evidence that no other cause has been identified is not sufficient to compel a finding that the vaccine caused the injuries in question. Dr. Snyder’s inability to identify an alternative possible cause of Rachael’s condition thus does not undermine the force of his testimony or render it improper for the special master to credit and rely on that testimony
E
Finally, citing this court’s decision in Shyface v. Secretary of Health & Human Services, 165 F.3d 1344, 1351-53 (Fed.Cir.1999), the Lampes argue that the special master did not apply the traditional principles of tort causation to the issue of actual causation in this case and that he therefore reached the wrong result in finding an absence of causation. We see no error in the special master’s decision in that regard. He found that the first DPT vaccination had no recognized effect on Rachael, that the second DPT vaccination did not cause her seizure condition, and that the third DPT vaccination did not aggravate that condition. Nothing in the special master’s opinion suggests that he applied an improper standard of causation.
The Lampes contend that the extensive testing done on Rachael, which revealed no alternative cause for her condition, raises a strong inference that the DPT vaccinations caused her condition and that it was error for the special master to conclude otherwise. In addition, they argue that then-experts’ theory of a progressive allergic reaction finds compelling support from the fact that Rachael suffered a grand mal seizure one day after receiving a DPT booster shot in January 1977. Dr..Snyder, *1368however, testified that even in light of those circumstances, it was not possible to conclude that the DPT vaccinations caused or significantly aggravated Rachael’s seizure condition and her subsequent mental retardation. His conclusion finds support in Rachael’s medical records, which suggest that Rachael’s recurrent ear infections and tonsillitis may have made her seizure condition considerably worse. His conclusion also finds support in the records of the pediatrician who attended Rachael at the time of the booster shot, which indicate that she had three seizures within that general period and that her medication for seizure control had recently been increased. Those records also do not indicate the severity or duration of the January 1977 seizure or whether Rachael was hospitalized as a result, but they do indicate that a seizure-related hospitalization occurred in October 1976, and that Rachael’s allergies may have been causing problems even before the booster shot was administered.
At bottom, the Lampes’ argument is that because Rachael was healthy before the DPT series commenced, because no other cause has been identified after extensive testing, and because Rachael experienced a seizure within seven days of her third DPT vaccination, the special master was required as a matter of law to find causation. Acceptance of that argument would in effect create a new Table injury based on onset or significant aggravation within seven days, instead of the statutory three days. We decline to adopt such a rigid standard for determining actual causation, particularly in light of the 1994 decision of the National Vaccine Advisory Committee that the available scientific evidence did not support making a similar change in the Vaccine Injury Table. See National Vaccine Advisory Comm., Report of the Ad Hoc Subcommittee on Childhood Vaccines 8 (1994).
Ill
In the end, this case, like so many Vaccine Act cases, turns on its facts. The factual presentation made by the Lampes was a strong one in many respects — the Lampes’ experts presented plausible theories of causation, and the evidence showed that medical examinations of Rachael eliminated many of the possible non-vaccine-related causes of her seizure condition. Based on the evidence the Lampes’ presented, a finder of fact might well have been persuaded that the Lampes had shown that the third DPT vaccination — or the entire series of DPT vaccinations— caused Rachael’s seizure condition, which in turn ultimately led to her mental retardation. But the special master, after carefully weighing the evidence in favor of, and contrary to, that hypothesis, reached the contrary conclusion. We cannot say that the special master’s conclusion was so clearly wrong as to be arbitrary or capricious; nor do we find that the special master committed any legal errors in reaching his conclusion. We therefore must uphold the special master’s decision denying compensation.
Each party will bear its own costs for this appeal.
AFFIRMED.