Institute of Pennsylvania Hospital and University of Pennsylvania
ABSTRACT
Fifteen college students attempted to deceive a professional polygraph examiner, while 15 others who had nothing to hide also submitted to the examination. The examiner was blind as to whether each subject was deceptive or truthful. Using the skin conductance response (SCR), significant discrimination was made between deceptive and truthful subjects with both "guilty person" and "guilty knowledge" polygraph tests. On both types of test, however, subjects who were not detected were significantly less socialized (Socialization Scale of the California Psychological Inventory) than those who were detected. This reduced susceptibility to detection was mediated by a reduced SCR to deception among low-socialization subjects. Among innocent subjects the highly socialized were more responsive electrodermally throughout the test, leading some of them to be misclassified as deceptive on at least one test. Implications of the results for both detection of deception and the construct of socialization are discussed.
DESCRIPTORS: Socialization, Skin conductance, Detection of deception.
The psychophysiological detection of deception depends upon the subject giving larger autonomic responses to questions associated with his deception than to appropriate control questions. Research over the past two decades has led to a better understanding of the factors which produce such differential responsivity among deceptive subjects. For example, the attempt to deceive (Gustafson & Orne, 1963; 1965a; Kugelmass, Lieblich, & Bergman, 1967) and mere exposure to questions associated with the subject's deception (Gustafson & Orne, 1965b; Kugelmass et al., 1967) have been shown to be sufficient to produce differential responsivity, and the more thoroughly a subject processes the test items, as indexed by later memory, the more often he is detected (Waid, Orne, Cook, & Orne, 1978). Such research has been important in revealing the components of the subject's behavior and the test procedures which are responsible for the differential responsivity which lead to detection.
Another class of variables which is also likely to influence differential responsivity -- attributes of the subject such as personality -- has been less thoroughly investigated (Barland & Raskin, 1973; Orne, Thackray, & Paskewitz, 1972). One personality construct of particular interest, both on the basis of face validity and because of relevant prior work, is socialization. Poorly socialized individuals are characterized by tendencies toward impulsivity, lack of restraint, superficial interpersonal relationships, and a history of interpersonal conflict despite normal intelligence, good superficial social skills, and an absence of neuroticism and social anxiety (Gough, 1960). The behavior patterns of poorly socialized individuals suggest that deception is not unusual for them and they might be less aroused while attempting deception and consequently less easily detected.
Preliminary reports of the present study were presented at the meetings of the Society for Psychophysiological Research, San Diego, 1976, and the Eastern Psychological Association, Boston, 1977.
We wish to thank Joseph Brophy for lending his professional expertise to the role of polygraph examiner and several colleagues for their substantive comments during the preparation of the manuscript: Frederick J. Evans, A. Gordon Hammer, Pamela A. Markowsky, Emily Carota Orne, Helen M. Pettinati, Betsy E. Lawrence, and R. Lynn Horne.
Special appreciation is due to Jeremy P. DeLong and Anna M. Waid for their scoring of the present data.
Address requests for reprints to: William M. Waid, Unit for Experimental Psychiatry, 111 N. 49th Street, Philadelphia, PA 19139.
15
16 WAID, ORNE, AND WILSON
From an empirical perspective, the potential importance of individual differences in socialization is underlined by the finding of Waid (1976) that college students scoring low on the socialization scale (Gough, 1964) of the California Psychological Inventory (CPI) gave smaller skin conductance responses (SCRs) to noxious noise bursts than did subjects scoring high on socialization. The high and low groups did not differ, however, in the amplitude of the SCR to innocuous 68 dB tones, indicating that the low-socialization subjects were characterized by a reduced SCR specifically to noxious stimuli rather than by a generalized reduction of electrodermal responsivity. Similar findings have been reported for the related construct of psychopathy. Hare (1978) has recently reviewed the many studies finding prison inmates diagnosed as psychopathic to give smaller SCRs than nonpsychopathic prisoners or nonprisoner control subjects in anticipation and in response to noxious stimuli.
