In the Matter of the Complaint Against OMNITRONICS Omnitronics Research Corporation 3085 West Market Street at Akron, Ohio 44313 P.S. Docket No. 12/9; 10/07/82 Cohen, James A. APPEARANCES FOR COMPLAINANT: H. Richard Hefner. Esq. Steven B.Caver, Esq. Consumer Protection Division Law Department United States Postal Service Washington, DC 20260 APPEARANCES FOR RESPONDENT: Gary D. Pearch, Esq. William R. Holland, Esq. Omnitronics Research Corporation 3085 West Market Street Akron, Ohio 44313
Respondent has appealed from the Initial Decision of an Administrative Law Judge which holds that, with regard to the advertising and sale of a "Voice Stress Computer," it is engaged in a scheme to obtain money or property through the mail by means of materially false representations in violation of 39 U.S.C. § 3005.
On August 18, 1981, the Consumer Protection Division of the Law Department, United States Postal Service (Complainant), filed a Complaint alleging that Respondent was selling a voice stress analyzer through the mail, marketed under the trade name Voice Stress Computer (VSC), by means of false representations in violation of 39 U.S.C. § 3005. Specifically, Paragraph III of the Complaint alleges that Respondent falsely represents that:
"(a) The Voice Stress Computer is an effective and reliable means of detecting falsehood.
(b) The Voice Stress Computer may be relied upon in the conduct of business and personal affairs.
(c) The Voice Stress Computer is an accurate, reliable means of detecting and registering stress communicated by vocal utterance."
Respondent filed an Answer in which it denied that it makes the representations alleged in the Complaint or that it makes any representations which are materially false as a matter of fact. Thereafter, a hearing was held before an Administrative Law Judge at which both parties presented testimony with respect to the alleged representations. On April 23, 1982, the Administrative Law Judge issued an Initial Decision which found that Respondent makes the representations alleged in the Complaint and that they are false. Respondent appealed the Initial Decision to the Judicial Officer. Both parties have filed briefs.
Respondent argues that the Initial Decision should be reversed because:
1. The determination of the administrative law judge that the representations alleged in paragraph 3(a) and (b) of the Complaint are made in its advertisement is erroneous;
2. The determination of the administrative law judge that complainant satisfied its burden of proof and that issuance of a False Representation Order pursuant to 39 USC § 3005 was thereby justified is erroneous and not supported by even a preponderance of reliable and probative evidence.
Each of these grounds for appeal is discussed below.
1. The determination of the administrative law judge that the representations alleged in paragraph 3(a) and (b) of the complaint are made in its advertisement is erroneous.
In connection with this argument, Respondent contends that Findings of Fact II (a) and (b) and Conclusion of Law III are erroneous because the ordinary reader would not interpret its advertising as making the representations alleged in paragraphs III(a) and (b) of the Complaint. At the outset, it is noted that Respondent does not challenge the Administrative Law Judge's conclusion that the ordinary reader would interpret its advertising as making the representation set forth in paragraph III (c) of the Complaint. Moreover, for the reasons stated in the Initial Decision, it is clear this representation is made in Respondent's advertisement (CX-1; I.D. at 4).
With respect to paragraphs III (a) and (b) of the Complaint, Respondent argues that the advertisement on which the Administrative Law Judge based his findings does not expressly represent that the VSC detects falsehood or is a lie detector or that the VSC may be relied on in the conduct of business and personal affairs (Resp. Brief at 7). Respondent states that a fair reading of its advertisement, CX-1, in its entirety would indicate that the impact and thrust of the language utilized is that the VSC is a device for detecting stress (id.). Similarly, Respondent argues that the Administrative Law Judge misapplied Donaldson v. Read Magazine, Inc., 333 U.S. 178 (1948) in concluding that the average reader would interpret its advertisement as making the representations alleged in paragraph III of the Complaint.
