dissenting.
The United States Court of Appeals for the Armed Forces held that the President violated the Constitution in June 1991, when he promulgated Rule 707 of the Military Rules of Evidence. Had I been a member of that court, I would not have decided that question without first requiring the parties to brief and argue the antecedent question whether Rule 707 violates Article 86(a) of the Uniform Code of Military Justice, 10 U. S. C. § 836(a). As presently advised, I am persuaded that the Rule does violate the statute and should be held invalid for that reason. I also agree with the Court of Appeals that the Rule is unconstitutional. This Court’s contrary holding rests on a serious undervaluation of the importance of the citizen’s constitutional right to present a de*321fense to a criminal charge and an unrealistic appraisal of the importance of the governmental interests that undergird the Rule. Before discussing the constitutional issue, I shall comment briefly on the statutory question.
I
Rule 707 is a blanket rule of exclusion.1 No matter how reliable and how probative the results of a polygraph test may be, Rule 707 categorically denies the defendant any opportunity to persuade the court that the evidence should be received for any purpose. Indeed, even if the parties stipulate in advance that the results of a lie detector test may be admitted, the Rule requires exclusion.
The principal charge against the respondent in this case was that he had knowingly used methamphetamine. His principal defense was “innocent ingestion”; even if the urinalysis test conducted on April 7, 1992, correctly indicated that he did ingest the substance, he claims to have been unaware of that fact. The results of the lie detector test conducted three days later, if accurate, constitute factual evidence that his physical condition at that time was consistent with the theory of his defense and inconsistent with the theory of the prosecution. The results were also relevant because they tended to confirm the credibility of his testimony. Under Rule 707, even if the results of the polygraph test were more reliable than the results of the urinalysis, the weaker evidence is admissible and the stronger evidence is inadmissible.
Under the now discredited reasoning in a case decided 75 years ago, Frye v. United States, 54 App. D. C. 46, 293 *322F. 1013 (1923), that anomalous result would also have been reached in nonmilitary cases tried in the federal courts. In recent years, however, we have not only repudiated Frye’s general approach to scientific evidence, but the federal courts have also been engaged in the process of rejecting the once-popular view that all lie detector evidence should be categorically inadmissible.2 Well reasoned opinions are concluding, consistently with this Court’s decisions in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U. S. 579 (1993), and General Electric Co. v. Joiner, 522 U. S. 136 (1997), that the federal rules wisely allow district judges to exercise broad discretion when evaluating the admissibility of scientific evidence.3 Those opinions correctly observe that the rules of evidence generally recognized in the trial of civil and criminal cases in the federal courts do not contain any blanket prohibition against the admissibility of polygraph evidence.
*323In accord with the modern trend of decisions on this admissibility issue, in 1987 the Court of Military Appeals held that an accused was “entitled to attempt to lay” the foundation for admission of favorable polygraph evidence. United States v. Gipson, 24 M. J. 246, 253 (1987). The President responded to Gipson by adopting Rule 707. The governing statute authorized him to promulgate evidentiary rules “which shall, so far as he considers practicable, apply the principles of law and the rules of evidence generally recognized in the trial of criminal eases in the United States district courts.” 10 U. S. C. § 886(a).4 Thus, if there are military concerns that warrant a special rule for military tribunals, the statute gives him ample authority to promulgate special rules that take such concerns into account.
Rule 707 has no counterpart in either the Federal Rules of Evidence or the Federal Rules of Criminal Procedure. Moreover, to the extent that the use of the lie detector plays a special role in the military establishment, military practices are more favorable to a rule of admissibility than is the less structured use of lie detectors in the civilian sector of our society. That is so because the military carefully regulates the administration of polygraph tests to ensure reliable results. The military maintains “very stringent standards for polygraph examiners”5 and has established its own Poly*324graph Institute, which is “generally considered to be the best training facility for polygraph examiners in the United States.”6 The military has administered hundreds of thousands of such tests and routinely uses their results for a wide variety of official decisions.7
*325The stated reasons for the adoption of Rule 707 do not rely on any special military concern. They merely invoke three interests: (1) the interest in excluding unreliable evidence; (2) the interest in protecting the trier of fact from being misled by an unwarranted assumption that the polygraph evidence has “an aura of near infallibility”; and (3) the interest in avoiding collateral debates about the admissibility of particular test results.
