UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of white noise on state complexity and evaluative importance Lim, David Teck-Kai 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1987_A8 L55.pdf [ 4.38MB ]
JSON: 831-1.0097274.json
JSON-LD: 831-1.0097274-ld.json
RDF/XML (Pretty): 831-1.0097274-rdf.xml
RDF/JSON: 831-1.0097274-rdf.json
Turtle: 831-1.0097274-turtle.txt
N-Triples: 831-1.0097274-rdf-ntriples.txt
Original Record: 831-1.0097274-source.json
Full Text

Full Text

T H E E F F E C T S OF WHITE NOISE O N S T A T E C O M P L E X I T Y A N D E V A L U A T I V E I M P O R T A N C E By D A V I D T E C K - K A I L I M B.A. (Hons.), The University of Waterloo, 1984 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF M A S T E R OF ARTS in T H E F A C U L T Y OF G R A D U A T E STUDIES Department of Psychology We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A September, 1987 © David T. K . L im, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 1956 Main Mall Vancouver, Canada DE-6(3/81) Abstract The effects of cortical arousal on state complexity and evaluative importance were examined. Arousal was manipulated using two levels of white noise. In Study 1, a three-dimensional social domain was created using behavioral descriptions of eight fictitious people. In Study 2 , subjects memorized these descriptions, and later, from memory, made similarity judgments among these eight targets while being exposed to either loud or soft white noise. The first hypothesis was that loud noise would effect an increase in the relative importance of the evaluation dimension. The second hypothesis was that this increased use of evaluation would be a result of a reduction in state complexity-evidenced by the other dimensions becoming less important. The results fully supported the first hypothesis and partially supported the second. There was also some support for the hypothesis that trait complex compared to trait simple individuals would be more affected by loud noise. However, the prediction that sensitizers would be more affected by the loud noise than repressors was not supported. The limitations of the second stud-y and new directions for research are discussed. i i i TABLES OF CONTENTS Abstrac t i i L i s t of Tables v i L i s t of F igures v i i Acknowledgement v i i i Introduct ion - 1 Arousal 2 White Noise and Arousal 3 White Noise and Se lec t ive At tent ion 6 Cogn i t ive Complexity 8 Stress and Complexity 9 Dynamic Complexity Theory 10 Paulhus and Lim (1985) . 1 1 I n d i v i d u a l Di f ference Measures. 14 T r a i t Complexity 14 R e p r e s s i o n - S e n s i t i z a t i o n 15 The Present Studies 17 Study 1 18 Method 18 Subjects and Procedure 19 Results 20 Di scuss ion 25 Study 2 26 Method. 27 Subjects 27 i v Independent Measures 27 A. T r a i t Complexity. 27 B. Repres s ion -Sens i t i za t ion 28 Arousal Manipulation 28 D i s s i m i l a r i t y Rating Task 29 Procedure 30 R e s u l t s . 33 Discuss ion 49 Arousal and the Dynamic Complexity Model 49 A l t e r n a t i v e Explanat ions 50 Ind iv idua l Di f ferences and the Dynamic Complexity Model 51 General Discussion 54 Limitat ions 54 New Direc t ions 55 Footnotes 57 References. 59 V Appendix A Descr ipt ions of the 8 Soc ia l S t imul i 67 Appendix B C h a r a c t e r i s t i c s Rating Scale 68 Appendix C The 2 Blocks of the 28 pa irs 69 Appendix D Post-Experimental Questionnaire 70 Appendix E Corre la t ion Matrix for soft noise cond i t i on 71 Appendix F Corre la t ion Matrix for loud noise condi t ion 72 Appendix G ANOVA Summary Tables 73 i . Block and Time Ef fec t s 73 i i . ANCOVA: Unsealed E l - R a t i o 74 i i i . ANCOVA: Scaled E l - R a t i o 74 i v . ANCOVA: Unsealed Dimension Strength 75 v . ANCOVA: Scaled Dimension Strength 76 Appendix H Indiv idual d i f f erence Measures A. Paragraph Completion Test (PCT) 77 B. Mini Repress ion-Sens i t i za t ion S c a l e . . 79 v i LIST OF TABLES 1. Mean Ratings of the 8 targets on the three Dimensions 21 2. C o r r e l a t i o n among the independent and dependent measures 35 3. Mean o v e r a l l E l - R a t i o for the two Blocks and the two Times 37 v i i LIST OF FIGURES 1. Mean Ratings of the 8 Targets on the Evaluat ion and Potency Dimensions 22 2. Mean Ratings of the 8 Targets on the Evaluat ion and Industrious Dimensions 23 3. Conf igurat ion of the Eight S o c i a l St imul i along Dimensions One (Evaluation) and Two (Dominance) 38 4. Conf igurat ion of the Eight S o c i a l S t imul i along Dimensions One (Evaluation) and Three (Industriousness) 39 5. Conf igurat ion of the Eight S o c i a l St imul i along Dimensions" Two (Dominance) and Three (Industriousness) 40 6. E f f e c t s of Noise on the Dimension Strength of E v a l u a t i o n , Dominance, and Industriousness 42 7. E f f e c t s of Noise on the MULTISCALE Weights of E v a l u a t i o n , Dominance, and Industriousness 44 8. In terac t ion between T r a i t Complexity and Noise Condi t ion on the Unsealed E l - R a t i o 47 9. In terac t ion between T r a i t Complexity and Noise Condi t ion on the Scaled E l - R a t i o . . . . 4 8 Acknowledgements I would like to express my gratitude to a number of individuals who provided support and assistance in the completion of this thesis. First, to my adviser and friend Del Paulhus, for his constant support, encouragement, probing questions, and midnight meetings. Without his guiding hand, I would not have overcome my dominant responses and excesses. I would also like to thank my committee members: Peter Suedfeld, for his astute questions and comments that constantly kept me on my toes, and to Jennifer Campbell, for her helpful comments on statistics and experimental design. Finally, I'd like to thank Anita Sam: her support and encouragement kept me going that extra mile. Her patience in waiting for me till the early hours of the morning will always be remembered and appreciated. 1 The Effects of White Noise on State Complexity and Evaluative Importance. Introduction August 21st, 1981... Hill 306...Echo company had finally dug in for the night. Sentries were posted in anticipation of the company's final defense exercise. The sun went down quickly that night, and as supper rolled around, the sudden explosion of an anti-personnel mine sounded the start of our defense exercise. The ensuing machine-gun fire accompanied by the spurts of M-16s revealed that the main thrust of the attack came from the North-West Sector. My runner received a message from Company HQ ordering me to attend an immediate operations order. Company Commander Tan informed us, the platoon commanders, that the reconnaisance patrol was trapped between the enemy and couldn't reach us. A platoon was needed to rescue them. Amidst the constant barrage of gunfire and explosions, and the emotional arousal created by the sudden attack, Tan chose Officer Cadet (OCT) Cheng and his men. Tan liked Cheng, and although Cheng was a good soldier, he did not have the assertiveness, confidence or the military "sense" to be able to carry out the rescue mission successfully. On the other hand, OCT Wong had all these qualities- the one thing that kept Wong from being chosen was that Tan did not like him. Needless to say, the rescue operation was very disorganized and very nearly failed. Although Cheng managed to return with the patrol, he lost half his platoon, and eighty percent of the remainder were "casualties". I have since left the military, but I have always wondered whether Commander Tan would have made the same choice if he had not been under pressure. Perhaps he would have based his decision on criteria more relevant than personal liking. As a graduate student in psychology, I now find it possible to address such questions using scientific methods. In this paper, I attempt to operationalize and test several questions. The most general question was whether emotional arousal affects one's judgments about other people. Do the judgments really become cognitively simple ? Does arousal change the criteria used in one's social judgments ? Does arousal affect some people more than others ? Before attempting to answer these questions empirically, I will begin with a review of the arousal literature. 2 Arousal This section will first focus on the concept of arousal, and then on how it affects cognition, and subsequent decisions. Following this, I will outline a model that explains the effects of arousal on cognition and performance. Arousal may be defined as "a general drive state which potentiates all behavior. The idea is that all activity must be driven by some internal energy and the availability of this energy corresponds to the arousal of the organism" (Posner, 1975, pp 444). A number of theories have assumed the existence of such a nonspecific arousal. However, researchers who compared various arousal agents and their effects on performance.found that the effects of increasing arousal depend on current levels of arousal, and the types of arousal agents (Glass & Singer, 1972; Grings & Dawson, 1978; see Eysenck, 1982, for a more recent review). Consequently, a number of theorists have attempted to resolve these different and often equivocal findings by proposing multifaceted arousal models (e.g., Eysenck, 1982; Kahneman, 1973; Hamilton, Hockey, & Rejman, 1977; Lacey, 1967). One such model was Lacey's (1967) three-factor arousal model. He proposed that" there were three different types of arousal systems, each affected by different arousal agents and each indexed by a different set of arousal indices. His three systems were cortical arousal, autonomic arousal, and behavioral arousal. The first system involved brain (cortical) activity, as indexed by electroencephalogram (EEG)* measures and critical flicker fusion (CFF). The second system involved the physiological activity of autonomic functioning, as measured by galvanic skin response (GSR), heart rate (HR), and blood pressure (BP), etc. The third system involved behavioral performance, such as reaction time latency and accuracy. This last factor can be described in terms of the Yerkes-Dodson law which proposed an inverted U-shaped relation between arousal and performance (Yerkes & Dodson, 1908). These two researchers stated that there is an optimal level of arousal where performance is best. As 3 one becomes aroused, one's performance improves up to a point and then when one becomes overaroused, performance deproves (see also Eysenck, 1982; Glass & Singer, 1972). White Noise and Arousal One commonly used arousal agent is white noise (WN) (e.g., Edsell, 1976; Eysenck, 1982; Hartley, 1973, 1974; Glass & Singer, 1972; Gulian & Thomas, 1986; Rosen, 1970; Smith, Jones, & Broadbent, 1981; Thayer & Carey, 1974). In a review, Michael Eysenck (1982) reported that W N , up to 95 dBA, causes a measurable increase in physiological arousal. For example, Glass and Singer (1972) in a series of studies showed high galvanic skin responses to loud noise. Hockey (1970) and Rosen (1970) found similar results using other physiological measures. However, numerous other researchers have found inconsistent evidence regarding the general effects of noise on a wide variety of arousal indices. For instance, in his review, Kryter (1985) reported studies that failed to show a high correlation between numerous arousal indices (EEG, HR, for example) obtained during noise exposure. Some of these indices showed an initial intercorrelation, but with continued exposure to noise became weaker and in some cases, non-existent. The reason for this decrease in the interrelation was that certain indices habituated to the noise, while others failed to habituate and remained above baseline. Kryter (1985) concluded that noise had different effects on different measures of arousal, with certain indices being more sensitive to noise effects than others (see Eysenck, 1982, Hancock, 1984, for similar discussions). Roth, Dorato, and Koppell (1984) also compared the effects of loud noise on numerous indices of arousal. These included heart rate (HR), skin conductance response (SCR), and E E G . These researchers found that the cortical measures (e.g., EEG) did not habituate to the noise at all, while the autonomic measures (e.g., HR, GSR) did so almost immediately, by quickly returning to baseline levels. This finding provides further evidence of the presence of differences in the reactions of different arousal indices to the same 4 arousal agent. Swenson and Tucker (1983) also found that loud white noise led to an increase in cortical arousal. They measured brain wave activity during exposure to loud noise, and discovered increases in brain wave activity, as indexed by the E E C Voicu (1970) reported similar results when he observed that E E G increased during exposure to loud noise. In a series of studies conducted by Paulhus and his colleagues, loud white noise did not affect heart rate, although physical exercise did (Paulhus, Murphy, & Reid, 1986). On the other hand, Paulhus et al (1986) found that loud white noise had cognitive effects, namely, enhanced performance on a simple letter transformation task and impaired performance on a difficult transformation task. There was no effect for physical exercise on performance on the letter transformation task. In short, white noise seemed to have affected the cortical component of arousal and not the autonomic