The reduced differential electrodermal responsivity of low socialization subjects strongly suggests the relevance of socialization to the detection of deception since the latter depends precisely upon a differentially large response to relevant as opposed to control questions. The present paper examines the susceptibility to electrodermal detection of deception of subjects differing on the CPI index of socialization. The results have implications for an understanding of socialization as well as for the detection of deception. If the reduced SCR to noxious, disturbing stimuli shown by low-socialization subjects (Waid, 1976; Hare, 1978) plays a functional role in their social behavior, than it should be observed in response to social stimuli or behavior such as deception which might be conceptualized as noxious or disturbing.
Since the comparative utility of laboratory models reflecting two basic approaches to the detection of deception is a matter of some controversy (Lykken, 1974; Podlesny & Raskin, 1977), tests representative of both approaches were administered. The "Guilty Person Test" simulates the "control-question" field lie detector test (Reid & Inbau, 1977). 1 For example, if the theft of $600 were the subject of the test, the subject might be asked, "Did you take the $600?" The subject would be considered deceptive if he responded more to this question than to a control question about other possible thefts in his life. In the "Guilty Knowledge Test, " in contrast, the subject is asked about several items of information which a guilty but not an innocent person would know. Using the same example, but assuming that the amount of money stolen had not been made public, one critical item could be tested by asking the subject, "Was the amount stolen $100... $400... $500... $600... $700... or $800?" The guilty individual would be expected to respond more to $600 than to the control amounts, whereas for innocent individuals there should be no difference among the amounts. (This procedure is referred to as "peak-of-tension" in the field literature.) Lykken (1974) has urged the wider use of this approach, but with multiple critical items, in the field, and Reid and Inbau (1977) describe its increased, apparently successful, use in the field. Unfortunately, the necessary conditions for the guilty knowledge test are frequently difficult to attain in field situations. Nonetheless, the present purpose was not to compare these approaches but to permit the best possible test of the effect of level of socialization. Socialization might affect detection on one type of test but not on the other. Since we were not seeking to compare the two types of test and since, when used in the field, the guilty knowledge procedure is usually given after at least two guilty person tests, we administered the guilty person tests first to each subject followed by the guilty knowledge tests. A peak-of-tension test, i. e., a guilty-knowledge-type test with only a single item, was also administered.
Method
Subjects
Subjects were 30 male college students, aged 18 to 28, who volunteered to participate in a polygraph-test study. Age was uncorrelated with socialization (r = -.004) and was not considered further.
Apparatus
Skin conductance recordings were obtained using a constant .75-V source (Beckman Skin Conductance Coupler) applied to a pair of .64 cm2 Beckman biopotential electrodes attached to the palmar surfaces of the first phalanx of the third and fourth fingers of the left hand. Johnson & Johnson K-Y Jelly was used as the electrode medium (Edelberg, 1967). The skin was cleaned with acetone prior to electrode placement, and adhesive rings were used to maintain a uniform area of contact with the skin.
Procedures
Each subject served as either a "guilty" person carrying a memorized list of "code words" (N=15) or an "innocent" person suspected of carrying "code" words (N= 15). Experimenter 1 explained the subject's task and supervised guilty subjects' "overlearning" of 6 code words. To learn the 6 code words, they were required to associate them, reproduce them in the original, reverse and alphabetical order, and perform a number of interpolated tasks
17 SOCIALIZATION AND DETECTION OF DECEPTION
prior to again reproducing them.2 Subjects were told that failure to reproduce all 6 words correctly would disqualify them from the experiment. The procedure took approximately 1 hr and was designed to ensure overleaming as well as increase the subject's ego involvement in the task. "Innocent" subjects were informed that the polygraph examiner would suspect them and that it was often very difficult to prove one's innocence in a "lie detection" test. As a test of their "ability to perform under stress," the innocent subjects completed the same timed, interpolated tasks as did their guilty counterparts, but they, of course, learned no words. Since in a life situation the innocent suspect is often as concerned as the guilty, it is crucial in a laboratory model to involve the control subjects in the procedure lest one detect the latter on the basis of their lack of concern and relatively low arousal. Therefore, care was taken to have innocent subjects view their role as the experimental group by emphasizing our interest in determining the likelihood of false positives.
At the completion of the hour, subjects were informed that they would now be examined. Innocent subjects were reminded that they had to prove their innocence by not responding physiologically. Guilty subjects were told that the polygrapher would do his best to obtain a confession, that it would be difficult to deceive him, but that it was possible to "beat the polygraph"; highly intelligent, mature individuals would be able to control their emotions well enough to succeed.