It seems clear that those interests pose less serious concerns in the military than in the civilian context. Disputes about the qualifications of the examiners, the equipment, and the testing procedures should seldom arise with respect to the tests conducted by the military. Moreover, there surely is no reason to assume that military personnel who perform the factfinding function are less competent than ordinary jurors to assess the reliability of particular results, or their relevance to the issues.8 Thus, there is no identifiable military concern that justifies the President’s promulgation of a special military rule that is more burdensome to the accused in military trials than the evidentiary rules applicable to the trial of civilians.
It, therefore, seems fairly clear that Rule 707 does not comply with the statute. I do not rest on this ground, however, because briefing might persuade me to change my views, and because the Court has decided only the constitutional question.
II
The Court’s opinion barely acknowledges that a person accused of a crime has a constitutional right to present a *326defense. It is not necessary to point to “any particular language in the Sixth Amendment,” ante, at 307, to support the conclusion that the right is firmly established. It is, however, appropriate to comment on the importance of that right before discussing the three interests that the Government relies upon to justify Rule 707.
The Sixth Amendment provides that “the accused shall enjoy the right... to have compulsory process for obtaining witnesses in his favor.” Because this right “is an essential attribute of the adversary system itself,” we have repeatedly stated that few rights “are more fundamental than that of an accused to present witnesses in his own defense.”9 According to Joseph Story, that provision was included in the Bill of Rights in reaction to a notorious common-law rule categorically excluding defense evidence in treason and felony cases.10 Our holding in Washington v. Texas, 388 U. S. 14 (1967), that this right is applicable to the States, rested on the premises that it “is in plain terms the right to present a defense” and that it “is a fundamental element of due proc*327ess of law.”11 Consistent with the history of the provisión, the Court in that ease held that a state rule of evidence that excluded “whole categories” of testimony on the basis of a presumption of unreliability was unconstitutional.12
The blanket rule of inadmissibility held invalid in Washington v. Texas covered the testimony of alleged accomplices. Both before and after that decision, the Court has recognized the potential injustice produced by rules that exclude entire categories of relevant evidence that is potentially unreliable. At common law interested parties such as defendants,13 their spouses,14 and their eo-eonspirators15 were not competent *328witnesses. “Nor were those named the only grounds of exclusion from the witness stand; conviction of crime, want of religious belief, and other matters were held sufficient. Indeed, the theory of the common law was to admit to the witness stand only those presumably honest, appreciating the sanctity of an oath, unaffected as a party by the result, and free from any of the temptations of interest. The courts were afraid to trust the intelligence of jurors.” Benson v. United States, 146 U. S. 325, 336 (1892). And, of course, under the regime established by Frye v. United States, scientific evidence was inadmissible unless it met a stringent “general acceptance” test. Over the years, with respect to category after category, strict rules of exclusion have been replaced by rules that broaden the discretion of trial judges to admit potentially unreliable evidence and to allow properly instructed juries to evaluate its weight. While that trend has included both rulemaking and noneonstitutional judicial decisions, the direction of the trend has been consistent and it has been manifested in constitutional holdings as well.
Commenting on the trend that had followed the decision in Benson, the Court in 1918 observed that in the
“years which have elapsed since the decision of the Benson Case, the disposition of courts and of legislative bodies to remove disabilities from witnesses has continued, as that decision shows it had been going forward before, under dominance of the conviction of our time that the *329truth is more likely to be arrived at by hearing the testimony of all persons of competent understanding who may seem to have knowledge of the facts involved in a ease, leaving the credit and weight of such testimony to be determined by the jury or by the court, rather than by rejecting witnesses as incompetent, with the result that this principle has come to be widely, almost universally, accepted in this country and in Great Britain.” Rosen v. United States, 245 U. S. 467, 471.
See also Funk v. United States, 290 U. S. 371, 377-378 (1933). It was in a case involving the disqualification of spousal testimony that Justice Stewart stated: “Any rule that impedes the discovery of truth in a court of law impedes as well the doing of justice.” Hawkins v. United States, 358 U. S. 74, 81 (1958) (concurring opinion).