component. In addition, cortical arousal appears to underlie performance divergence on cognitive tasks: That is, improved performance on simple tasks, and impaired performance on difficult tasks. In conclusion, the evidence that different physiological indices do not show similar reactions to the same arousal agent lend support to Lacey's proposition that there are three different types of arousal systems (Lacey, 1967), . Furthermore, noise affects the cortical component (Roth et al, 1984; Swenson & Tucker, 1983), while not affecting the autonomic component (Kryter, 1985; Roth et al, 1984). A number of researchers have looked at the relation between white noise and critical flicker fusion (CFF). C F F is the fastest rate at which an intermittent light source still appears to the observer to be flickering (Bobon, Lecoq, von Frenkell, & Lavergne, 1982, p. 1). Numerous researchers have indicated that a high C F F value was an indicant of cortical arousal (Gortelmeyer & Wiemann, 1982; Kranda, 1982; Smith & Misiak, 1976), and was positively related to measures of self-reported alertness (Grandjean, Baschera, Martin, & Weber, 1977; Parrott, 1982). Paulhus, Reid, and L im (1987) exposed subjects to either continuous loud (95 dBA) or soft (55 dBA) white noise while measuring their C F F o level. They found that subjects exposed to the loud noise had significantly higher C F F values than those in the soft noise condition. They concluded that loud noise led to an increase in cortical arousal, as evidenced by the higher C F F . Taken together, these studies clarify the effects of white noise on the arousal system: It seems to affect the cortical component while not affecting the autonomic • component. Furthermore, this increase in cortical arousal appears to underlie the observed cognitive effects (i.e., the differential processing of easy and difficult tasks). For example, Loeb and Alluisi (1984) found that high E E G affected level of performance and this decrement in performance was directly associated with time on task. On the other hand, Anastasi (1964) noted that noise effects may result from effects of noise other than arousal. Similarly, Dylan Jones (1983) and Schroder, Driver, and Streufert (1967) have suggested that it may be these other effects of white noise, working either independently or in an interactive fashion with arousal, rather than arousal per se, that caused the predicted effects. One such effect is aversiveness. If aversiveness casuses people t o stop processing information (e.g., closing eyes, thinking about something else), then loud noise should impair all tasks. However, many studies (Eysenck, 1982; Hockey & Hamilton, 1970; Paulhus et al., 1986) have found that loud noise actually improves performance on simple tasks. Therefore, if aversiveness is the mediator, its effects must be more subtle. Another potential confound is distraction. Subjects may focus on the noise and spend less time or concentrate less effectively on the primary task (Mandler & Sarason, 1952). These researchers conceived of distraction as a gross interference that affected performance on all tasks. If loud white noise were simply distracting, then it should impair all tasks, not improve some. Therefore, this distraction hypothesis again cannot account for the abovementioned findings of Hockey and Hamilton (1970) and Paulhus et al. (1987). A more subtle form of distraction-selective attention-is discussed later. 6 Finally, the decibel levels used in most WN research have ranged from 50 dBA to 130 dBA (see Eysenck, 1982; Glass & Singer, 1972; Hamilton, Hockey, & Quinn, 1972; Hockey, 1970). In the most extensive review, Kryter (1985, 1970) concluded that arousal would not occur in noise levels below 65 dBA, and that levels above 130 dBA would not only elicit arousal, but also pain and cutaneous stimulation to the ear. White Noise and Selective Attention A number of studies have shown that when subjects were required to attend to two tasks simultaneously, W N affected the secondary task more than it affected the primary task. Indeed the primary task sometime improves (see Jones, D., 1983, for a review of these studies). Hence it became evident that W N effects on task performance were mediated by a restructuring of information processing (Cohen, 1978; Eysenck, 1983; Glass & Singer, 1972; Jones, 1983). Consistent with this notion of cognitive factors playing a role in the effects of W N on task performance, Eysenck (1982) concluded that the effects of W N may be such that it requires attention, and consequently one is forced to change strategies to deal with the task at hand. Eysenck went on to report supportive findings by numerous researchers (Berlyne, 1960; Blechman & Dannemiller, 1976; Broadbent & Gregory, 1965; Davis & Jones, 1975; Hockey, 1970; Hockey & Hamilton, 1970). One explanation used to account for the findings of differential noise effects on secondary and primary tasks is that the arousal component of W N is affecting encoding capacity (Davis & Jones, 1975; Hockey & Hamilton, 1970). These researchers postulated that the observed effects of W N on attentional capacity may be due to the fact that information encoded under arousal may be more simple in structure (i.e., more physical properties of the stimuli are encoded than semantic properties). On the other hand, information encoded in normal conditions may be more complex in structure (i.e., more contextual and semantic properties of the stimuli are encoded). In short, arousal, caused by the loud W N , may bias one towards using a lower level of information processing and 7 away from using a higher and more complex system of information processing (see also Davis, 1983; Jones, 1983, Loeb, 1981, and Paulhus, 1984, for similar conclusions). The most influential theory of arousal effects on cognitive processing is that of Easterbrook (1959), who claimed that arousal led to attentional selectivity, i.e., one became selective about the cues available by focusing on a smaller number. This model has been supported by the findings of Boggs and Simon (1968), Hockey and Hamilton (1970), O'Malley and Poplawsky (1971), and more recently, by Cohen and Lezak (1977), who showed that arousal led to an increase in attention paid to the information needed for the primary task at the expense of the secondary task. In line with the Easterbrook (1959) hypothesis, Kahneman (1973) and later, Sheldon Cohen (1978), combined both selectivity and capacity components of attention into a "capacity model of attention". This model proposed four tenets: (1) We have limited attentional capacities; (2) when environmental demand exceeds capacity, we develop a set of priorities by focusing on relevant cues at the expense of others (attentional selectivity); (3) we then commit some of our capacity "units" to evaluating the stimuli; and (4) prolonged demands in capacity lead to a temporary depletion in capacity. Simply then, when overloaded (or when aroused) by an external stimulus, the organism allocates attention to relevant cues at the expense of irrelevant or subsidiary cues, both social as well as physical, that would otherwise have been perceived and used under less demanding situations. One limiting feature of all these models is the ambiguous meaning of "selectivity". When subjects are told what the.primary task is, they can normally maintain performance on that task under white noise. However, certain stimuli are known to overide such instructions (e.g., one's name or threatening stimuli) (Paulhus & Levitt, 1987). Moreover, without explicit instructions these models do not make any predictions about what attributes or dimensions predominate under arousing circumstances. This next section deals more specifically with this issue. 8 Cognitive Complexity I will use the term "cognitive complexity" in the most general sense to encompass all the different definitions of complexity (for example, "conceptual complexity", "differentiation", "dimensionality", "intregration", etc.) This usage is consistent with Streufert and Streufert (1978) who also used this term to encompass all the different concepts of cognitive complexity (see p. 12 and 15). The concept of cognitive complexity originated with George Kelly (1955) and James Bieri (1955) who considered the number of dimensions (or constructs) available to the organism. In social judgments, for example, cognitive complexity refers to the number of dimensions used to judge the traits and behavior of others. An extension of the concept arose from the work conducted by a school of researchers at Purdue, Princeton and Syracuse led by Harold Schroder. Schroder and his colleagues coined the term "conceptual complexity" as the degree to which one's cognition is highly structured (Harvey, Hunt, & Schroder, 1961; Schroder, Driver, & Streufert, 1967; Schroder & Suedfeld, 1971). They distinguished two aspects-differentiation and integration. Differentiation referred to the number of attributes or dimensions used by the individual when judging stimuli. Integration referred to the way in which these attributes were combined together. These authors further stated that integration required differentiation. In other words, if I used only one attribute to judge an entity, I could not be measured on integration because integration involved using more than one attribute. To measure individual differences, Schroder et al. (1967) developed a measure of conceptual complexity termed the "Paragraph Completion Test (PCT)". This test comprised six opening stems (e.g., When I am in doubt, Criticism, etc.) representing the three social dimensions of uncertainty, criticism, and attitude towards authority. Responses were scored in terms of the number of distinct points of view (differentiation), and how they were integrated. The higher the score, the more complex the individual. An individual who used only one dimension would be given a score of 1. An individual who used more than 9 one dimension would obtain a score ranging from 2 to 4. Scores above 4 depended on how well he or she integrated the dimensions (see Schroder et al., 1967, for a more detailed discussion of the scoring scheme and system). A series of reports have shown the reliability and validity of this measure (e.g., Epting & Wilkins, 1974; Harris, 1981; Kammer, 1984; Slugoski, Marcia, & Koopman, 1984; Vannoy, 1965). Others have shown its predictive ability and its relations with other theoretical personality constructs (see Schroder & Suedfeld, 1971, for a review of some of these studies). Stress and Complexity Guetzkow (1959) showed, through a simulation study, that threat of war reduced one's complexity (or dimensionality) from many to two main dimensions-- evaluation and dominance. Likewise, Suedfeld and Tetlock and their colleagues (Porter & Suedfeld, 1981; Suedfeld, 1981; Suedfeld & Rank, 1976; Suedfeld & Tetlock, 1977; Suedfeld, Tetlock, & Ramirez, 1977; Tetlock, 1979), using archival material to examine the effects of stress on integrative complexity level, found that the sophisticated judgments of political leaders reduced in complexity just before the onset of conflict. In a laboratory study, Driver (1962) found the same trend in complexity reduction during simulated war games conducted in the laboratory. Finally, Wallsten and Barton (1982) showed that under speeded conditions, dimensionality was reduced from three dimensions to only one. It is not unreasonable to suggest that a state of emotional arousal mediated these effects of stress, threat, and urgency. Although couched in different terminology, these studies are consistent with Easterbrook's (1959) proposition that arousal leads to increased selectivity of cues in the decision making process. Only the last group of researchers (e.g., Guetzkow, 1959; Suedfeld & Tetlock, 1977; Wallsten & Barton, 1982), speak to the kinds of attributes or dimensions that are utilized under arousal. I will now introduce a new model of information processing that directly addresses this question. 10 Dynamic Complexity Theory This theory has been described to varying degrees in a number of paper presentations and articles (Paulhus, 1984; Paulhus & L im, 1985; Paulhus, Lim, Reid & Murphy, 1986; Paulhus & Suedfeld, in press). The theory draws from the research on conceptual complexity (Schroder et al., 1967; Schroder & Suedfeld, 1971). However, dynamic complexity theory focuses only on differentiation-the number of dimensions or attributes one uses to judge an entity. This model has three main propositions. First, arousal tends to reduce the complexity of information processing. Second, this reduction in one's complexity level leads to an amplification of the role of the evaluation dimension in decision making. Third, psychological threat or stress triggers a fast-rising arousal that leads to a reduction in complexity before any other information is processed. Fourth, this reduced complexity is behaviorally and cognitively adaptive. Finally, trait complexity moderates the effects of arousal on state complexity and consequently, the effectiveness of defense (see Paulhus & Suedfeld, in press, for a more in-depth discussion of these propositions). One final point to note is that the relation between arousal and cognitive complexity is probably an inverted-U similar to the one proposed by Yerkes & Dodson (1908), Easterbrook (1959), and Schroder et al., (1967). This claim is supported by Paulhus, Murphy, and Reid (1986) who showed that after subjects worked on the task for more than 5 minutes, the arousal effects disappeared. Presumably subjects became bored and/or fatigued. In any case, the dynamic complexity model expressly focuses on reductions in complexity beyond the optimal point of arousal—i.e., on the descending slope of the inverted-U function (Paulhus & Suedfeld, in press). This section of my paper will now describe the evidence for the first and second propositions, and then go on to propose a line of research that attempts to provide more direct evidence to substantiate these first two propositions. The other three propositions are currently being investigated in the laboratory by Paulhus and his colleagues. 