Polygraph Test. The subject was then turned over to Experimenter 2, a professional polygraph examiner, who had no prior knowledge of whether he was innocent or guilty nor of the six code words. He attached several transducers, including those connected to a Stoelting field polygraph situated in the room.3 Two electrodes unobtrusively terminated in an adjoining room, where Experimenter 3 simultaneously recorded the SCR on a Beckman laboratory polygraph. Experimenter 2 first conducted a pretest interview which included biographical and basic medical-history questions followed by a review of the matter under investigation and a review of the relevant questions which would be asked on the first test. The control questions were not reviewed at this time to permit comparison of a test with unreviewed control questions with a subsequent test with reviewed control questions. Experimenter 2 administered several distinct tests to each subject. Between tests the experimenter left the room for 3 min. Upon returning he told the subject there was still suspicion about possible deception and that another series of questions would be asked. The interval between questions within a test was 12 to 14 sec. Every question could be answered yes or no. The tests were as follows:
a) Guilty Person Test. The subject was asked 3 questions directly bearing on his guilt or innocence (e.g., "Are you a courier carrying critical code words?"), interpolated with 2 control questions (e.g., "In your entire life did you ever steal even one thing?") and several irrelevant questions (e.g., "Are you in the United States?").4 Following the test the examiner questioned the subject briefly about his answers, particularly to the control questions. Then a "card test" was presented as a demonstration of the accuracy of the polygraph, as in field polygraphy. Since some subjects report purposely increasing their response on the card test in order to give a false impression of their response to deception, the results are not comparable to the other tests and are not presented. A second guilty person test was then conducted. The first guilty person test involved unreviewed control questions, whereas on the second test the subject did know the wording of the control questions.
b) Peak-of-Tension Test. The subject was asked, "Are there 2 critical code words?" and subsequently, 3, 4, 5, 6, or 7 critical code words. The same sequence was then asked again, but in the reverse order. Since couriers were concealing 6 code words, they were lying about information incidental to their role when they responded "no" to the number 6, whereas innocent subjects had no knowledge of the number of code words. The questions were reviewed prior to the test so the subject knew when the critical item would occur.
c) Guilty Knowledge Test. Subjects were asked if any of 24 words had special meaning to them as code words. Six of the words were the subject's code words. The list of words was repeated 4 times, so the number of detections per subject could range from 0 to 24.
Startle Stimulus. Ten to 15 sec following the last question on the final test, Experimenter 2 instructed the subject to remain still with eyes closed. Five to 10 sec later Experimenter 2 clapped his hands loudly. (This pro-
3 The Stoelting field polygraph served primarily as part of the stimulus context in the present study since its presence enhances the subject's awareness of the monitoring of his "involuntary" responses. The data presented here, however, are from the Beckman skin conductance recordings.
4 The use of the control-question test in laboratory research represents a semantic problem which appears irreconcilable within the limits of current laboratory approaches. According to field polygraphers the control question should deal with a subject similar to that of the relevant questions, as described in the example above concerning theft. In the laboratory "mock crime" or "courier" context, however, the subject has not actually stolen anything or committed any other transgression. Thus, the control question concerning the possible thefts in his life does not deal with the same subject matter as the relevant question. The control question deals with actual transgressions whereas the "mock crime" or "courier" questions obviously do not. Despite this inconsistency, the data presented below indicate that such control questions permit significant discrimination of truthful and deceptive laboratory subjects.
18 WAID, ORNE, AND WILSON
cedure was added only for the last 26 subjects.) Perfect standardization of this stimulus through tape recording was sacrificed in order to keep equipment in the room at a minimum and to preclude subject suspicion of physiological recording or other activity in the adjacent room. Nonetheless, Experimenter 2 was reasonably consistent. Measurement of 10 hand claps gave a mean sound pressure level of 91.3 dB (re .0002 dynes/cm 2) at the location of the subject's ears with a standard deviation of 1.8.