State evidentiary rules may so seriously impede the discovery of truth, “as well as the doing of justice,” that they preclude the “meaningful opportunity to present a complete defense” that is guaranteed by the Constitution, Crane v. Kentucky, 476 U. S. 683, 690 (1986) (internal quotation marks omitted).16 In Chambers v. Mississippi, 410 U. S. 284, 302 *330(1973), we concluded that “where constitutional rights directly affecting the ascertainment of guilt are implicated, the hearsay rule may not be applied mechanistically to defeat the ends of justice.”17 As the Court notes today, restrictions on the “defendant’s right to present relevant evidence,” ante, at 308, must comply with the admonition in Rock v. Arkansas, 483 U. S. 44, 56 (1987), that they “may not be arbitrary or disproportionate to the purposes they are designed to serve.” Applying that admonition to Arkansas’ blanket rule prohibiting the admission of hypnotically refreshed testimony, we concluded that a “State’s legitimate interest in barring unreliable evidence does not extend to per se exclusions that may be reliable in an individual case.” Id., at 61. That statement of constitutional law is directly relevant to this case.
*331rH KH f-H
The constitutional requirement that a blanket exclusion of potentially unreliable evidence must be proportionate to the purposes served by the rule obviously makes it necessary to evaluate the interests on both sides of the balance. Today the Court all but ignores the strength of the defendant’s interest in having polygraph evidence admitted in certain cases. As the facts of this case illustrate, the Court is quite •wrong in assuming that the impact of Rule 707 on respondent’s defense was not significant because it did not preclude the introduction of any “factual evidence” or prevent him from conveying “his version of the facts to the court-martial members.” Ante, at 317. Under such reasoning, a rule that excluded the testimony of alibi witnesses would not be significant as long as the defendant is free to testify himself. But given the defendant’s strong interest in the outcome— an interest that was sufficient to make his testimony presumptively untrustworthy and therefore inadmissible at common law — his uncorroborated testimony is certain to be less persuasive than that of a third-party witness. A rule that bars him “from introducing expert opinion testimony to bolster his own credibility,” ibid., unquestionably impairs any “meaningful opportunity to present a complete defense”; indeed, it is sure to be outcome determinative in many cases.
Moreover, in this case the results of the polygraph test, taken just three days after the urinalysis, constitute independent factual evidence that is not otherwise available and that strongly supports his defense of “innocent ingestion.” Just as flight or other evidence of “consciousness of guilt” may sometimes be relevant, on some occasions evidence of “consciousness of innocence” may also be relevant to the central issue at trial. Both the answers to the questions propounded by the examiner, and the physical manifestations produced by those utterances, were probative of an innocent state of mind shortly after he ingested the drugs. In Dean Wigmore’s view, both “conduct” and “utterances” may con*332stitute factual evidence of a “consciousness of innocence.”18 As the Second Circuit has held, when there is a serious factual dispute over the “basic defense [that defendant] was unaware of any criminal wrongdoing,” evidence of his innocent state of mind is “critical to a fair adjudication of criminal charges.”19 The exclusion of the test results in this ease-cannot be fairly equated with a ruling that merely prevented the defendant from encumbering the record with cumulative evidence. Because the Rule may well have affected the outcome of the trial, it unquestionably “infringed upon a weighty interest of the accused.” Ante, at 308.
The question, then, is whether the three interests on which the Government relies are powerful enough to support a categorical rule excluding the results of all polygraph tests no matter how unfair such a rule may be in particular eases.
*333 Reliability
There are a host of studies that place the reliability of polygraph tests at 85% to 90%.20 While critics of the polygraph argue that accuracy is much lower, even the studies cited by the critics place polygraph accuracy at 70%.21 Moreover, to the extent that the polygraph errs, studies have repeatedly shown that the polygraph is more likely to find innocent people guilty than vice versa.22 Thus, exculpatory polygraphs — like the one in this case — are likely to be more reliable than inculpatory ones.
Of course, within the broad category of lie detector evidence, there may be a wide variation in both the validity and the relevance23 of particular test results. Questions about the examiner’s integrity, independence, choice of questions, or training in the detection of deliberate attempts to provoke misleading physiological responses may justify exclusion of *334specific evidence. But such questions are properly addressed in adversary proceedings; they fall far short of justifying a blanket exclusion of this type of expert testimony.