11 In most domains involving human judgment, a number of attributes are utilized concurrently, and the judgment can be considered to be multidimensional in nature (Jones, L . , 1983). In line with this, Charles Osgood, using the semantic differential technique, reported studies that supported this contention (Osgood, Suci, & Tannebaum, 1957). He concluded from his factor analytic studies that the three dimensions predominantly used to judge any stimulus (or set of stimuli) were Evaluation, Potency, and Activity (Osgood et al, 1957). The primary dimension used is evaluation--i.e., entities are judged in terms of likability, or good-bad. The second dimension, potency, focuses on the power aspect of the entity. That is, whether the stimulus is strong, dominant, or powerful. In the social domain, this dimension has been alternatively termed dominance. The third dimension, activity, concerns the degree of movement or excitement perceived in the stimulus. Osgood (1976; Osgood et al., 1957) has reported the consistency and robustness of these three dimensions across a variety of domains-countries, languages, paintings, sonar signals. The last two dimensions have been found to be fairly highly related to each other in social domains. Evaluation, on the other hand, has usually been found to be orthogonal to the other two. Paulhus and Lim (1985). Using Osgood's "Big Three", Paulhus and L im (1985) tested the first proposition of the dynamic complexity model. Arousal was manipulated by varying the intensity of white noise (50 vs. 95 dBA). In all these studies, male subjects rated the pairwise dissimilarity of six targets. State-complexity level was operationalized as the number of dimensions used. A multidimensional scaling (MDS) procedure was used to configure the subjects' similarity ratings of a particular set of stimuli (Carroll & Wish, 1974; Kruskal & Wish, 1978; Schiffman, Reynolds & Young, 1982). This MDS procedure was used to reveal the group o space that subjects used . In addition, using an individual differences MDS procedure-I N D S C A L (Carroll & Chang, 1970),-individual subject weights were derived for each 12 dimension appearing in the group space. These subject weights reflected the amount of weighting each individual placed on each dimension. Two indexes of evaluative importance were developed to measure the relative importance of the primary dimension in relation to the other two dimensions. The first index, termed the "scaled evaluative importance ratio (scaled El-ratio)", was obtained for each subject by dividing the M D S weight for the first dimension by the sum of the M D S weights of the other dimensions. The second index, termed the "unsealed Evaluative Importance Ratio (unsealed El-ratio)", is defined as the dissimilarity rating of the two targets representing the evaluation dimension divided by the sum of the corresponding dissimilarity ratings on the second and third dimensions. Using these two indices, we conducted three studies, each addressing a different stimulus domain (acquaintances, university courses, and self-roles). In the first two studies, subjects were asked to identify one exemplar for each of the six categories representing Osgood's evaluation and dominance dimensions, and a third dimension, industriousness. Subsequently, subjects did similarity ratings of these six exemplars. In Study 1, subjects supplied names of six of their acquaintances representing the six categories-likable, unlikable; forceful, not forceful; industrious, not industrious. Subjects were then presented with all possible pairs of these six aquaintances and instructed to do similarity ratings of each pair. During the rating task, half the subjects were exposed to continuous loud W N (95 dBA), while the other half were exposed to continuous soft W N (55 dbA). Subjects in the loud condition, compared to subjects in the soft condition, placed relatively more emphasis on evaluation than the other two supplied dimensions-i.e., the former group had a higher El-ratio. However, the two MDS group spaces did not appear to differ in state-complexity. In Study 2, we tested the robustness of the dynamic complexity model by looking at a different stimulus domain-university courses. However, the last two dimensions used in Study 1 seemed inappropriate in judging this domain. Therefore, we kept the evaluation 13 dimension, and added on course difficulty and class size. The procedure used was basically the same as Study 1, except subjects were asked to name six university courses, one for each of the six categories-likable,, dislikable; difficult, not difficult; large, not large. The group space did not clearly reveal the three intended dimensions of evaluation, course difficulty, and class size. Moreover, all the three dissimilarity ratings (i.e., the ratings for the three "critical" pairs used in the El-ratio) did not change in loud white noise compared to soft noise. In order to determine what dimension(s) subjects were actually using under arousal, we looked back at the courses each subject had originally named. It appeared that subjects were reporting large dissimilarities between arts and science courses. Consequently, an Arts vs. Science importance ratio was calculated: the ratio of the dissimilarity rating for arts vs science courses, divided by the sum of the dissimilarity ratings between the arts vs arts courses and the science vs science courses. This ratio was significantly larger for subjects in the loud condition, and was also larger than the El-ratio. We concluded that the primary dimension used in judging university courses was not evaluation, but rather the arts-science distinction. Although the primary dimension differed from that in Study 1, the same cognitive process was observed. In Study 3, the procedure was again similar to that in the first two studies. Rather than coming up with six names, subjects were given six self-roles based on Linville's (1982) social roles: son, student, friend to men, friend to women, leader, healthy person. The three most dissimilar pairs obtained under arousal were, in order of dissimilarity, student-friend to women, son-friend to women, and lastly, leader-friend to women. By using the El-ratio index, we compared the relative importance of the most dissimilar pair with that of the second and third most dissimilar pairs. The results showed that the EI-ratio was larger in the loud condition than the soft condition. To verify the dimension(s) subjects used to judge these role pairs, subjects' ratings of the six roles on the evaluation dimension were analyzed. We found that "student" was 14 least liked, and "friend to women" was most liked. These same two roles were also found to be the pair that represented the first dimension in the group space. Therefore, we concluded that the primary dimension used to judge self-roles was evaluation and that it performed according to the dynamic complexity model. In summary, arousal seemed to inflate the relative importance of the primary dimension-in the case of the social domain, evaluation, and for university courses, the arts-science distinction. However, the predicted reduction in dimensionality (state-complexity) was not clearly supported. There are a number of possible reasons. First, in Studies 1 and 2, the freedom by which subjects were asked to come up with their own stimuli in terms of the given dimensions failed to preclude use of extraneous dimensions that they brought in to the studies. That is, subjects may have reduced complexity on their own idiosyncratic dimensions. Moreover, having only one dissimilarity rating for each dimension was not psychometrically reliable enough. These factors may explain the inconclusiveness of this set of studies in supporting the contention that under arousal, the primary dimension became more important as a result of the other dimensions dropping off completely. Individual Difference Measures Numerous researchers have indicated the potential moderating effects of individual difference variables on arousal effects on task performance (e.g., Bell & Byrne, 1978; Eysenck, 1982; Schroder & Suedfeld, 1971). This section will examine two such individual difference variables. Trait-Complexity. Using the PCT to measure trait integrative complexity, Schroder et al. (1967) found consistent differences between trait-complex and trait-simple subjects across a variety of tasks. Likewise, Driver (.1962) showed that complex subjects judged others along more dimensions than did the simple. This finding is consistent with Brock's (1962) formulation that simple structures rely mainly on a single salient component of information (or dimension) that eventually becomes central in governing general attitudes. 15 Furthermore, Driver (1962) found that simple subjects performed worse than complex subjects under stressful conditions. He attributed this difference in performance to the fact that the simple were not able to use the available amount of information equally, whereas the complex were able to. However, he also concluded that the trait-complex, when placed under stress, reduced their complexity more than did trait-simples, although the former never fell below the complexity level of the latter (Driver, 1962, pp 112). The study most similar to the present one found mixed support for the Driver (1962) conclusion. Paulhus and L i m (Study 3, 1985) used the PCT as the trait measure, and number of constructs generated to classify a set of stimuli as the state measure of complexity (dimensionality). Comparing the top one-third with the bottom one-third subjects on the PCT scores, we found that under arousal, trait-complex individuals reduced in complexity marginally more than did trait-simples. However, there was no significant difference between these groups on the importance placed on evaluation under arousal (Paulhus & Lim, Study 3, 1985). The mixed findings can perhaps be best explained by the fact that in the Paulhus and L im (1985) studies, there was no control for the number of dimensions used by subjects. There was some indication that some subjects brought in new dimensions under arousal. Secondly, the instrument used to distinguish trait complexity measured the integration component and not the differentiation component. This may have perhaps reduced the sensitivity of the trait measure to arousal effects. Therefore, the present Study 2 again tested the Driver (1962) results, but this time both trait complexity groups (based on differentiation) were exposed to the same amount of information (i.e., three prescribed dimensions). Repression-Sensitization (R-S). This construct has its roots in the Freudian tradition, as an ego defense mechanism (Bell & Byrne, 1978; Krohne & Rogner, 1982). Simply, repressors are individuals who tend to deny or avoid any form of threat, whereas sensitizers are individuals who tend to approach the threat by ruminating or verbalizing 16 about it (Gordon, 1957). This R-S construct is related to the concept of perceptual defense (Bruner & Postman, 1947) in that repressors are known to adopt perceptual defense against threatening stimuli, while sensitizers are known to adopt perceptual vigilance. Byrne (1961), using items from the M M P I (Hathaway & McKinley, 1951), developed the Repression-Sensitization (R-S) Scale. This scale comprised 156 items which were later reduced to 127 items. This scale comprised statements that were to be answered either "True" or "False". Responses were then scored based on a key such that a high score indicated a sensitizer. Recently, Paulhus and Levitt (1983) reduced the number of items in this scale to 43, and termed it the mini R-S. They found almost as high an internal consistency coefficient (.85) as that of the 156-item form (.89). This R-S construct has been widely tested and compared with other individual difference measures, and used to assess performance on various tasks. The interested reader is asked to refer to the chapter by Bell and Byrne (1978) for a complete review of this construct. The research area of primary concern to this author is the relation between the R-5 construct and arousal. Sensitizers have been found to be more prone to be affected by irrelevant aspects of a situation or task than repressors, and consequently are unable to focus on the task at hand (see Krohne, 1978; Krohne & Schroder, 1972; Mischel, Ebbesen, 6 Zeiss, 1976). These "cognitive" differences between the two defense styles may be a result of differences in information selectivity during the