Conclusion. Experimenter 2 then informed the subject that the testing was at an end, and, without communicating anything concerning his decision, conducted the subject to the reception room. The subject then completed the Socialization Scale (which was identified to the subject only as California Psychological Inventory: Scale VII) and several other questionnaires assessing his reactions to the experiment. Finally, Experimenter 4 conducted a post-experimental interview with each subject. During that time, all questions concerning the experiment were answered.
Data Quantification
Amplitude of the SCR. Amplitude of the SCR was scored as the change in µmhos from onset to peak of the phasic change in conductance beginning in a period 1.5 sec following stimulus onset. The criterion for a scorable response was change of at least .025 µmhos conductance. If no scorable response occurred, a value of 0 was given that response. On the Guilty Person Tests this period extended 9 sec following stimulus onset since the questions were relatively long. On the remaining tests the period extended to only 5 sec following stimulus onset since each "question" was merely a single word or number. The scoring was done twice, independently, by two assistants who were blind as to subjects' "guilt" or "innocence," code words, and level of socialization. Discrepancies greater than .5 mm were resolved by Experimenter 3.
Detection. Since each test had a different format, a different criterion for detection was required for each. On both guilty person tests subjects were classified as "deception indicated" if they gave a larger SCR to any of the 3 critical questions than the largest SCR to any of the control questions. This is similar to some field polygraph procedures (Reid & Inbau, 1977). In some scoring systems (Podlesny & Raskin, 1977) subjects who are rated as responding equally to control and relevant questions are classified as "inconclusive. " For research purposes we view it more desirable to use the actual, objective, amplitude values, rather than a rating scheme (Podlesny & Raskin, 1977), and classify subjects as "no deception indicated" if their responses are not larger to the relevant than to the control questions. Further, to permit a sensitive test of the effect of socialization we report results separately for each test rather than a global final decision reflecting a combination of all the tests as typically would be done in the field.
On the Peak-of-Tension Test subjects were classified as deception indicated if they gave a larger SCR to the critical number than the largest SCR to any of the remaining numbers (except the first number in each series) on either or both presentations of the two series. On the Guilty Knowledge Test a code word was considered detected if it evoked a larger response than the largest response to any of the 3 control words in that category. Subjects who had code words detected 7 or more times (6 = chance rate of detection) were classified as deception indicated. Subjects who had code words detected 6 or fewer times were classified as "no deception indicated."
Results
Accuracy of Detection
Each type of test except the Peak-of-Tension Test significantly discriminated between guilty and innocent subjects (Table 1). On all tests, a much larger percentage of guilty than of innocent subjects were detected as deceptive. That is, guilty subjects produced larger SCRs to critical than to control stimuli, whereas innocent subjects did not. The present results are not intended as a comparison of the guilty person, peak-of-tension, and guilty knowledge techniques. Detection was depressed on the peak-of-tension and guilty knowledge tests because they always occurred after guilty person tests, as would be done in most current field procedures. There was no difference between Guilty Person Test 1, in which the control questions were unreviewed, and Guilty Person Test 2 in which these questions had been reviewed.
Socialization and Detection
Guilty Subjects. On each type of test except the second guilty person test, deceptive subjects who escaped detection scored significantly lower on the Socialization Scale than subjects who were detected. That is, subjects who tended to respond electrodermally more when deceiving than when answering control questions were significantly more so-
19 SOCIALIZATION AND DETECTION OF DECEPTION
cialized than those who did not. These results are presented in Table 2. 5
It might be hypothesized that either the demonstration of the polygraph's effectiveness which preceded the second guilty person test or the use of reviewed control questions on this test may have counteracted the effects of socialization. The present study was not designed to test this hypothesis. However, since the effects of socialization were present on later tests, in which both critical and control items were reviewed, any such effects were only transient.
Innocent Subjects. As can be seen in Table 1, on each separate test few innocent subjects were misclassified as deceptive. However, 6 of 15 subjects were misclassified as deceptive on one guilty person test, 3 on the first guilty person test, and 3 different subjects on the second guilty person test. The mean socialization score of these 6 subjects (M=38.8, SD=4.1) was significantly higher than that of those correctly classified as innocent on both tests (M=32.8, SD=7.1, t(14)=2.06, p<.05). One other was misclassified on the peak-of-tension and 1 on the guilty knowledge test. (Two of the 3 subjects misclassified on the peak-of-tension test were among those misclassified on the guilty person tests.) When all 8 misclassified subjects (M=38.5, SD=4.6) were compared with the 7 remaining subjects (M=31.6, SD=6.9), the results were also significant (t(14)=2.26, p<.025).