There is no legal requirement that expert testimony must satisfy a particular degree of reliability to be admissible. Expert testimony about a defendant’s “future dangerousness” to determine his eligibility for the death penalty, even if wrong “most of the time,” is routinely admitted. Barefoot v. Estelle, 468 U. S. 880, 898-901 (1983). Studies indicate that handwriting analysis, and even fingerprint identifications, may be less trustworthy than polygraph evidence in certain eases.24 And, of course, even highly dubious eyewit*335ness testimony is, and should be, admitted and tested in the crucible of cross-examination. The Court’s reliance on potential unreliability as a justification for a categorical rule of inadmissibility reveals that it is “overly pessimistic about the capabilities of the jury and of the adversary system generally. "Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.” Daubert, 509 U. S., at 596.25
*336 The Role of the Jury
It is the function of the jury to make credibility determinations. In my judgment evidence that tends to establish either a consciousness of guilt or a consciousness of innocence may be of assistance to the jury in making such determinations. That also was the opinion of Dean Wigmore:
“Let the accused’s whole conduct come in; and whether it tells for consciousness of guilt or for consciousness of innocence, let us take it for what it is worth, remembering that in either case it is open to varying explanations and is not to be emphasized. Let us not deprive an innocent person, falsely accused, of the inference which common sense draws from a consciousness of innocence and its natural manifestations.” 2 J. Wigmore, Evidence § 293, p. 232 (J. Chadbourn rev. ed. 1979).
There is, of course, some risk that some “juries will give excessive weight to the opinions of a polygrapher, clothed as they are in scientific expertise,” ante, at 313-314. In my judgment, however, it is much more likely that juries will be guided by the instructions of the trial judge concerning the credibility of expert as well as lay witnesses. The strong presumption that juries will follow the court’s instructions, see, e. g., Richardson v. Marsh, 481 U. S. 200, 211 (1987), applies to exculpatory as well as inculpatory evidence. Com*337mon sense suggests that the testimony of disinterested third parties that is relevant to the jury’s credibility determination will assist rather than impair the jury’s deliberations. As with the reliance on the potential unreliability of this type of evidence, the reliance on a fear that the average jury is not able to assess the weight of this testimony reflects a distressing lack of confidence in the intelligence of the average American.26
Collateral Litigation
The potential burden of collateral proceedings to determine the examiner’s qualifications is a manifestly insufficient justification for a categorical exclusion of expert testimony. Such proceedings are a routine predicate for the admission of any expert testimony, and may always give rise to searching cross-examination. If testimony that is critical to a fair determination of gufit or innocence could be excluded for that reason, the right to a meaningful opportunity to present a defense would be an illusion.
It is incongruous for the party that selected the examiner, the equipment, the testing procedures, and the questions asked of the defendant to complain about the examinee’s burden of proving that the test was properly conducted. While there may well be a need for substantial collateral proceedings when the party objecting to admissibility has a basis for questioning some aspect of the examination, it seems quite obvious that the Government is in no position to challenge *338the competence of the procedures that it has developed and relied upon in hundreds of thousands of cases.
In all events the concern about the burden of collateral debates about the integrity of a particular examination, or the competence of a particular examiner, provides no support for a categorical rule that requires exclusion even when the test is taken pursuant to a stipulation and even when there has been a stipulation resolving all potential collateral issues. Indeed, in this very case there would have been no need for any collateral proceedings because respondent did not question the qualifications of the expert who examined him, and surely the Government is in no position to argue that one who has successfully completed its carefully developed training program27 is unqualified. The interest in avoiding burdensome collateral proceedings might support a rule prescribing minimum standards that must be met before any test is admissible,28 but it surely does not support the blunderbuss at issue.29
IV
The Government’s concerns would unquestionably support the exclusion of polygraph evidence in particular eases, and may well be sufficient to support a narrower rule designed to respond to specific concerns. In my judgment, however, *339those concerns are plainly insufficient to support a categorical rule that prohibits the admission of polygraph evidence in all cases, no matter how reliable or probative the evidence may be. Accordingly, I respectfully dissent.
Rule 707 states, in relevant part:
“Notwithstanding any other provision of law, the results of a polygraph examination, the opinion of a polygraph examiner, or any reference to an offer to take, failure to take, or taking of a polygraph examination, shall not be admitted into evidence.” Mil. Rule Evid. 707(a).
“There is no question that in recent years polygraph testing has gained increasingly widespread acceptance as a useful and reliable scientific tool. Because of the advances that have been achieved in the field which have led to the greater use of polygraph examination, coupled with a lack of evidence that juries are unduly swayed by polygraph evidence, we agree with those courts which have found that a per se rule disallowing polygraph evidence is no longer warranted. . . . Thus, we believe the best approach in this area is one which balances the need to admit all relevant and reliable evidence against the danger that the admission of the evidence for a given purpose will be unfairly prejudicial.” United States v. Piccinonna, 885 F. 2d 1529, 1535 (CA11 1989). “[W]e do not now hold that polygraph examinations are scientifically valid or that they will always assist the trier of fact, in this or any other individual case. We merely remove the obstacle of the per se rule against admissibility, which was based on antiquated concepts about the technical ability of the polygraph and legal precepts that have been expressly overruled by the Supreme Court.” United States v. Posado, 57 F. 3d 428, 434 (CA5 1995).