stages of encoding and retrieval. For example, during the retrieval stage, because repressors "block out" task-irrelevant stimuli (eg., any form of threat or arousal), they are more likely able to recall all the task-relevant information and to perform the task well. On the other hand, sensitizers will evidence poorer memory recall as well as poorer task performance, a result of them being unable to "block out" task-irrelevant aspects of the situation (Haley, 1974). Support for this contention was found in a study done by Krohne and Schroder (1972). They reported 17 that sensitizers were more easily affected by stress, and consequently their level of information processing decreased to a greater extent than did repressors. The Present Studies Two studies were designed to answer more clearly the question of what happens to state complexity level and evaluative importance under arousal. Study 1 was designed to develop behavioral descriptions of eight fictitious people to represent all possible combinations of three dimensions- evaluation, dominance, and industriousness. For example, a target might be unlikable (evaluation), dominant (dominance), and industrious (industriousness). These descriptions controlled the amount of information given to subjects, thereby discouraging subjects from bringing in other extraneous dimensions when judging the eight targets. Study 2 was designed to test the hypothesis that evaluation would become primary as a result of a reduction in complexity under arousal. Subjects rated the dissimilarity of these eight fictitious people while being exposed to either loud or soft WN. Furthermore, this study tested the hypothesis that both the trait complex subjects and the sensitizers would be most affected by the loud white noise. Study One The primary aim was to develop descriptions of eight fictitious people configured on three relatively orthogonal dimensions: evaluation, dominance (potency), and industriousness. These eight descriptions were refined until (a) they were quite easily recalled during an experimental session and (b) they were configured correctly, that is, the targets were correctly placed in their hypothesized quadrants. In short, I hoped to develop an eight-person social domain that was configured in three dimensions. This was an attempt to ensure that subjects used only the three primed dimensions when making judgments about these social targets, and that the evaluation dimension would be more important than the other two dimensions under moderately arousing conditions. Method The first step was to develop eight one-sentence behavioral descriptions to include all possible combinations of high and low on the three dimensions. The development of these one-sentence descriptions was guided by the lists of prototypical acts provided by Buss and Craik (1983). These researchers asked judges to rate the most protot3'pical acts for eight traits around the interpersonal circle (Leary, 1957; Sullivan, 1953; Wiggins, 1979). In my case, I chose behavioral descriptions rather than trait-words, as I wanted subjects to encode the concepts of evaluation, potency, and industriousness rather than simply memorizing the trait adjectives themselves. Note also that each dimension has four targets scoring high and four targets scoring low. I chose four different behaviors to ensure that the four targets on the same pole of one dimension would be behaviorally different but conceptually similar. For example, the dominant behavior of "Alan" is he demands that others let him copy their assignments. Likewise, the dominant behavior of "Frank" is that he usually gives advice even when none is requested. Both these behaviors were chosen to be high on dominance 19 and low on evaluation (i.e., dislikable). Again this was to force subjects to make their judgments at a conceptual level (Winter, Uleman, & Cunniff, 1985). After writing the descriptions, the next step was to come up with a helpful mnemonic for each target to assist the subject in remembering the targets and their behaviors. Therefore, each target was given a name. Precaution was taken to ensure that the first letter of each name would not be the same as that of the critical behavioral descriptors. I worried that the easiest strategy for remembering a person's behavior was to find a word in the description that started with the same letter as the person's name. Indeed, I discovered in a pilot study that many subjects did use this mnemonic method to recall each target's behavior. To direct subjects' use of mnemonics, I included a relatively neutral context word beginning with the same letter as the target's name. To test the impact of these eight descriptions, I administered them to several student samples under classroom (moderately arousing) conditions. This was an attempt to see if subjects were able to memorize and recall these eight descriptions, and whether the targets appeared in the appropriate octant (see Appendix A for the eight targets and their behavioral descriptions). The procedure was actually repeated several times to different samples: The final administration is described below. Subjects Sixty-six students (28 F, 37 M) enrolled in an undergraduate class participated as subjects to obtain extra course credit. Procedure This study was administered during class time and it took ten to twelve minutes to complete the whole task. Subjects were given four minutes to read and memorize the eight target descriptions . After four minutes, subjects were told to stop reading the descriptions, and to hand in the description sheets to the experimenter. Subsequently, a sheet of 7-point rating scales was distributed and subjects were instructed to rate the eight 20 targets on three characteristics-- likable, assertive, and industrious (see Appendix B for the rating scale). Upon completion of this rating task, subjects were debriefed by the experimenter. Results The mean rating for each target on each of the three characteristics was computed (see Table 1). As can be seen from Table 1, the mean ratings of these eight targets on the three dimensions were in the hypothesized direction. For example, "Alan" had a mean rating of 1.97 on likability, 5.46 on assertiveness, and 2.75 on industriousness, values that were in line with his predetermined character portfolio of being unlikable, dominant, and lazy (refer to Appendix A for descriptions). Likewise, all the other targets had ratings that matched their predetermined character portfolios. These mean ratings were then plotted in three-dimensional space to show the overall configuration (see Figures 1 and 2). The correlations among the three dimensions were as follows: The correlation between evaluation and dominance was -.28, the correlation between evaluation and industriousness was .33, and the correlation between dominance and industriousness was zero. Although not completely orthogonal, these dimensions were apparently distinct enough to represent three separate dimensions. 21 Table 1 . Mean ratings of the 8 t a r g e t s on the three dimensions (N*66). Dimensions St i m u l i Evaluat ion Dominance Industriousness ALAN (DDL) 1.97 5.46 2.75 BOB (LDL) 6.28 4.49 3.54 CHRIS (LDI ) 5.95 4.78 5.98 DAVE (LSI ) 4.95 1 .98 4.56 ED (LSL) 5.44 2.33 2.42 FRANK (UDI ) 3.02 5.47 4.53 GARY (USI ) 3.82 2.77 5.50 HARRY (USD 2.51 3.31 2.43 Variance 2.68 2.00 1 .93 D i f f e r e n c e Score 2.83 2.48 2.36 Character-type of Target: UTJL— U n l i k a b l e , Dominant, Lazy; LDL-- L i k a b l e , Dominant, Lazy; LDI- L i k a b l e , Dominant, In d u s t r i o u s ; LSI-- L i k a b l e , Submissive, Industrious; LSL-- L i k a b l e , Submissive, Lazy; UDI-- U n l i k a b l e , Dominant, Industrious; USI-- U n l i k a b l e , Submissive, Industrious; USL-- U n l i k a b l e , Submissive, Lazy. D i f f e r e n c e Score: Mean of 4 targets anchoring the p o s i t i v e p ie of one dimension minus mean of 4 t a r g e t s anchoring the negative pole of the same dimension. For example, Eval D i f f Score= M(Bob,Chris,Dave,Ed) - M(Alan,Frank,Gary,Harry) F i g u r e 1 M e a n R a t i n g s O f T h e E i g h t T a r g e t s O n T h e E v a l u a t i o n A n d P o t e n c y D i m e n s i o n s . POTENCY >UDL >UDI >LDI •LDL >USL >USI >LSL •LSI i r 1 2 T 4 T T 7 to to EVALUATION F i g u r e 2 M e a n R a t i n g s O f T h e E i g h t T a r g e t s O n T h e E v a l u a t i o n A n d I n d u s t r i o u s D i m e n s i o n s . INDUSTRIOUS • L D I >USI •UDI • L S I >LDL • UDL • U S L • L S L To evaluate the importance placed on each dimension, the variances of the eight ratings on each dimension were computed. They were as follows: evaluation, 2.68; dominance, 2.00; and industriousness, 1.93 (see Table 1). These variances were then subjected to a one-way analysis of variance to test if one dimension showed a wider range of ratings on the eight targets than the others. There were no significant differences between them, although the variances were in the predicted order. As another index of dimensional importance, the mean for the sum of the four targets representing each pole of each dimension was computed. For example, the ratings of the four likable targets were summed and averaged, and this was done for all the other five sets of four targets anchoring each pole of the three dimensions. Subsequently, a difference score was obtained by taking the mean rating of the four targets anchoring the positive pole of each dimension minus the mean rating of the four targets anchoring the negative pole of each dimension. For example, the mean rating for the 4 "likable" targets minus the mean value for the 4 "unlikable" targets (see bottom of Table 1). The three difference scores were for evaluation, 2.825; for dominance, 2.48; and for industriousness, 2.355. These difference scores were subjected to a one-way repeated measures A N O V A to test if one dimension was given more importance than the others. There were significant differences between these three difference scores, with the evaluation difference score being significantly larger than that of dominance, F(l,65) = 4.55, p < .04, and that of industriousness, F(l ,65)= 5.58, p < .02. The latter two did not differ significantly. As with the case of the variances, these difference scores showed the same trend with evaluation having the largest difference score followed by dominance, and industriousness. A secondary analysis looking at gender differences showed no significant differences between the two genders in their ratings and subsequent placement of the eight targets. Discussion The results showed that subjects were indeed able to recall the eight targets and to place these targets accurately on the three-dimensional space. In short, I have successfully constructed a social domain where subjects used the evaluation dimension more than dominance and industrious under moderately arousing conditions. This configuration is consistent with other findings of evaluation being the primary dimension social domains (e.g., Osgood et al, 1957; Fiske, 1980; Rosenberg & Sedlak, 1972). Finally, I discovered large individual differences among the subjects in terms of memorization ability. Some of them reported that they found it hard to memorize the descriptions while others found it easy. Because subjects will be required to memorize these descriptions in Study 2, I decided to include a measure of this individual difference variable in the event that this variable may interfere with the manipulated variable used in the study. Study Two On the basis of the literature reviewed earlier, cortical arousal (henceforth termed "arousal", unless otherwise stated) was manipulated by comparing loud to soft white noise conditions. The first hypothesis was that subjects in the loud compared to subjects in the soft noise condition would place relatively more importance on the evaluation dimension than either of the other two dimensions. The second hypothesis was that this increase in the relative importance of evaluation under arousal would be a result of the other dimensions diminishing in importance. In short, a reduction in state-complexity level under arousal. A further prediction was that sensitizers and trait-complex (high differentiation) individuals would most clearly show the above effects. More specifically, they would show the largest reductions in complexity, and the largest increase in the emphasis of the evaluation dimension under arousal. Method Subjects Forty male subjects enrolled in an introductory psychology class participated in this study for course credit. Females were not used because in previous studies, several of them had complained about the noise and subsequently discontinued participating in the experiment. Thus, the difficulty of analyzing a self-selected sample (i.e., those who could tolerate the noise and remained in the study) was avoided. Stimulus Materials The stimuli used in this study were the eight fictitious people whose descriptions were developed in Study 1. To recapitulate, each target was either high or low on each of the three dimensions-evaluation, dominance, and industriousness. For example, target "Alan" is unlikable, dominant, and lazy. For a given dimension, each target differed from only one other target on that dimension and that dimension alone. For example, "Alan" differed from "Bob" only on the evaluation dimension as both of them were characterized as dominant and lazy. There were a total of 12 such comparisons (four for each dimension), and they were termed the "critical" comparisons (see Appendix C). By the term "critical" I mean that these 12 comparisons (or pairs) anchored the respective three dimensions, as they were the only comparisons that differed on only one dimension while "controlling" for the other two dimensions. Independent Measures Trait Complexity. Subjects were scored according to their responses to the six PCT stems (see Appendix H for the six PCT stems). A score was derived for each subject-the mean value of all six stems. A cutoff score of 2.5 was based on the differentiation (DPCT) rather than the integration component of cognitive complexity. That is, scores 2.5 and below reflected low differentiation (low number of dimensions used), while scores above 2.5 reflected high differentiation. 