The absence of an express statement that the VSC detects falsehood is not dispositive of whether Respondent's advertisement makes this representation. In reviewing an advertisement under 39 U.S.C. § 3005, both the advertisement's express and implied representations must be considered. N. Van Dyne Advertising, Inc. v. United States Postal Service, 371 F.Supp. 1373 (S.D.N.Y. 1974); Vibrabrush Corp. v. Schaffer, 152 F.Supp. 461 (S.D.N.Y. 1957).
The Administrative Law Judge correctly found that the headline in CX-1, "WHAT ARE THEY REALLY SAYING?", and the statements, in bold type, that the VSC " r eveals the hidden message in every voice" and will enable the user to " f ind out what they're really saying," as well as the statement that "the mere presence of the VSC promotes honesty," portrayed the VSC as a reliable means of detecting falsehood (I.D. at 2-3). Furthermore, the advertisement's discussion of the VSC's development also links the product's detection of stress with deception. The advertisement notes that the VSC was "developed by security scientist John Welsh, the man who coinvented the electronic merchandise-tagging system that foils thousands of shoplifters every day" and used "certain principles of voice stress discovered by American intelligence groups during World War II." Finally, it is noted that CX-1 appeared in the June 1981 issue of Law Enforcement Communications magazine (Tr. 5). The average reader of this magazine would have been primarily interested in purchasing the device as a means of preventing or spotting deception and would have interpreted the advertisement within this context.
The Administrative Law Judge also correctly found that Respondent's advertisement represents that the VSC may be relied on in business and personal situations. In reaching this conclusion the Administrative Law Judge cited, among other language, the advertisement's statements that ". . . Now you can detect hidden stress in others . . . before it costs you money, time and trouble." . . . "Now you can unmask this hidden stress, and you can do it easily and affordably" . . . and ". . . to take the first step away from those people-problems that can eat up so much time and money fill out the coupon and mail it today" (I.D. at 3).
Accordingly, Findings of Fact II(a) and (b) and Conclusion of Law III are affirmed.
2. The determination of the administrative law judge that complainant satisfied its burden of proof and that issuance of a False Representation Order pursuant to 39 U.S.C. § 3005 was thereby justified is erroneous and not supported by even a preponderance of reliable and probative evidence.
Respondent contends that Finding of Fact III and Conclusions of Law 4 and 5 of the Initial Decision are erroneous because the falsity of the representations alleged in the complaint have not been established by a preponderance of the evidence. Respondent argues that the Administrative Law Judge erred in discounting the testimony of its witnesses in favor of that given by Marilyn VanGraber, Ph.D., a communications consultant, and Postal
Inspector Norman Robbins, both of whom testified for the Complainant. It contends that the tests performed with the VSC by these witnesses, which served as the basis for their conclusions, were not reliable because " n either witness had ever used the VSC prior to such tests . . . neither witness possessed any expertise or background in medicine, physiology, audio phenomenon, electronics or electrical engineering . . . they had no idea what the VSC measured, . . . they conducted all tests without having first check the recording unit or tapes involved for non-coherent noise signal to noise ratio and other quantitative and qualitative qualities of recorded sound input . . . and neither witness had any idea whatsoever as to what was contained in the VSC - the embodied technology and how it worked" (Resp. Brief at 9-10).
Expert opinion testimony is admissible even where not based on personal knowledge or testing of the specific product in issue. Standard Research Labs, P.S. Docket Nos. 7/78, 7/86 (P.S.D. Oct. 27, 1980); Michigan Bulb Co., P.S. Docket No. 7/43 (P.S.D. Oct. 30, 1979), and cases cited therein. However, in this case, both witnesses tested the VSC. Their lack of prior experience with this device did not affect their ability to gauge its effectiveness. In this regard, Respondent's advertisement proclaims that " i t's simple for you . . . After reading the easy-to-follow instruction manual and practicing a little, you'll be able to distinguish high stress in the voice of almost anyone who is trying to control it" (CX-1).