“The per se . .. rule excluding unstipulated polygraph evidence is inconsistent with the ^flexible inquiry3 assigned to the trial judge by Dau-bert. This is particularly evident because Frye, which was overruled by Daubert, involved the admissibility of polygraph evidence.” United States v. Cordoba, 104 F. 3d 225, 227 (CA9 1997).
“Pretrial, trial, and post-trial procedures, including modes of proof, for cases arising under this chapter triable in courts-martial, military commissions and other military tribunals, and procedures for courts of inquiry, may be prescribed by the President by regulations which shall, so far as he considers practicable, apply the principles of law and the rules of evidence generally recognized in the trial of criminal eases in the United States district courts, but which may not be contrary to or inconsistent with this chapter.” 10 U. S. C. § 836(a).
According to the Department of Defense’s 1996 Report to Congress: “The Department of Defense maintains very stringent standards for polygraph examiners. The Department of Defense Polygraph Institute’s basic polygraph program is the only program known to base its curriculum on forensic psychophysiology, and conceptual, abstract, and applied knowl*324edge that meet the requirements of a master’s degree-level of study. Candidates selected for the Department of Defense polygraph positions must meet the following minimum requirements:
“1. Be a United States citizen.
“2. Be at least 25 years of age.
“3. Be a graduate of an accredited four-year college or have equivalent experience that demonstrates the ability to master graduate-level academic courses.
"4. Have two years of experience as an investigator with a Federal or other law enforcement agency....
“5. Be of high moral character and sound emotional temperament, as confirmed by a background investigation.
“6. Complete a Department of Defense-approved course of polygraph instruction.
“7. Be adjudged suitable for the position after being administered a polygraph examination designed to ensure that the candidate realizes, and. is sensitive to, the personal impact of such examinations.
“All federal polygraph examiners receive their basic polygraph training at the Department of Defense Polygraph Institute. After completing the basic polygraph training, DoD personnel must serve an internship consisting of a minimum of six months on-the-job-training and conduct at least 25 polygraph examinations under the supervision of a certified polygraph examiner before being certified as a Department of Defense polygraph examiner. In addition, DoD polygraph examiners are required to complete 80 hours of continuing education every two years.” Department of Defense Polygraph Program, Annual Polygraph Report to Congress, Fiscal Year 1996, pp. 14-15; see also Yankee, The Current Status of Research in Forensic Psychophysiology and Its Application in the Psychophysiological Detection of Deception, 40 J. Forensic Sciences 63 (1995).
Honts & Perry, Polygraph Admissibility: Changes and Challenges, 16 Law and Human Behavior 357, 359, n. 1 (1992) (hereinafter Honts & Perry).
Between 1981 and 1997, the Department of Defense conducted over 400,000 polygraph examinations to resolve issues arising in counterintelligence, security, and criminal investigations. Department of Defense Polygraph Program, Annual Polygraph Report to Congress, Fiscal Year *3251997, p. 1; id., Fiscal Year 1996, p. 1; id., Fiscal Year 1995, p. 1; id., Fiscal Year 1994, p. 1; id., Fiscal Year 1993, App. A; id., Fiscal Year 1992, App. A; id., Fiscal Year 1991, App. A-1 (reporting information for 1981-1991).
When the members of the court-martial are officers, as was true in this case, they typically have at least a college degree as well as significant military service. See 10 U. S. C. § 825(d)(2); see also, e. g., United States v. Carter, 22 M. J. 771, 776 (A. C. M. R. 1986).
“Few rights are more fundamental than that of an accused to present witnesses in his own defense, see, e. g., Chambers v. Mississippi, 410 U. S. 284, 302 (1973). Indeed, this right is an essential attribute of the adversary system itself. . . . The right to compel a witness’ presence in the courtroom could not protect the integrity of the adversary process if it did not embrace the right to have the witness’ testimony heard by the trier of fact. The right to offer testimony is thus grounded in the Sixth Amendment....” Taylor v. Illinois, 484 U. S. 400, 408-409 (1988).