28 Repression-Sensitization. Subjects completed the mini R-S scale (Paulhus & Levitt, 1983) (see Appendix H for a copy of the R-S scale). A high score indicates a sensitizer. This study adopted the median split to distinguish two groups of individuals- repressors (below the median value), and sensitizers (above the median). Arousal Manipulation The arousal agent was continuous white noise (WN). High arousal was induced by exposing subjects to loud W N (95 dBA), and moderate arousal by soft W N (55 dBA). Low or no arousal has been found to be difficult to induce in the laboratory (Duffy, 1962; Linden, 1987; Paulhus, Murphy, & Reid, 1986). Furthermreo, I chose a soft noise condition over a no noise condition for the simple fact that if there were significant differences between no noise and loud noise, it would be considerably more difficult to explain these differences (the subjects would be non-comparable in several respects). On the other hand, by exposing subjects to different intensities of white noise, any significant differences would have to be attributable to the differences in noise intensity. In the study conducted by Paulhus et al. (1986), the most profound effects of W N on task performance occurred within the first three minutes of the experiment. After that, the noise effects on performance became less predictable and consistent. That study concerned the effects of W N on letter transformations-a nonsocial domain. Nevertheless, it suggests that any experiment using W N should be completed within the first three minutes or so for maximal W N effects. A post-experimental questionnaire was administered to the subjects as an experimental check to ensure that the W N manipulation worked. Using 7-point Likert scales- "1" (not at all) to "7" (very much so)--subjects rated the W N on the following attributes: distraction, physical arousal, irritability, and loudness. The W N was presented to the subjects through the use of a Sony cassette recorder (model TC 110-B), and a pair of headphones. The volume levels on the cassette recorder were preset to correspond to the respective dBA levels-55 dBA for the soft condition, and 29 95 dBA for the loud condition. The dBA levels were assessed using a Radio Shack sound level meter (model 33-2050). Subjects wore the headphones and were exposed to the W N throughout the dissimilarity rating task. Dissimilarity Rating Task The dissimilarity rating task was performed on an Apple 11+ personal computer. The micro-software used in this dissimilarity rating task was developed by Dong (1981). The program, "SIMRRT", allows input up to a maximum of 50 stimuli to be rated pair-wise along a dissimilarity dimension. This program then records each subject's responses (or judgments) and reaction times (RTs) to all comparisons. This SIMRRT program was modified by Paulhus and Lim (1985) to include various instructions and a ten-second time delay between pair presentation and subject response"*. In the present study, the program presented all possible pairs of the eight targets in 2 blocks of 14 pairs each (see Appendix C for the 2 Blocks). Each block comprised 6 "critical" pairs or comparisons (2 for each dimension) and eight others. Each target was equally represented in each block. Each block was randomly presented to the subjects, and within each block, the pairs were randomly presented. The reason for dividing the 28 pairs into two blocks was the finding by Paulhus et al. (1986) that most effects of W N on performance occurred within the first three minutes of exposure. Because the entire task would take about six minutes to complete (including the 10-second delay for each pair), only half of the 28 pairs would be judged within the first three minutes and consequently be under the maximal influence of the W N . Therefore, by dividing the 28 pairs into two blocks of 14 and counterbalancing the block order across subjects, all 28 pairs were judged within the first three minutes. Subjects provided dissimilarity ratings of the eight targets by pressing a number ranging from "0" (no difference) to "8" (very different) to represent their judgments. Subjects were instructed to press a number only when the command "Respond Now" 30 appeared on the screen after the time delay. Following a subject response, a new pair was presented. Procedure Subjects were tested individually in the laboratory. Upon entering the lab, subject met the experimenter who then ushered subject to a seat by a table. Subject then read the following description: "This research study is designed to examine how people rate others along a similarity dimension-- i.e., which people are similar to one another and which are not. In order to examine this question, you will be asked to memorize eight fictitious people. Subsequently, you will be presented with a pair of these eight fictitious people on a T V screen, and asked to rate how similar each pair-member is to the other. You will be exposed to some non-painful "noise" while you are making these judgments." After reading the instructions, the subject signed the consent form indicating that he had agreed to participate in the study. Subject then completed two questionnaires. The first was the short Repression-Sensitization scale (mini R-S) and the second, the Paragraph Completion Test (PCT). Upon completion of the first questionnaire, experimenter explained the instructions of the PCT to subject. Experimenter then informed subject that he had a total of 12 minutes (2 minutes per stem) to complete all six PCT stems, and that experimenter would signal to subject at the half-way point that subject had six minutes left. After completion of the PCT, experimenter informed subject of the next two phases of the experiment. The first phase involved the memory task; the second, the computer dissimilarity rating task. Experimenter then explained the second phase by turning the subject's attention to the blackboard in the lab. Experimenter read the following instructions to the subject: 31 "The next two phases of this experiment involve first, a memory task, and then a computer task. The memory task will require that you memorize the written descriptions of eight fictitious people, which will be presented to you in a moment. You will be given four minutes, which is the average amount of time required to memorize the descriptions of the eight targets. After you have completed the memory task, you will be seated in front of the computer, and presented with all possible pairs of the eight targets to be presented in this format: (Experimenter pointed to the board) HOW D I F F E R E N T A R E ... * * * * * * * * * * * * * * * * * * * * * * A L A N : BOB * * * :t * * * * :|: :f * * * * * * * * * * * * Experimenter explained to subject that his task was to rate how different each member of the pair was to the other by using a 9-point scale-- "0", having the descriptor "no difference" , to "8", having the descriptor "very different". Subject was instructed to press one of the nine numbers on the keyboard only after the command "Respond Now" appeared at the bottom left-hand corner, of the T V screen. Subject was told that this command would appear ten seconds after a pair appeared on the screen. Experimenter then further stated that a new pair would appear after a number is pressed, and the same procedure continued till all the pairs were presented and rated. Experimenter, instructed subject to respond fairly quickly upon the appearance of the command on the screen. Experimenter asked subject if he had any questions regarding the dissimilarity rating task. Once subject was clear about the computer dissimilarity rating task, experimenter presented the description sheet of the eight social targets to subject who was then instructed to read the sheet instructions: "Below are descriptions of eight fictitious people. Please read the descriptions carefully, and remember as much about each person as possible. The names as well as the contexts attached are meant to assist you in remembering these people and their behaviors". Subject was then given 4 minutes to memorize the descriptions of all eight targets. Then he was instructed to rate the eight targets on three characteristics- likeability, assertiveness, and industriousness, using a 7-point rating scale- 1 meaning "not at all" to 7 meaning "very much so" (this is the same rating sheet used in Study 1). Upon completion of this rating task, he then sat in front of computer, and told that in a case where he completely forgot a particular target, he was to use a dissimilarity rating of "4" every time that target was compared with another. This attempted to reduce the amount of "noise" in the ratings attributable to guesses. Experimenter tossed a coin to determine which W N condition subject would be in. The following instructions were given to subject: "While you are rating each pair, you will be exposed to continuous white noise. It will not hurt your ears or cause any permanent damage." Subsequently, subject wore the headphones and began the computer task. Upon completion of the computer task, subject completed a post-experiment questionnaire by rating the difficulty of the the dissimilarity (computer) task, how distracting the W N was, how physically arousing it was, how irritating it was, and how loud it was (see Appendix D for the questionnaire), on a 7-point scale. These questions were arousal manipulation checks. Subject also rated the difficulty of the memorization task, in order to obtain a measure of individual differences in memorization ability. Subject was then debriefed by experimenter, and allowed to leave. - 33 Results Dependent Measures Altogether there were four dependent measures. The first two were obtained directly from the raw dissimilarity ratings. Accordingly, they were termed "unsealed" measures. The other two were obtained by subjecting the dissimilarity ratings to multidimensional scaling, and deriving dimension weights. Accordingly, this latter set was termed "scaled" measures. Unsealed Measures a. Dimension Strength (D-strength). The strength of each dimension was indexed by the sum of the four critical dissimilarity ratings. A critical dissimilarity was a rating of two targets that differed only on the relevant dimension (e.g., in the case of evaluation: the likable-dominant-industriousness target versus the unlikable-dominant-industriousness target). b. Evaluative Importance Ratio (El-ratio). This ratio indexed the importance of evaluation relative to the other two dimensions. The El-ratio was calculated by dividing the D-strength of evaluation by the sum of the D-strengths of the dominance and industriousness dimensions 0. Scaled Measures The dissimilarity ratings were subjected to two individual differences multidimensional scaling programs. The first program was developed by Carroll and Chang (1970) and titled " INDSCAL" . The other was that developed by Ramsey (1982; 1977) and titled " M U L T I S C A L E " . Both programs produce (a) a group "space" that best configures the eight social stimuli, and (b) a set of weights indicating each subject's emphasis on the dimensions appearing in the group space . a. Dimension Strength. The strength of each subject's emphasis on a dimension was indexed by the individual difference weight derived by M U L T I S C A L E . 34 b. El-ratio. Again this was a measure of the relative importance of evaluation. The weight of the first dimension was divided by the sum of the weights of the other two dimensions. Intercorrelations among the dependent measures From Table 2, it can be seen that four dependent measures (the two El-ratios, and the two D-strengths for evaluation) were all highly intercorrelated (correlations ranging from .68 to .96). This finding supports the contention that the four were convergent measures of evaluative importance (see Appendixes E and F for the same correlation matrix within each noise condition). Arousal Manipulation Checks. On the post-experimental questionnaire, subjects rated their experience of the white noise on four 7-point Likert scales (ranging from 1= "not at all", to 7= "very much so"). Compared to soft noise subjects, subjects in the loud white noise condition reported noise as being more physically arousing, t(38)= 2.92, p <.006 (loud mean= 3.70, soft mean= 2.25); distracting, t(38)= 3.54, p <.001 (loud mean= 4.45, soft mean= 2.40); loud t(38)= 9.26, p <.001 (loud= 4.55, soft= 1.75); and irritating t(38)= 3.99, p <.001 (loud= 3.90, soft = 1.90). Clearly, the arousal manipulation was successful in that the loud noise condition was experienced as significantly more arousing than the soft noise condition. However, other attributes were simultaneously being varied. As can be seen from Table 2, the manipulation checks were fairly highly correlated with each other (correlations ranging between .47 to .78). These intercorrelations were similar within each noise condition, although generally smaller (see Appendixes E and F). T a b l e 2. C o r r e l a t i o n among the independent and dependent measures ( N * 4 0 ) . 