Dr. Van Graber's lack of experience in the areas cited by Respondent did not disqualify her from evaluating the VSC. She was qualified at the hearing as an expert in the field of voice stress analysis without objection by Respondent's counsel (Tr. 19). While the four tests she performed with the VSC are not beyond question as to the procedures used and their statistical validity, the care with which these tests were undertaken, and the observations made, make them relevant, and, in light of all of the evidence in the record, persuasive. In this regard, Dr. VanGraber testified that: the testing was performed in an environment free from background noise (Tr. 20, 50); she used a sound deadening surface (Tr. 60); in conducting her tests, she used the guidelines given by the instruction manual (Tr. 38, 62); and her testing constituted "a very rigorous examination" (Tr. 35).
The question of technical expertise is more relevant with respect to Postal Inspector Robbins, who was not qualified as a voice stress analyst. He tested the VSC by using polygraph tape recordings which contained confirmed deceptive and non-deceptive information (Tr. 96). He was qualified to interpret the polygraph recordings used in the tests because of his experience as a polygraph examiner (Tr. 93-94, 105, 108). He also is presumed qualified to operate the VSC since, as noted, Respondent's advertisement represents that anyone can work the device. In this regard, Inspector Robbins testified that his tests were performed by following the instruction manual (Tr. 97, 106). In conducting these tests, he was as qualified as Respondent's expert witness, Sergeant Pennington, who is also a polygraph examiner and on whose tests and conclusions Respondent heavily relies.
Respondent notes that Dr. Williamson, inventor of the VSC, testified that the device is sensitive to background noises. When asked, in view of this sensitivity, how the device would be used to monitor business affairs, he stated: "Presumably you would not operate it where there is a lot of background noise" (Tr. 178). As noted, Dr. VanGraber testified that her testing was conducted in an environment free from background noise. Inspector Robbins testified that, in conducting his test, he adjusted the tape recorder's volume so that it would not affect the VSC (Tr. 106). If the degree of background noise existing during Inspector Robbins' test was sufficient to invalidate the VSC's responses, it is questionable whether the device is suitable for the purpose for which it is advertised, regardless of any merit it might have under purely clinical circumstances. Respondent's advertisement does not warn its readers that the device is unreliable where normal background noise exists. Indeed, the advertisement states: "Another plus: the computer's shirt-pocket size enables you to use it wherever you like" (CX-1). Moreover, the instruction manual states that " t he Voice Stress Computer will perform efficiently in any reasonably quiet surroundings as long as speech sounds are louder than background noise" (CX-5 at 5).
Respondent next argues that the four tests performed by Dr. VanGraber were not properly conducted. With regard to the first and third tests, involving face-to-face oral examinations and reactions to words, Respondent argues they should be discounted because she "prejudged that a subject or word would create psychological stress even though the subjects were friends and associates and were responding in nonjeopardy, confidential circumstances . . . Moreover, the oral examination tests concerned past recollection and not in vivo stress situations" (Resp. Brief at 10). However, Dr. VanGraber testified that the stress that a person feels with regard to a word is present stress, even though it may be engendered by a past experience (Tr. 42). The word association test involved 10 subjects, three of whom had marriage problems, one of whom had been accused of rape, and another who had been a rape victim (Tr. 84). Respondent's counsel agreed with Dr. VanGraber's statement that, with respect to rape, "stress is not only apparent during the act, it is something that's carried on for quite some time" (Tr. 43-44).
Respondent argues that Dr. VanGraber's second test, which consisted of reading numbers and letters, should be discounted because it "was admittedly not a test to determine the presence or absence of stress, but only to 'see how the unit was working'" (Resp. Brief at 10). To the contrary, the results of this test are very relevant to consideration of how the device functions. Based on 30 repetitions over a period of hours, Dr. VanGraber determined that the angle of the microphone to the subject had an impact on the number of red light responses by the VSC; that, because pitch was higher, female subjects generated more red light responses than male subjects; and that, if an utterance was elongated, more red lights would result (Tr. 82).