“Joseph Story, in his famous Commentaries on the Constitution of the United States, observed that the right to compulsory process was included in the Bill of Rights in reaction to the notorious common-law rule that in cases of treason or felony the accused was not allowed to introduce witnesses in his defense at all. Although the absolute prohibition of witnesses for the defense had been abolished in England by statute before 1787, the Framers of the Constitution felt it necessary specifically to provide that defendants in criminal cases should be provided the means of obtaining witnesses so that their own evidence, as well as the prosecution’s, might be evaluated by the jury.” Washington v. Texas, 388 U. S. 14, 19-20 (1967) (footnotes omitted).
“The right to offer the testimony of witnesses, and to compel their attendance, if necessary, is in plain terms the right to present a defense, the right to present the defendant’s version of the facts as well as the prosecution’s to the jury so it may decide where the truth lies. Just as an accused has the right to confront the prosecution’s witnesses for the purpose of challenging their testimony, he has the right to present his own witnesses to establish a defense. This right is a fundamental element of due process of law.” Id., at 19.
“It is difficult to see how the Constitution is any less violated by arbitrary rules that prevent whole categories of defense witnesses from testifying on the basis of a priori categories that presume them unworthy of belief
“The rule disqualifying an alleged accomplice from testifying on behalf of the defendant cannot even be defended on the ground that it rationally sets apart a group of persons who are particularly likely to commit perjury.” Id., at 22.
“It is familiar knowledge that the old common law carefully excluded from the witness stand parties to the record, and those who were interested in the result; and this rule extended to both civil and criminal cases. Fear of perjury was the reason for the rule.” Benson v. United States, 146 U. S. 325, 335 (1892).
“The common-law rule, accepted at an early date as controlling in this country, was that husband and wife were incompetent as witnesses for or against each other....
“The Court recognized that the basic reason underlying th[e] exclusion [of one spouse’s testimony on behalf of the other] had been the practice of disqualifying witnesses with a personal interest in the outcome of a ease. Widespread disqualifications because of interest, however, had long since *328been abolished both in this country and in England in accordance with the modern trend which permitted interested witnesses to testify and left it for the jury to assess their credibility. Certainly, since defendants were uniformly allowed to testify in their own behalf, there was no longer a good reason to prevent them from using their spouses as witnesses. With the original reason for barring favorable testimony of spouses gone the Court concluded that this aspect of the old rule should go too.” Hawkins v. United States, 358 U. S. 74, 75-76 (1958).
See Washington v. Texas, 388 U. S., at 20-21.
"whether rooted directly in the Due Process Clause of the Fourteenth Amendment, Chambers v. Mississippi, [410 U. S. 284 (1973)], or in the Compulsory Process or Confrontation clauses of the Sixth Amendment, Washington v. Texas, 388 U. S. 14, 23 (1967); Davis v. Alaska, 415 U. S. 308 (1974), the Constitution guarantees criminal defendants ‘a meaningful opportunity to present a complete defense.’ California v. Trombetta, 467 U. S. [479, 485 (1984)]; cf. Strickland v. Washington, 466 U. S. 668, 684-685 (1984) (‘The Constitution guarantees a fair trial through the Due Process Clauses, but it defines the basic elements of a fair trial largely through the several provisions of the Sixth Amendment’). We break no new ground in observing that an essential component of procedural fairness is an opportunity to be heard. In re Oliver, 333 U. S. 257, 273 (1948); Grannis v. Ordean, 234 U. S. 385, 394 (1914). That opportunity would be an empty one if the State were permitted to exclude competent, reliable evidence *330bearing on the credibility of a confession when such evidence is central to the defendant’s claim of innocence. In the absence of any valid state justification, exclusion of this kind of exculpatory evidence deprives a defendant of the basic right to have the prosecutor’s case encounter and 'survive the crucible of meaningful adversarial testing.’ United States v. Cronic, 466 U. S. 648, 656 (1984). See also Washington v. Texas, supra, at 22-23.” Crane v. Kentucky, 476 U. S., at 690-691.
“Few rights are more fundamental than that of an accused to present witnesses in his own defense. E. g., Webb v. Texas, 409 U. S. 95 (1972); Washington v. Texas, 388 U. S. 14, 19 (1967); In re Oliver, 333 U. S. 257 (1948). In the exercise of this right, the accused, as is required of the State, must comply with established rules of procedure and evidence designed to assure both fairness and reliability in the ascertainment of guilt and innocence. Although perhaps no rule of evidence has been moré respected or more frequently applied in jury trials than that applicable to the exclusion of hearsay, exceptions tailored to allow the introduction of evidence which in fact is likely to be trustworthy have long existed. The testimony rejected by the trial court here bore persuasive assurances of trustworthiness and thus was well within the basic rationale of the exception for declarations against interest. That testimony also was critical to Chambers’ defense. In these circumstances, where constitutional rights directly affecting the ascertainment of guilt are implicated, the hearsay rule may not be applied mechanistically to defeat the ends of justice.” Chambers v. Mississippi, 410 U. S., at 302.