18) (19) (1 ) (2 ) (3) (4) (5 ) (6) (7) (8) (9) (10) (11) (12) (13) (14) ( 15) ( 16) (17 ) WN (1 ) DPCT (2 ) - 0 .04 PCT (3 ) 0 . 15 0 .66 PCT2 (4 ) 0 .04 0 .86 0 . 74 PCT6 (5 ) 0 .04 0 . 85 0. .83 0 .87 RS (6 ) 0 .03 0 . 23 0 .05 0 . 25 0 . 15 MEMY (7) o . 20 - 0 . 30 - 0 . 2 1 - 0 .21 - 0 .21 - 0 .20 COMP (8 ) - 0 . 14 - 0 . 25 -o . 28 - 0 .28 - 0 . 22 0 . 14 0 . 48 DIST (9 ) 0. 50 - 0 . 12 - 0 . 10 - 0 . 10 - 0 . 12 0 . 13 0 23 0 .21 LOUD(10) 0 83 - 0 .04 0 . 10 0 .01 0 .04 0 . 1 1 0 . 27 - 0 .04 0 . 56 IRRG(11) 0. .54 -o . 15 - 0 .01 - 0 . 10 - 0 .07 0 .20 0 . 19 0 .22 0 .78 0 .65 ARS ( 12) 0. 43 0 . 18 0 . 19 0 .08 0 .09 0 .29 -o .09 0 .02 0 .48 0 52 0. ,50 OEV ( 13) 0. 39 0 .05 0 . 16 0 .08 0 .03 - 0 .07 - 0 .03 -o .27 0 .07 0. .35 0. 15 0 . 10 DDO ( 14) 0. 01 - 0 .01 0 .24 0 . 13 0 . 10 0 . 15 0 .30 0 . 34 0 .07 0. .01 0. 18 0 . 17 - 0 . 12 DIN ( 15) - 0 . 25 - 0 . 19 - 0 . 32 - 0 . 1 1 - 0 . 26 0 .20 0 . 22 0 .08 - 0 . 27 - 0 . 16 -o. 25 - 0 . , 1 1 - 0 . 30 0 oo UEI ( 16) o. 30 0 .08 0 .09 - 0 .03 0 .03 -0 . 14 - 0 . 28 - 0 . 33 0 . 17 0. 19 0. 1 1 0. 09 0. 7 1 -O 48 - 0 . 6 9 MWE (17) 0. 24 0 . 13 - 0 .04 0 .03 - 0 .02 - 0 . 1 1 -o .25 - 0 .35 0 .04 0. .20 - 0 . 01 0 .03 0. 68 - 0 . .41 - 0 . 3 9 0 .73 MWD ( 18) - 0 . 21 - 0 . 18 - 0 .26 - 0 .24 - 0 . 19 0 .02 0 . 18 0 . 15 - 0 .22 - 0 . 12 - 0 . 18 - 0 , , 1 1 -o. 39 - 0 . .30 0 .62 - 0 . 4 0 - 0 . 4 7 MWI ( 19) 0. 00 0 . 13 0 .39 0 .25 0. .29 0 .07 0 . 19 0 .21 0 .26 0. 00 0. 33 0. 1 1 - 0 . 22 0. 72 - 0 . 25 - 0 . 3 0 - 0 . 4 2 SEI (20) 0. 26 0 . 1 1 - 0 .04 - 0 .01 - 0 .02 - 0 . 1 1 - 0 . 26 - 0 . 38 - 0 .01 0. 24 - 0 . 03 0. 02 0. 71 -o. 43 - 0 . 4 0 0. 74 0 .96 WN» w h i t e n o i s e c o n d i t i o n ; DPCT = s c o r e on d i f f e r e n t i a t i o n PCT; PCT» s c o r e on i n t e g r a t i v e PCT; PCT2» s c o r e on t op 2 s tems; P C T 6 » s c o r e on a l l s i x s tems; RS = r e p r e s s 1 o n - s e n s 11 i zat1 on ; MEMY» d i f f i c u l t y o f memory t a s k ; COMP= d i f f i c u l t y o f computer d i s s i m i l a r i t y t a s k ; 0 I S T - d i s t r a c t i o n r a t i n g o f n o i s e ; LOUD- l oudnes s r a t i n g o f n o i s e ; IRRG" i r r i t a t i n g r a t i n g o f n o i s e ; ARS= p h y s i c a l a r o u s a l r a t i n g o f n o i s e ; DEV» D - s t r e n g t h of e v a l u a t i o n ; DD0= D - s t r e n g t h o f dominance ; DIN= D - s t r e n g t h o f i n d u s t r i o u s n e s s ; UEI= u n s e a l e d E l - r a t i o ; MWE= MDS weight f o r e v a l u a t i o n ; MWD = MDS we i gh t f o r dominance ; MWI» MDS we ight f o r i n d u s t r i o u s n e s s ; SEI= l o g s c a l e d E l - r a t i o . 30 < |r| < . 35 , p < 0 5 |r| > .36. p < .01 Block and Time Effects See Table 3 for the mean overall El-ratios for the two Block conditions and the two Times: There were no significant block or time effects on any of the dependent measures (see A N O V A in Appendix G-Table i for analysis on overall evaluative-importance ratio). MDS Group Space The two multidimensional scaling (MDS) procedures yielded similar three-dimensional group spaces. The correlations between the corresponding dimensions in the two spaces were .98, .92, and .81, respectively. Given that the two programs produced such similar results, I shall subsequently present only one set of results, namely, the M U L T I S C A L E analyses. An elbow in the stress curve suggested that a 3-dimensional solution was most appropriate. From the Figures 3, 4, and 5, it can be seen that all three dimensions appeared as predicted (i.e., evaluation followed by dominance and industriousness) . For example, in Figure 3, the likable and unlikable targets separated into distinct clusters on dimension 1. Table 3. The mean E l - ra t io s for the two Blocks and the two Times. Time Time 1 Time 2 Mn Block EI A 0.56 0.57 0.57 Block B 0.55 0.55 0.55 Mean Time EI 0.55 0.56 F i g u r e 3. C o n f i g u r a t i o n of the e i g h t s o c i a l s t i m u l i a l o n g d i m e n s i o n s one ( e v a l u a t i o n ) and two ( d o m i n a n c e ) . D i m e n s i o n 2 D i m e n s i o n 1 Co F i g u r e 4. C o n f i g u r a t i o n of the e i g h t s o c i a l s t i m u l i on d i m e n s i o n s one ( e v a l u a t i o n ) and t h r e e ( i n d u s t r i o u s n e s s ) D i m e n s i o n 3 I * * • * • * . . • • • * * « * * . * » * * * * , . , » » * i LDL UDL USL LSL D i m e n s i o n 1 USI UDI LSI LDI CO CO I I F i g u r e 5 . C o n f i g u r a t i o n of the e i g h t s o c i a l I****••* + * * * * * * * * * * * * * * * * * * * s t i m u l i on d i m e n s i o n s two (dominance ) and t h r e e D i m e n s i o n 3 ( I n d u s t r l o u s n e s s ) LDL UDL USL LSL D i m e n s i o n 2 UDI LDI USI LSI I I I 41 Effects of noise on the dependent variables I will present the noise effects separately from the individual difference results, although they were analyzed simultaneously in the same A N O V A s . Unsealed measures. a. El-ratio. A DPCT(low, high) x RS(low, high) x White Noise(soft, loud) Analysis of Variance (ANOVA) yielded a significantly higher El-ratio in the loud noise than in the soft noise ' condition, one-tailed F(l,32)= 1.95, p <.04 (loud condition = 0.80, soft= 0.66). After covarying out difficulty of memory task, this noise effect became even stronger, F(l,31) = 7.24, p <.01 (Details in Appendix G-Table ii). (There was no correlation between the covariate and the manipulated variable, in this case white noise.) b. Dimension-strengths (D-strengths). Recall that the D-strength for each dimension is the mean value of the four critical dissimilarities for each respective dimension. A repeated measures A N O V A comparing the three D-strengths across the two noise conditions produced a significant interaction between noise condition and D-strength, F(2,64)= 2.66, p <.04. When difficulty of memory task was covaried out, this interaction effect became even stronger, F(2,62) = 4.11, p <. 01 (see Figure 6) (Details of the A N C O V A are in Appendix G-Table iv). To determine the source of the interaction, separate covariance analyses of the D-strengths were conducted. The D-strength for evaluation was significantly higher in the loud than in the soft noise condition, F(l,39)= 6.73, p <.005 (loud= 5.42, soft= 4.74). There was no noise effect on the D-strength for dominance, but the D-strength for industriousness was significantly lower in the loud noise condition than in the soft condition, F(l,39)= 2.81, p <.05 (soft= 3.85, loud= 3.29). Finally, a repeated measures A N O V A within each noise condition produced significant differences in the D-strengths, with the D-strength for evaluation being significantly larger than the D-strengths of dominance and industriousness for both noise conditions, (p's <.01). 2.5 Figure 6 Effects Of Noise On The Dimension Strength Of Evaluation, Dominance, And Industriousness. Legend LZ2 SOFT NOISE B« LOUD NOISE Evaluation Dominance DIMENSIONS Industriousness 43 Scaled measures. These were the measures emerging from the MDS procedure. M U L T I S C A L E calculated three weights for each subject to indicate how important the three dimensions were in the subject's dissimilarity ratings. a. El-ratio. For each subject, the scaled El-ratio was derived by dividing the weight on evaluation by the sum of the weights on dominance and industrious. The variances of the scaled El-ratio for two noise conditions, however, turned out to be grossly dissimilar. Consequently, these ratios were transformed using the log transformation as suggested by Kirk (1968) to attain homogeneity of variances. This scaled El-ratio was then subjected to a DPCTOow, high) x RS(low, high) x White noise(soft, loud) Analysis of Variance (ANOVA), and a marginally significant noise effect emerged in the predicted direction, F(l ,32)= 1.78, p <.09. After covarying difficulty of memory task, this effect became highly significant, one-tailed F(l,31) = 4.35, p <.02 (loud = -0.11, soft= -0.48). That is, the loud-noise subjects had a significantly higher scaled El-ratio than soft-noise subjects (Details in Appendix G-Table ii). b. Dimension Strengths. The dimension weights were entered into a repeated measures A N O V A . There was a marginally significant interaction between the scaled D-strengths and noise condition, after covarying difficulty of memory task, F(2,62)= 1.88, p <.08 (see Figure 7) (see Appendix G-Table v for details). Further A N C O V A analyses showed that subjects in the loud noise condition placed significantly more weight on evaluation than subjects in the soft condition, F( l ,32)= 3.79, p <.03, one-tailed (loud= 1.06, soft= 0.73). There was no difference between the two noise conditions on the weights placed on dominance and a marginally significant difference between the two noise conditions on the weights placed on industrious, one-tailed F(l,32)= 1.88, p <.08 (soft= 0.92, soft= 0.63), with the soft noise subjects placing more weight on industriousness than did loud noise subjects. 44 From Figure 7, it appears that industriousness has the largest weight in the soft noise condition. However, this weight was not significantly different from the others. In short, subjects in the soft noise condition placed roughly equal weight on all three dimensions. On the other hand, subjects in the loud noise condition weighted evaluation significantly higher than industriousness, F(1,18)= 7.07, p <.02, but not significantly higher than dominance, although it was in the predicted direction. Figure 7 Effects Of Noise On The MULTISCALE Weights Of Evaluation, D o m i n a n c e , A n d Industr iousness. 1.2-1 Legend tZ2 SOFT NOISE m LOUD NOISE Evaluation Dominance DIMENSIONS Industr iousness 45 Effects of individual difference variables. Because the median (1.67) of the six PCT stems was too far below the theoretical cut-off point (2.5) used to distinguish high from low differentiation subjects, I resorted to using the scores of the top two stems instead. The latter measure had a larger range, enabling me to divide the subjects into two fairly equal sized groups using the differentiation value of 2.5. The correlation between the scores using all six stems with the scores using the top two stems was a highly significant .87, indicating that the two scores were closely related. Using the median score of 11 obtained on the RS scale, the subjects were divided into two groups—sensitizers (a score of 11 and higher) and repressors (a score below 11). Unsealed El-ratio. These results are from the same three-factor (DPCT, RS, White noise) A N O V A was used to test the effects of white noise (see Appendix G-Table ii for details). When difficulty of memory task was covaried out, the predicted DPCT X noise condition interaction was marginally significant, F(l,31)= 1.72, p < .08. Also emerging was a marginally significant main effect for DPCT: F(l,31) = 3.32, p < .07, two-tailed, with trait simple (M= .81) scoring higher than the trait complex (M= .68) (See Figure 8). As predicted, tests of the simple main effects showed that the trait complex group evidenced a significant increase in this El-ratio from the soft to the loud noise condition, F(l ,20)= 13.47, p < .002, while the trait simple group did not show a significant change. Scaled El-ratio. The scaled El-ratio was subjected to a three-factor A N O V A model. When difficulty of memory task was covaried out, there was a significant interaction between noise and trait complexity, one-tailed F(l ,31)= 2.92, p <.05 (see Appendix G-Table iii). Again, most of the interaction was accounted for by the significant increase in the scaled El-ratio for the trait-complex from soft to loud noise condition (see Figure 9). These analyses were repeated using scores on the R-S scale as the third factor. There were no RS effects on any of the four dependent measures. In addition, the trait complexity analysis was repeated using the more traditional PCT scoring--i.e., the median value of 1.67 of all six stems was used to divide the group into high integrative complexity and low integrative complexity. A three-factor A N O V A model including RS and noise condition yielded no significant effects of integrative complexity on any of the dependent variables. 47 Figure 8 Interaction Between Trait Complex i ty A n d Noise Condi t ion On The Unsea led Evaluative Impor tance Ratio. 