Comparison with Normative Data. The mean socialization scale score reported by Gough (1964) for 1,745 male college students was 37.41 (SD=5.28). This is similar to the mean socialization scale scores (36.1 to 40.3) of "guilty" subjects classified as "deception indicated," in Table 2. Excluding guilty person test 2, the means of deceptive subjects classified as no deception indicated (30.3-33.1) are close to or more than one standard deviation below the normative mean for college males and approximate the normative score for "high school disciplinary problems" (M=31.25, SD=5.4).
Socialization and Amplitude of the Skin Conductance Response
The mean SCR of each guilty subject to critical questions and the mean SCR of each innocent subject to all questions on the four tests was computed and expressed as the square root of the mean SCR.
As anticipated, the reduced susceptibility to detection of low-socialization subjects was mediated by a reduced SCR accompanying deception. Among guilty subjects the correlation between socialization and the mean SCR accompanying deception across the four tests was r (13)=.45, p <.05. The correlation between Socialization and the mean SCR to neutral questions (i.e., the irrelevant question on the guilty person tests and the control items on the remaining tests) was not significant, however, r(13)=.31. These results are analogous to those of Waid (1976) who found low socialization subjects to give significantly smaller SCRs to noxious noise bursts than high socialization subjects, but found no significant difference between groups in response to innocuous tones.
The tendency of high-socialization, innocent subjects to be misclassified as deceptive was mediated by their larger SCRs to the test questions. The correlation of socialization, with the mean SCR to all test questions across the four tests was r(13)= .56, p <.025.
Socialization and Response to a Startle Stimulus
Low-socialization subjects showed a significantly smaller SCR to an unexpected, loud stimulus -- a hand clap -- than did high-socialization subjects. The correlation between socialization and the SCR for the 26 subjects for whom the latter measure was available was r (24)=.46, p <.01. This result replicates related findings by Waid (1976) and Hare (1978).
20 WAID, ORNE, AND WILSON
Socialization and Skin Conductance Levels
The amplitude of the SCR is typically independent of skin conductance level (SCL) (Edelberg, 1967; Venables & Christie, 1973). Nonetheless, SCL just prior to the initial question of each test was measured and correlated with Socialization score to test for any adventitious effect of base level of conductance. For guilty subjects these four correlations were -.02, -.03, -.03, and -.02. For innocents the correlations were .48, .54, .49, and .56, none of which were significant (two-tailed). None of the correlations are negative, confirming that the relationship between socialization and detection is not an artifact of the law of initial values.
Discussion
From the present data, it appears that the relatively poorly socialized individual is less likely to be detected by his differential SCR while deceiving than his highly socialized counterpart. A highly socialized, innocent subject, on the other hand, may, due to his greater overall electrodermal responsivity in the situation, be more likely to be misjudged as deceptive than his less socialized counterpart. Although the effect of socialization might not be the same with cardiovascular or respiratory measures, it should be noted that the skin conductance response has been found to be the most accurate indicator of deception both in the laboratory (Barland & Raskin, 1975; Cutrow, Parks, Lucas & Thomas, 1972; Thackray & Orne, 1968) and in one field study which focused on this issue (Barland, 1975). Whether the relationship between socialization and detection would manifest itself in most field settings, however, can be determined only by further research.