With respect to Dr. VanGraber's fourth test, which involved comparing the device's readings with 18 pre-recorded psychological evaluation tapes and two polygraph tapes, Respondent argues that the tapes could have been altered and that she did not know the results of subsequent polygraph examinations. These arguments are not persuasive. Dr. VanGraber testified that the tapes were continually in her custody and control from the time they were made until the test was performed and that they remained unaltered (Tr. 50-51). Her unfamiliarity with the results of polygraph tests given to some subjects of her psychological evaluation tapes does ot affect the validity of her testimony because the individuals were known to be deceptive (Tr. 54).
Respondent alleges that Inspector Robbins was biased because he admitted that he favors polygraph machines as opposed to voice stress analyzers. The witness stated, however, " w ell, I'm not exactly against voice stress analyzers . It's just that so far as my knowledge is concerned, I've never been able to find that the voice stress equipment really works the way the voice stress advocates say that it does" (Tr. 118). He went on to state that, in his opinion, voice stress analysis is ot as reliable as polygraph analysis (id.). This testimony is hardly sufficient to establish bias for the purpose of impeachment.
As Respondent further points out, Inspector Robbins admitted that his tests would be invalid if the VSC was found to be unresponsive to "yes" or "no" answers. However, Sergeant Pennington testified that the device worked with respect to "yes" or "no" answers during his tests (Tr. 228). Moreover, neither Respondent's advertisement nor the instruction manual states that the device will not work with "yes" or "no" answers.
The conclusions of Dr. VanGraber and Inspector Robbins, based on independent tests performed with the VSC, are consistent that the device does not accurately track instances of known deception (Tr. 81-88, 122-127). While Respondent argues that the Administrative Law Judge "totally ignored" contrary evidence presented by Mr. Williamson and Sergeant Pennington, the Initial Decision reflects that the Administrative Law Judge considered the testimony of these witnesses (I.D. at 5, 8-11). While he found that Sergeant Pennington "presented the most impressive testimony for Respondent," the Administrative Law Judge also concluded that this witness's enthusiasm was directed primarily at Respondent's larger unit, SA-3 (I.D. at 8-11) and that Respondent's witnesses did not overcome the testimony presented by Dr. VanGraber and Inspector Robbins.
A review of Sergeant Pennington's testimony supports the Administrative Law Judge's conclusion that the witness's endorsement of the VSC was less than enthusiastic. When asked whether the "Voice Stress Computer is as effective as any component of the polygraph machine in detecting stress" (Tr. 219), Sergeant Pennington turned the question around, stating that " s ingularly any one of the components would equal what the computer does" (id.). Finally, when asked whether he "will continue to use and rely on the Voice Stress Computer in conducting future examinations?", he stated: "Yes, sir. The SA-3 and the voice computer" (Tr. 220, emphasis added).
Moreover, Sergeant Pennington testified that the hand-held device "is not nearly as useful as the SA-3" (Tr. 223); that " t he small one has no way of keeping a recorded chart. There is no way to keep an accurate record of when a flash occurred" (Tr. 224); and indicated that extensive practice would be required before the average layman could use the VSC in his everyday business situation (Tr. 228). These statements contrast with the assurances in Respondent's advertisement that "It's siple for you. And tough for them" and that the VSC "shows you stress in voices immediately." While the witness also stated that the VSC showed 100% consistency on 8 tests when compared with the polygraph (Tr. 233-238), the prior comments offset this testimony.
Regarding Mr. Williamson's testimony, that witness also admitted that the VSC did not have the "high degree of precision" of the larger SA-3 (Tr. 158) and stated that, in the hand-held model, stress measurement "becomes more of a subjective impression" (Tr. 161).
Respondent also contends that the Administrative Law Judge unjustifiably rejected the testimony of Dr. Merle Szucs, who testified to the VSC's ability to measure stress relating to disease. The Administrative Law Judge found that Dr. Szucs' testimony was "unreliable" (I.D. at 11). Dr. Szucs admitted that he had no opinion as to whether the VSC detects stress caused by deception (Tr. 197). Since, as discussed previously, the thrust of Respondent's advertising, and the interest of the readers at which that advertising was aimed, is the device's ability to detect stress caused by deception, Dr. Szucs' testimony is to a large extent irrelevant.