“Moreover, there are other principles by which a defendant may occasionally avail himself of conduct as evidence in his favor — in particular, of conduct indicating consciousness of innocence,... of utterances asserting his innocence . .. , and, in sedition charges, of conduct indicating a loyal state of mind ....” 1A J. Wigmore, Evidence §56.1, p. 1180 (Tillers rev. ed. 1983); see United States v. Reifsteck, 841 F. 2d 701, 705 (CA6 1988).
“Mariotta’s basic defense was that he was unaware of any criminal wrongdoing at Wedteeh, that he was an innocent victim of the machinations of the sophisticated businessmen whom he had brought into the company to handle its financial affair's. That defense was seriously in issue as to most of the charges against him, drawing considerable support from the evidence....
“With the credibility of the accusations about Mariotta’s knowledge of wrongdoing seriously challenged, evidence of his denial of such knowledge in response to an opportunity to obtain immunity by admitting it and implicating others became highly significant to a fair presentation of his defense....
‘Where evidence of a defendant’s innocent state of mind, critical to a fair adjudication of criminal charges, is excluded, we have not hesitated to order a new trial.” United States v. Biaggi, 909 F. 2d 662, 691-692 (CA2 1990); see also United States v. Bucur, 194 F. 2d 297 (CA7 1952); Herman v. United States, 48 F. 2d 479 (CA5 1931).
Raskin, Honts, & Kireher, The Scientific Status of Research on Polygraph Techniques: The Case for Polygraph Tests, in 1 Modern Scientific Evidence 572 (D. Faigman, D. Kaye, M. Saks, & J. Sanders eds. 1997) (hereinafter Faigman) (compiling eight laboratory studies that place mean accuracy at approximately 90%); id., at 575 (compiling four field studies, scored by independent examiners, that place mean accuracy at 90.5%); Raskin, Honts, & Kireher, A Response to Professors Iacono and Lykken, in Faig-man 627 (compiling six field studies, scored by original examiners, that place mean accuracy at 97.5%); Abrams, The Complete Polygraph Handbook 190-191 (1989) (compiling 13 laboratory studies that, excluding inconclusive results, place mean accuracy at 87%).
Iacono & Lykken, The Scientific Status of Research on Polygraph Techniques: The Case Against Polygraph Tests, in Faigman 608 (compiling three studies that place mean accuracy at 70%).
E. g., Iacono & Lykken, The Case Against Polygraph Tests, in Faig-man 608-609; Raskin, Honts, & Kireher, A Response to Professors Iacono and Lykken, in Faigman 621; Honts & Perry 362; Abrams, The Complete Polygraph Handbook, at 187-188,191.
See, e.g., Judge Gonzalez’s careful attention to the relevance inquiry in the proceedings on remand from the Court of Appeals decision in Picd-noima. 729 F. Supp. 1336 (SD Fla. 1990).
One study compared the accuracy of fingerprinting, handwriting analysis, polygraph tests, and eyewitness identification. The study consisted of 80 volunteers divided into 20 groups of 4. Fingerprints and handwriting samples were taken from all of the participants.
In each group of four, one person was randomly assigned the role of “perpetrator.” The perpetrator was instructed to take an envelope to a building doorkeeper (who knew that he would later need to identify the perpetrator), sign a receipt, and pick up a package. After the “crime,” all participants were given a polygraph examination.
The fingerprinting expert (comparing the original fingerprints with those on the envelope), the handwriting expert (comparing the original samples with the signed receipt), and the polygrapher (analyzing the tests) sought to identify the perpetrator of each group. In addition, two days after the "crime,” the doorkeeper was asked to pick the picture of the perpetrator out of a set of four pictures.