Raskin and Hare (1978) recently reported finding no effect of socialization on the detection of deception among prison inmates. The apparent conflict between their results and the present ones may be due to the large number of methodological differences between the two studies. Lykken (1978) has described some of the potential confounding factors involved in having prison inmates attempt to deceive a polygraph examiner in order to win prize money, as was the case in Raskin and Hare (1978). Further, since the focus of that study was the clinical syndrome of psychopathy, all of the subjects were preselected as high or low on clinical ratings of psychopathy. High and low socialization subjects within each of these categories were then compared. In contrast, after separating prison inmates solely on the basis of socialization, Schalling, Lidberg, Levander, and Dahlin (1973) found, as predicted, less skin conductance activity between presentations of a noxious stimulus among low-socialization prisoners. These results suggest that the negative results of Raskin and Hare (1978) probably relate to either the selection problems mentioned above or to procedures used in the detection of deception rather than to an ineffectiveness of the Socialization Scale in prison samples. Clinical diagnosis of psychopathy was also unrelated to detection. This result does not necessarily conflict with the present results since the correlation between socialization and Raskin and Hare's clinical diagnosis was only -.31. Furthermore, as Lykken emphasized (1978) in a discussion of these data, psychopathy might have had an effect in the absence of monetary inducements (Lykken, 1978).6
Podlesny and Raskin (1977) have asserted that the setting used in the present experiment is not an adequately motivating context for the detection of deception. From this view, it might be argued that socialization had its present effects only because of relatively unmotivating conditions. Such an interpretation, however, is difficult to reconcile with the accuracy of classification of both "guilty" and "innocent" subjects. The present accuracy of 80% of both innocents and guilties on each guilty person test falls in the range of those of other studies using the guilty person test in a "mock crime" context (Barland & Raskin, 1975; Podlesny & Raskin, 1978; Raskin & Hare, 1978).
The present findings appear particularly stable in it light of the fact that they held true under three quite different types of tests. Low-socialization, deceptive subjects were less susceptible to detection regardless of whether the type of test was guilty person, peak-of-tension, or guilty knowledge. Further support for the reliability of these findings comes from the fact that low-socialization subjects in the present sample were less responsive to a loud, unexpected noise than were high-socialization subjects, replicating findings by Waid (1976) and Hare (1978).
Nonetheless, socialization score did not predict a deceptive subject's detection on each test. Only 3 deceptive subjects, each scoring relatively low on socialization, escaped detection on the first guilty person test, and the 3 subjects who escaped detection on the second guilty person test were no less socialized than subjects who were detected on that test. Thus, despite the significant tendency for low-socialization subjects to escape detection, some low-socialization subjects were detected on some tests.
Although there was a tendency for innocent, high-socialization subjects to appear deceptive on
21 SOCIALIZATION AND DETECTION OF DECEPTION
individual tests, the results also suggest that present detection of deception procedures may minimize this tendency. Thus, it is noteworthy that no innocent subject was misclassified as deceptive on both guilty person tests, and none of the innocent subjects were misclassified on three or more tests. Consequently, basing final decisions on several tests reduces the effects of socialization on false positive decisions.
Several other technical issues concerning polygraph tests must be commented upon viz a viz the present results. First, the peak-of-tension test did not discriminate significantly between deceptive and truthful subjects. Undoubtedly, this was partially due to the fact that it was always administered relatively late in the session. It is also likely, however, that the incidental information under consideration was simply not very salient to the deceptive subjects. Secondly, although the guilty knowledge test discriminated between truthful and deceptive subjects, there was a high number of false negatives. The present study, however, was not designed as a comparison of guilty knowledge and guilty person tests. The guilty knowledge test was always administered last, resulting in reduced detection rates.
The present results also have broader implications regarding the construct of socialization and the role of electrodermal arousal in social behavior. If a person is less aroused electrodermally by lying, and related behaviors, it may be that such actions are easier for him and consequently he may be more likely to engage in them. The result of such reduced arousal to antinormative behavior might be the emergence of the behavior patterns characterized as under-socialized. The present findings that less socialized subjects give smaller SCRs when deceiving than do more socialized subjects, is consistent with this view. Only further research, however, can determine whether the reduced physiological component should be conceptualized as an index of an arousal-dampening process which facilitates such behavior or whether it should be conceptualized as a result of the facility with which the less-socialized person deceives.
In summary, the predicted effect of level of socialization on the electrodermal detection of deception was confirmed in a normal population: on individual tests, deceptive subjects who escaped detection were significantly less socialized than those who were detected. In addition, among innocent subjects there was a tendency for highly socialized subjects to be more responsive electrodermally throughout the test, leading some of them to be misclassified as deceptive on individual tests.
REFERENCES
Barland, G. H. Detection of deception in criminal suspects: A field validation study. Doctoral dissertation, University of Utah, 1975.
Barland, G. H., & Raskin, D. C. Detection of deception. In W. F. Prokasy & D. C. Raskin (Eds.), Electrodermal activity in psychological research. New York: Academic Press, 1973. Pp. 417-477.