1-1 D _|_ Ld TJ D O <n c ZD 0.8-0.6 0.4 0.2 Legend TRAIT SIMPLE TRAIT COMPLEX SOFT LOUD White Noise Condition 48 Figure 9 Interaction Between Trait Comp lex i t y A n d Noise Condi t ion On The S c a l e d Evaluative Impor tance Ratio. -0.2--0.4 D rr, -0.6-1 O O -0.8- Legend TRAIT SIMPLE TRAIT COMPLEX -1 -1.2 , r SOFT LOUD White Noise Condition Discussion Before providing details, a summary of the results would be useful. First, as in Study 1, our subjects configured the eight social stimuli according to the prescribed dimensions with evaluation as the most important dimension followed by dominance and then industriousness. Second, in Study 2, the predicted noise effects were observed on state complexity: Noise reduced complexity and increased evaluative importance on both measures obtained from the raw dissimilarities (unsealed measures) and the MDS-derived dependent measures (scaled measures). Of the two individual difference variables, only trait complexity moderated the noise effects as evidenced by the predicted interactions between noise and trait complexity. Arousal and the Dynamic Complexity Model It was predicted that loud compared to soft white noise would lead to an increase in the relative importance of the evaluation dimension. It was also predicted that this increase in the relative importance of the evaluation dimension would be a result of a reduction in complexity where the other dimensions became less important in the judgment process. The first hypothesis was strongly supported by the results of this study. This finding held for both the unsealed and scaled El-ratios. These results complemented the earlier findings by Paulhus and Lim (1985), and by Driver (1962) who found increased reliance on evaluation in decision-making process under stress. The results of this study also provided partial support for the second hypothesis. That is, subjects in the loud noise condition compared to soft noise, evidenced a significant decrease in the dimension strength for industriousness. This decrease held for both the unsealed and scaled D-strengths. However, there was also a significantly higher dimension strength for evaluation for subjects in the loud noise condition compared to the soft. This increase in the dimension 50 strength for evaluation also contributed to the El-ratio becoming significantlyiarger. The absolute increase in evaluation may involve a separate process. Several researchers have discussed the polarizing effects of arousal (Scheier & Carver, 1982; Tomkins, 1980). Alternatively, the two effects may be linked: As industriousness drops out, the remaining dimensions, particularly evaluation, would then appear to engulf the judgment. Alternative Explanations At this juncture, I would like to address some alternative interpretations of the noise effects in the present study. We must consider the possibility that the distraction or aversiveness effects of loud noise can explain the findings in this study. Direct evidence is available from the analyses of the manipulation checks: self-report ratings of the physical arousal, distraction, and irritativeness of the noise. I partialed out each of these noise components from the correlation between the noise condition and the dependent measures. If either distraction or aversiveness were instrumental in causing the predicted noise effects, then the correlations between the noise condition and the dependent measures would be significantly smaller than the original correlations. None of these partial correlations was significantly different from the original ones. For example, the first-order correlation between noise condition and the unsealed EI-ratio was .30, and after partialing out self-report distraction became .25. The fact that partialing the arousal rating also failed to reduce this correlation will be dealt with later. A final test of these two alternative models was based on the assumption that if some subjects were distracted more than others, then perhaps, it was the former group that evidenced the observed effects while the latter group did not. I tested this prediction of the distraction model by comparing the distracted subjects with the nondistracted subjects only in the loud noise condition. I found no differences between them on the evaluative importance ratios. The same finding was also noted for aversiveness. In conclusion, neither the distraction nor the aversiveness model could account for the white noise effects on the dependent measures in the present study. 51 Individual differences and the Dynamic complexity model The literature on individual differences in sensitivity and response to general arousal and stress led me to propose two hypotheses. That is, trait-complex individuals and sensitizers should be most affected by the loud white noise, and should evidence the largest reductions in state complexity, and an increased emphasis on the evaluation dimension. In the case of trait complexity, the findings partially supported the hypothesis. That is, the trait complex subjects evidenced an increase in the relative importance of evaluation under arousal, while the trait simples did not. This present study lend further support to the conclusions drawn by both Paulhus and L im (1985) and Driver (1962) that the trait complex individuals would evidence the largest increase in evaluative importance under arousal. Although the individual difference results of the present study were generally promising and in the right direction, they must be considered tentative. The reason is that only marginal support was obtained with the unsealed El-ratio, while significant support was obtained with the scaled El-ratio. Secondly, there were no noise by trait complexity effects observed on the dimension strengths. Thus, it becomes impossible to ascertain if the increase in the relative importance of evaluation among the trait complex under arousal was indeed a result of a reduction in complexity. Finally, I used the PCT to calculate a measure of trait complexity along the differentiation continuum. This is not a common practice—the PCT has been traditionally used to measure integrative complexity. Consequently, my measure of trait complexity may not be a very good measure of differentiation, as evidenced by my having to use the scores of the top two stems instead of all six, because of a restricted range of scores. Nevertheless, the present study has added to the pool of studies showing the effects of arousal on trait complex individuals (Driver, 1962; Paulhus & L im, 1985; Schroder et al, 1967; Suedfeld & Tetlock, 1977). Future research should use a more direct measure of differentiation, for example, Bieri's (1955) modified Rep grid, Scott's " H " (Scott, 1966), or the Judd and Lusk (1984) dimensionality measure. Perhaps then the resultant findings would be a more powerful test of the predictions made by the dynamic complexity model that trait complex individuals would evidence a reduction in state-complexity. With regard to repression-sensitization, loud noise did not affect the sensitizers significantly more than repressors. Both groups placed relatively more emphasis on evaluation when exposed to loud noise. Numerous researchers have shown that cognitive and behavioral differences between these two groups have not always been so clearcut. For example, when subjects were instructed to learn and memorize words for later recall, no differences were found between these two groups on words recalled (Bergquist, Lewinsohn, Sue, & Flippo, 1968). Thus, when both groups were told to concentrate on the task at hand, and performance was measured on that task, existing RS differences disappeared (Holmes, 1974). In the case of the present study involving person perception and judgment, both groups were instructed to memorize the" stimuli for later recall in the judgment process. This instruction should not yield an encoding difference between these two groups. Moreover, because subjects were asked to judge targets other than themselves, a non-ego threatening exercise, retrieval differences were minimized (see Bell & Byrne, 1978). Perhaps if I had asked them to rate aspects of the "self , I would have discovered the expected differences. A final reason why no differences were found in the present study may be the non-threatening aspect of the arousal agent. White noise has been traditionally operationalized as an affectively neutral arousal agent (Eysenck, 1982): It has no affective meaning for the individual. Research has consistently shown that only ego-threat (affective arousal) led to expected RS differences in incidental learning, memory recall, and attraction preferences (Bell & Byrne, 1978; Haley, 1974; Krohne & Schroder, 1972). Perhaps, if I had attached some form of ego-threatening explanation with exposure to 53 white noise (for example, "loud noise will make you anxious"), I might have found the RS differences reported in the existing literature. In closing, it is interesting to note that I may have stumbled upon an even more powerful individual difference variable-namely, the self-reported difficulty of the memorization task. Apparently, this difference in memorizing difficulty masked the true effects of noise on the dependent measures, because after covarying this variable the predicted arousal effects appeared much stronger. This variable was more strongly related to the dependent measures than the other individual difference variables (correlations of -.26, and -.28 with unsealed and scaled El-ratio, respectively). These correlations indicate that subjects who placed relatively less importance on evaluation (i.e., complex subjects) found the memorization task difficult. Perhaps this self-reported memorizing difficulty is tapping a form of trait complexity called evaluative centrality (Scott, Osgood, & Peterson. 1979)—the tendency to use the evaluation dimension to encode and to classify multidimensional information. On the other hand, this individual difference may reflect a choice of memorization strategy. That is, some subjects memorized the targets by focusing on evaluation. Relevant data were available in subjects' ratings of the eights targets on the three characteristics (likable, assertive, and industrious) taken prior to the arousal manipulation. Evidence that a subject had encoded a dimension lay in large differences in ratings of targets originally designated high or low on that dimension. The data showed no differences in these ratings between the subjects who reported the memory task difficult and those who reported it easy. More specifically, the eight targets were accurate^ rated according to their prescribed character portfolios by both groups. In short, the information was encoded equally by the two groups. If evaluative predominance is a trait (Scott, Osgood, & Peterson, 1979), it does not impair recall of other dimensions. Alternatively, focusing on evaluation may be some kind of cognitive strategy instead, that makes it easy to memorize 54 multidimensional stimuli. However, the emphasis on evaluation also carried over to the dissimilarity rating task. In summary, this study could not answer the question of whether differences in memorizing difficulty was an indicant of an underlying trait or a cognitive strategy used by subjects when required to do comparisons of social stimuli. Future research may involve getting subjects to (a) memorize the same targets and (b) rate their acquaintances in a dissimilarity rating task (as in Paulhus & Lim, 1985). Use of evaluation in rating long-time acquaintances should reflect the trait of evaluative centrality. Thus, the observed relation between memorizing difficulty and use of evaluation would more directly answer the question of whether the former indicated a trait variable or a temporary strategy. General Discussion Limitations Because I knew of no practical method, the present study did not have an objective instrument to measure arousal. Instead, arousal was assessed using the self-report method. The measure was collected only once-after the entire experiment-and therefore failed to measure any temporal changes in arousal during continued exposure to the white noise. Consequently, there was no way of ascertaining the pattern of physiological effects of white noise. Note that the physical arousal index failed to mediate the effects of noise on any of the dependent measures when it was used as a covariate. Thus this variable was not the mediator behind the observed noise effects. In retrospect, this is not surprising: The self-report of physical arousal would tap only autonomic arousal. Perhaps I could have also asked subjects to report their level of cortical arousal-"how alert did you feel during the presentation of white noise?" An alternative and more effective cortical measure would be Thayer's (1978) Activation-Deactivation Adjective Checklist. This would have enabled me to test the mediating effects of cortical arousal. In considering an objective measure of cortical arousal, the issue of practicality and utility come into question. E E G is both an expensive and somewhat complicated measure to use. Interpretation of the E E G chart requires much practice. Furthermore, the instrument itself is bulky and nonportable. After the present study was in progress, I became aware of a more practical alternative-critical flicker fusion (CFF). Moreover, its responsiveness to white noise has been verified.