The results of the study demonstrate that polygraph evidence compares favorably with other types of evidence. Excluding “inconclusive” results from each test, the fingerprinting expert resolved 100% of the cases correctly, the polygrapher resolved 95% of the cases correctly, the handwriting expert resolved 94% of the eases correctly, and the eyewitness resolved only 64% of the eases correctly. Interestingly, when “inconclusive” results were included, the polygraph test was more accurate than any of the other methods: The polygrapher resolved 90% of the cases correctly, compared with 85% for the handwriting expert, 35% for the eyewitness, and 20% for the fingerprinting expert. WIdaeki & Horvath, An Experimental Investigation of the Relative Validity and Utility of the Polygraph Technique and Three Other Common Methods of Criminal Identification, 23 J. Forensic Sciences 596, 596-600 (1978); see also Honts & Perry 865.
The Government argues that there is a widespread danger that people will learn to “fool” the polygraph, and that this possibility undermines any claim of reliability. For example, the Government points to the availability of a book called Beat the Box: The Insider’s Guide to Outwitting the Lie Detector. Tr. of Oral Arg. 53; Brief for United States 25, n. 10. Beat the Box, however, actually cuts against a per se ban on polygraph evidence. As the preface to the book states:
“Dr. Kalashnikov [the author] is a polygraph professional. If you go up against him, or someone like him, he’ll probably catch you at your game. That’s because he knows his work and does it by the book.
“What most people don’t realize is that there are a lot of not so professional polygraph examiners out there. It’s very possible that you may be tested by someone who is more concerned about the number of tests he will run this week (and his Christmas bonus) than he is about the precision of each individual test.
“Remember, the adage is that you can’t beat the polygraph system but you can beat the operator. This book is gleefully dedicated to the idea of a sporting chance.” V. Kalashnikov, Beat the Box: The Insider’s Guide to Outwitting the Lie Detector (1983) (preface); id., at 9 (“[W]hile the system is all but unbeatable, you can surely beat the examiner”).
Thus, Beat the Box actually supports the notion that polygraphs are reliable when conducted by a highly trained examiner — like the one in this case.
Nonetheless, some research has indicated that people can be trained to use “countermeasures” to fool the polygraph. See, e. g., Honts, Raskin, & Kircher, Mental and Physical Countermeasures Reduce the Accuracy of Polygraph Tests, 79 J. Applied Psychology 252 (1994). This possibility, however, does not justify a per se ban. First, research indicates that individuals must receive specific training before they can fool the polygraph (i e., information alone is not enough). Honts, Hodes, & Raskin, Effects *336of Physical Countermeasures on the Physiological Detection of Deception, 70 J. Applied Psychology 177, 185 (1985); see also Honts, Raskin, Kircher, & Hodes, Effects of Spontaneous Countermeasures on the Physiological Detection of Deception, 16 J. Police Science and Administration 91, 93 (1988) (spontaneous countermeasures ineffective). Second, as countermeasures are discovered, it is fair to assume that polygraphers will develop ways to detect these countermeasures. See, e. g., Abrams & Davidson, Counter-Countermeasures in Polygraph Testing, 17 Polygraph 16, 17-19 (1988); Raskin, Honts, & Kircher, The Case for Polygraph Tests, in Faigman 577-578. Of course, in any trial, jurors would be instructed on the possibility of countermeasures and could give this possibility its appropriate weight.
Indeed, research indicates that jurors do not “blindly” accept polygraph evidence, but that they instead weigh polygraph evidence along with other evidence. Cavoukian & Heslegrave, The Admissibility of Polygraph Evidence in Court: Some Empirical Findings, 4 Law and Human Behavior 117, 123, 127-128, 130 (1980) (hereinafter Cavoukian & Hesle-grave); see also Honts & Perry 366-367. One study found that expert testimony about the limits of the polygraph “completely eliminated the effect of the polygraph evidence” on the jury. Cavoukian & Heslegrave 128-129 (emphasis added).
See n. 5, supra.
See N. M. Rule Evid. § 11-707.
It has been suggested that if exculpatory polygraph evidence may be adduced by the defendant, the prosecutor should also be allowed to introduce inculpatory test results. That conclusion would not be dictated by a holding that vindicates the defendant’s Sixth Amendment right to summon witnesses. Moreover, as noted above, studies indicate that exculpatory polygraphs are more reliable than inculpatory ones. See n. 22, supra. In any event, a concern about possible future legal developments is surely not implicated by the narrow issue presented by the holding of the Court of Appeals for the Armed Forces in this ease. Even if it were, I can see nothing fundamentally unfair about permitting the results of a test taken pursuant to stipulation being admitted into evidence to prove consciousness of guilt as well as consciousness of innocence.