Barland, G. H., & Raskin, D. C. An evaluation of field techniques in detection of deception. Psychophysiology, 1975, 12, 321-330.
Cutrow, R. J., Parks, A., Lucas, N., & Thomas, K. The objective use of multiple physiological indices in the detection of deception. Psychophysiology, 1972, 9, 578-588.
Edelberg, R. Electrical properties of the skin. In C. C. Brown (Ed,), Methods in psychophysiology. Baltimore: Williams & Wilkins, 1967. Pp. 1-53.
Gough, H. G. Theory and measurement of socialization. Journal of Consulting Psychology, 1960, 24, 23-30.
Gough, H. G. Manual for the California Psychological Inventory. Palo Alto, California: Consulting Psychologists Press, 1964.
Gustafson, L. A., & Orne, M. T. Effects of heightened motivation on the detection of deception. Journal of Applied Psychology, 1963, 47, 408-411.
Gustafson, L. A., & Ome, M. T. The effects of perceived role and role success on the detection of deception. Journal of Applied Psychology, 1965, 49, 412-417. (a)
Gustafson, L. A., & Ome, M. T. The effects of verbal responses on the laboratory detection of deception. Psychophysiology, 1965, 2, 10-13. (b)
Hare, R. D. Electrodermal and cardiovascular correlates of psychopathy. In R. D. Hare & D. Schalling (Eds.), Psychopathic behavior: Approaches to research. Chichester: Wiley, 1978. Pp. 107-144.
Ingersoll, B. D. Detection of deception in primary psychopaths. Unpublished doctoral dissertation, Pennsylvania State University, 1977.
Kugelmass, S., Lieblich, I., & Bergman, Z. The role of 'lying' in psychophysiological detection. Psychophysiology, 1967, 3, 312-315.
Lykken, D. T. Psychology and the lie detector industry. American Psychologist, 1974, 29, 725-739.
Lykken, D. T. The psychopath and the lie detector. Psychophysiology, 1978, 15, 137-142.
Ome, M. T., Thackray, R. I., & Paskewitz, D. A. On the detection of deception: A model for the study of the physiological effects of psychological stimuli. In N. S. Greenfield & R. A. Stembach (Eds.), Handbook of psychophysiology. New York: Holt, Rinehart & Winston, 1972. Pp. 743-786.
Podlesny, J. A., & Raskin, D. C. Physiological measures and the detection of deception. Psychological Bulletin, 1977, 84, 782-799.
Podlesny, J. A., & Raskin, D. C. Effectiveness of techniques and physiological measures in the detection of deception. Psychophysiology, 1978, 15, 344-359.
Raskin, D. C., & Hare, R. D. Psychopathy and detection of deception in a prison population. Psychophysiology, 1978, 15, 126-136.
Reid, J. E., & Inbau, F. E. Truth and deception. Baltimore: Williams & Wilkins, 1977.
22 WAID, ORNE, AND WILSON
Schalling, D., Lidberg, L., Levander, S. E., & Dahlin, Y. Spontaneous autonomic activity as related to psychopathy. Biological Psychology, 1973,1, 83-97.
Senter, R. J. Analysis of data. Glenview, Illinois: Scott Foresman and Company, 1969.
Thackray, R. I., & Ome, M. T. A comparison of physiological indices in detection of deception. Psychophysiology, 1968, 4, 329-339.
Thorndike, E. L., & Lorge, I. The teacher's word book of 30,000 words. New York: Columbia University Bureau of Publications, 1944.
Venables, P. H., & Christie, M. Mechanisms, instrumentation, recording techniques, and quantification of responses. In W. F. Prokasy & D. C. Raskin (Eds.), Electrodermal activity in psychological research. New York: Academic Press, 1973. Pp. 1-124.
Waid, W. M. Skin conductance response to both signalled and unsignalled noxious stimulation predicts level of socialization. Journal of Personality & Social Psychology, 1976, 34, 923929.
Waid, W. M., Ome, E. C., Cook, M. R., & Ome, M. T. Effects of attention,
as indexed by subsequent memory, on electrodermal detection of deception.
Journal of Applied Psychology, 1978, 63, 728-733.