Respondent further argues that "Voice stress analyzing devices are as effective and reliable as so-called polygraph devices as a means of detecting psychological stress and deception" (Resp. Brief at 17-18). However, Sergeant Pennington's testimony indicates that voice stress analyzers are not as accepted as polygraph devices. He stated that " i t's against the law in Alabama to use a voice stress computer analyzer by itself i.e., without a polygraph " (Tr. 229).
Respondent also argues that the Administrative Law Judge improperly relied on polygraph examination test results in considering the testimony of Dr. VanGraber and Postal Inspector Robbins. In support of this argument, Respondent asserts that " s ince the polygraph device was first used in 1919, the state and federal courts have uniformly ruled polygraph examination test results inadmissible as evidence for the very reason that the 'lie detector' has not attained scientific acceptance as a reliable and accurate means of ascertaining truth or deception" (Resp. Brief at 20). While polygraph test results may not be admissible in a criminal proceeding, it has not been shown they should be inadmissible in an administrative proceeding as a basis for comparison with the VSC. Furthermore, the polygraph test results relied on by Complainant's witnesses were confirmed by other evidence (Tr. 54, 96).
Respondent next argues that the VSC is entitled to a presumption of validity because it has been patented. In this regard, 35 U.S.C. § 282 states:
"A patent shall be presumed valid. Each claim of a patent (whether in independent, dependent, or multiple dependent form) shall be presumed valid independently of the validity of other claims; dependent or multiple dependent claims shall be presumed valid even though dependent upon an invalid claim. The burden of establishing invalidity of a patent or any claim thereof shall rest on the party asserting such invalidity."
Respondent has introduced two patents into evidence (Resp. Exhs. 2 and 3). The claims contained in these patents, in similar fashion, set forth the physical characteristics of the speech analyzers which Respondent desired to have patented. the speech analyzer referenced in Claim 1 of both patents is characterized as "determining the emotional state of a person." The speech analyzers described in Claim 7 of Respondent's Exhibit 2 and Claim 9 of Respondent's Exhibit 3 are characterized as having the purpose of "analyzing an FM demodulated speech signal." It is concluded, based on a review of these claims and the remainder of the patents, that the statutory presumption accorded to the validity of the patents themselves and the individual claims do not cover the much broader representations which are the subject of this proceeding.
Moreover, Howard Robbins, a patent attorney who testified for the Respondent, stated that Respondent's Exhibits 2 and 3 cover both the VSC and the larger SA-3 (Tr. 186) and that "The Voice Stress Computer design represents a portion of the technology implemented in the SA-3 desing" (Tr. 184, emphasis added). In view of this testimony, and that of Sergeant Pennington regarding the greater utility of the SA-3, supra, it is concluded that Respondent has not shown, on the basis of this record, that the VSC, as opposed to the SA-3, is entitled to a statutory presumption of validity. In this regard, it has been recognized that a claim, in order to be valid, should be "limited to a particular art or technology, to a particular apparatus or machinery, and to a particular end use." Arshall v. United States, 621 F.2d 421, 429 (Ct. Cl. 1980), cert. denied 449 U.S. 1077 (1981), reh. denied 450 U.S. 1050 (1981), quoting Gottschalk v. Benson, 409 U.S. 63, 64 (1972).
Finally, in view of the testimony discussed supra by both Respondent's and Complainant's witnesses, it is concluded that any statutory presumption regarding the validity of the VSC, as distinguished from the SA-3, that would pertain to the specific representations at issue has been rebutted by clear and convincing evidence. Cf. Saf-Guard Products, Inc. v. Service Parts, Inc., 532 F.2d 1266, 1271 (9th Cir.1976), cert. denied, 429 U.S. 896 (1976). Accordingly, Finding of Fact 3 and Conclusions of Law IV and V are affirmed.
After consideration of the entire record and Respondent's exceptions, it is concluded that Respondent is engaged in a scheme to obtain money through the mail by means of materially false representations. Accordingly, Respondent's appeal is denied. A remedial order under 39 U.S.C. § 3005 is being issued with this decision.