(see Paulhus et al., 1987). The C F F instrument is more portable, less expensive, and easier to use. Future research should include simultaneous measures of arousal-both autonomic as well as cortical. An example of the former would be a measure of heart rate (HR). The latter would be Thayer's (1978) self-report and C F F . By including a measure of autonomic arousal, one would be able to reconfirm the finding of habituation of physiological indices to continued exposure to noise. Another limitation of the present study is the use of only one cortical arousal agent. No two arousal agents produce the same effects. Therefore, the findings of the present study are limited in generalizability. In order to generalize to arousal, future research should test other cortical arousal agents. Furthermore, to extend the present findings to the "real" world, a variation of the present study should involve using a socially-oriented arousal agent (e.g., speaking in front of an audience). Other arousal agents to be considered for future use include physical exercise, threat of electric shock, and ego-threatening situations (success-failure paradigm). A final limitation is that the present study looked only at male subjects. Thus, it is impossible to generalize to the female population. Again, future research should include both genders to see if the same cognitive processes are present in the two groups under arousal. New Directions The findings of the present study suggest numerous further studies. The next study should involve administering noise during either the memorizing stage or the dissimilarity rating stage. This would shed light on the effects of arousal on both the encoding and retrieving of information. Another extension of the present study would involve using sets of stimuli varying in task complexity (each set having a different number of dimensions). Subjects would judge the two sets under noise, and the results would reveal which set evidences a larger reduction in complexity. For example, would a five-dimensional set of stimuli evidence a larger reduction in state complexity under arousal than a three-dimensional set of stimuli ? I have earlier suggested further lines of research involving arousal and the individual difference variables used in the present study. A final direction might also address the effects of arousal on integrative complexity by presenting subjects with conflicting information about a target: The time taken to present an integrative statement could be used as a measure of integrative complexity. Consideration might also be given to studies involving behavioral measures, e.g., choosing experimental partners under different levels of arousal. 57 Footnotes 1. O'Hanlon and Beatty (1977) considered E E G to be the best metric of what they call "brain arousal". 2. For some examples of its use in personality and social psychological research, the reader is referred to articles by Bryson (1977), Carroll (1972), Henley (1969), Jones, L. (1983), Passer, Kelley, & Michela (1978), Rosenberg & Sedlak, (1972), Shepard, Romney, & Nerlove (1972), Shoben (1983). 3. A pilot study showed that subjects were able to recall these eight targets fairly accurately after reading these eight descriptions for four minutes. 4. The main purpose for this ten-second delay was to rule out the possible effects of speededness on the dissimilarity responses. By equating the response times across the W N groups, any effects on the dissimilarity ratings will thus be a result of exposure to W N rather than the speed factor. A t-test comparing response times of both W N groups was performed, and no differences were observed in reaction time between the two W N groups. 5. An overall El-ratio was derived by dividing the mean of all 16 dissimilarities on the evaluation dimension with the mean of the sum of the dissimilarities of the dominance and industriousness dimensions. This measure did not control for the other dimensions in the computations, thus adding twelve more "noisy" elements to the index. In a supplemental set of A N O V A s and A N C O V A s , this index showed the same noise effects, but not as clearly. 6. Both MDS programs were used for two reasons. First, the two programs derived their respective group spaces based on different algorithms. I N D S C A L uses an alternating least squares solution, while M U L T I S C A L E uses a maximium likelihood solution (see Schiffman et al, 1982, for a more in depth discussion of these respective solutions). Second, using two M D S programs provides some convergent evidence for the stability of the three dimensions appearing in the group space. 7. The I N D S C A L procedure also yielded the 3-dimensional space as predicted-evaluation followed by dominance and then industriousness. The M U L T I S C A L E procedure, however, extracted industriousness before dominance. Yet, if mean dimension weight is used as a criterion, dominance is actually second and industriousness third as in I N D S C A L (mean weights of .83, and .80, respectively). By all criteria the importance of dominance and industriousness was approximately equal. For simplicity, I am presenting the group space based on the criterion of dimension weight importance. 59 References Anastasi, A . (1964). Fields of applied psychology. New York: MacGraw-Hill. Bell, P. A . , & Byrne, D. (1978). Repression-sensitization. In H . London, & J . E . Exner, Jr . (Eds.), Dimensions of personality (pp 449-485). New York: Wiley & Sons. Bergquist, W. H . , Lewinsohn, P. M . , Sue, D. W., & Flippo, J . P. (1968). Short and long term memory for various types of stimuli as a function of repression-sensitization. Journal of Experimental Research in Psychology, 3, 28-38. Berlyne, D. E. (1960). Conflict, arousal, and curiosity. London: McGraw-Hill. Berlyne, D. E . , Borsa, D. M . , Hamacher, J . H . , & Koenig, I. D. V . (1966). Paired-associated learning and the timing of arousal. Journal of Experimental Psychology, 72, 1-6. Bieri, J . (1955). Cognitive complexity-simplicity and predictie behavior. Journal of  Abnormal and Social Psychology, 51, 263-268. Blechman, E. A . & Dannemiller, E. A. (1976). Effects on performance of perceived control over noxious noise. Journal of Clinical and Consulting Psychology, 44, 601-607. Bobon, D. P., Lecoq, A. , von Frenckell, R., & Lavergne G. (1982). Critical flicker-fusion frequency: Introduction to the CINP session in Goteberg and methodological recommendations. Pharmacopsychiatria, 15, 1-4. Boggs, D. H . & Simon, J . R. (1968). Differential effects of noise on tasks of varying complexity. Journal of Applied Psychology, 52, 148-153. Broadbent, D. E . & Gregory, M . (1965). Effects of noise and of signal rate upon vigilance analyzed by means of decision theory. Human Factors, 7, 155-162. Brock, T. C. (1962). Cognitive restructuring and attitude change. Journal of Abnormal  Social Psychology, 64, 264-271. Bruner, J . S., & Postman, L. (1947). Emotional selectivity in perception and reaction. Journal of Personality, 16, 69-77. Bryson, R. B. (1977). The application of individual differences multidimensional scaling techniques to the study of interpersonal attraction. Multivariate Experimental  Clinical Research, 3, 123-131. Buss, D. M . , & Craik, K . H . (1983). The act frequency approach to personality. Psychological Review, 90, 105-126. Byrne, D. (1961). The Repression-Sensitization Scale: Rationale, reliability, and validity. Journal of Personality, 29, 334-349. Carroll, J . D. (1972). Individual differences and multidimensional scaling. In R. N . Shepard, A . K . Romney, & S. Nerlove (Eds.), Multidimensional scaling: Theory and  applications in behavioral sciences. Vol I. Theory. New York: Academic Press. 60 Carroll, J . D. Chang, J . J . (1970). Analysis of individual differences in multidimensional scaling using via an N-way generalization of Eckart-Young decomposition. Psychometrika, 35, 283-319. Carroll, J . D., & Wish, M . (1974). Multidimensional perceptual models and measurement methods. In E. C. Carterette & M . P. Friedman (Eds.), Handbook of perception (Vol.2) (pp 391-447). New York: Academic Press. Cohen, S. (1978). Environmental load and the allocation of attention. In A . Baum, J . E . Singer, & S. Valins (Eds.), Advances in Environmental Psychology (Vol.1), (pp 1-29). Hillsdale, N . J . : Erlbaum. Cohen, S. & Lezak, A . (1977). Noise and inattentiveness to social cues. Environmental Behavior, 9, 559-572. -Davis, D. R. (1983). Attention, arousal and effort. In A. Gayle & J . A . Edwards (Eds.), Physiological correlates of human behavior: Attention and performance (Vol. 2) (pp 9-34). London: Academic Press. Davis, D. R. & Jones, D. M . (1975). The effects of noise and incentives upon attention in short-term memory. British Journal of Psychology, 66, 61-68. Dong, H . K . (1981)."SIMRRT". The Ball Foundation, Glen Illyn, Illinois. Driver, M . J . (1962). Conceptual structure and group processes in an inter-nation simulation. Part one: The perception of simulated nations. Educational Testing  Service Research Bulletin, RB 62-15. Duffy, E. (1962). Activation and behavior. New York: John Wiley. Easterbrook, J . A -. (1959). The effect of emotion on cue utilization and the organization of behavior. Psychological Review, 66, 183-201. Edsell, R. (1976). Noise and social interaction as simultaneous stressors. Perceptual and  Motor Skills, 42, 1123-1129. Epting, F. R. & Wilkins, G. (1974). Comparison of cognitive structural variables for predicting person perception. Perceptual and Motor Skills, 38, 727-730. Eysenck, M . W. (1983). Memory and arousal. In A. Gayle & J . A . Edwards (Eds.), Physiological correlates of human behavior: Attention and performance (Vol. 2) (pp 187-202). London: Academic Press. Eysenck, M . W. (1982). Attention and arousal: Cognition and performance. New York: Springer-Verlag. Fiske, S. T. (1980). Attention and weight in person perception: The impact of negative and extreme behavior. Journal of Personality and Social Psychology, 38, 889-906. Glass, D. C. & Singer, J . E. (1972). Urban stress: Experiments on noise and social  stressors. New York: Academic Press. Gordon, J . E. (1957). Interpersonal prediction of repressors and sensitizers. Journal of  Personality, 25, 686-698. 61 Gortelmeyer R., & Wiemann, H . (1982). Retest reliability and construct validity of two psychophysical techniques. Pharmacopsychiatria, 15, 24-28. Grandjean, E., Baschera, P., Martin, E. , & Weber, A. (1977). The effects of various conditions on subjective states and critical flicker frequency. In R. R. Mackie (Ed.), Vigilance: Theory, operational performance, and physiological correlates (pp 331-339). New York: Plenum. Grings, W. W., & Dawson, M . E. (1978). Emotions and bodily responses: A  psychophysiological approach. New York: Academic Press. Guetzkow, H . S. (1959). A use of simulation in the study of inter-nation relations. Behavioral Science, 4, 183-190. Gulian, E. & Thomas, J . R. (1986). The effects of noise, cognitive set, and gender on mental arithmetic performance. British Journal of Psychology, 77, 503-511. Haley,G. A . (1974). Eye movement responses of repressors and sensitizers to a stressful film. Journal of Research in Personality, 8, 88-94. Hamilton, P., Hockey, G. R. J . , & Quinn, J . G. (1972). Information selection, arousal, and memory. British Journal of Psychology, 63, 181-189. Hamilton, P., Hockey, G. R. J . , & Rejman, M . (1977). The place of the concept of activation in human information processing theory: An integrative approach. In S. Dornic, (Ed.), Attention and performance (Vol. 6). Hillsdale, N J : Erlbaum. Hancock, P. A. (1984). Environmental stressors. In J . S. Warm (Ed.), Sustained attention  in human performance (pp 103-142). New York: Wiley. Harris, R. M . (1981). Conceptual complexity and preferred coping strategies in anticipation of temporally predictable and unpredictable threat. Journal of Personality and  Social Psychology, 41, 380-390. Hartley, L . R. (1973). Effect of prior noise or prior performance on serial reaction. Journal  of Experimental Psychology, 101, 255-261. Hartley, L . R. (1974). Performance during continuous and intermittent noise and wearing ear protection. Journal of Experimental Psychology, 102, 512-516. Harvey, O. J . , Hunt, D. E. , & Schroder, H . M . (1961). Conceptual systems and personality  organization. New York: Wiley. Hathaway, S. R. & McKinley, J . C. (1951). Manual for the Minnesota Multiphasic  Personality Inventory. New York: Psychological Corporation. Henley, N . M . (1969). A psychological study of the semantics of animal terms. Journal of  Verbal Learning and Verbal Behavior, 8, 176-184. Hockey, G. R. J . (1970). Effect of loud noise on attentional selectivity. Quarterly Journal  of Experimental Psychology, 22, 37-42. 62 Hockey, G. R. J . & Hamilton, P. (1970). Arousal and information selection in short-term memory. Nature, 226, 866-867. Holmes, D. S. (1974). Investigations of repression: Differential recall of material experimentally or naturally associated with ego threat. Psychological Bulletin, 81, 632-653. Jones, D. M . (1983). Noise. In G. R. J . Hockey (Ed.), Stress and fatigue in human  performance (pp 61-95). New York: Wiley. Jones, L . E . (1983). Multidimensional models of social perception, cognition, and behavior. Applied Psychological Measurement, 7, 451-472. Judd, C. M . , & Lusk, C. M . (1984). Knowledge structures and evaluative judgments: Effects of structural variables on judgmental extremity. Journal of Personality and  Social Psychology, 46, 1193-1207. Kahneman, D. (1973). Attention and effort. Englewood Cliff, N . J . : Prentice-Hall. Kammer, P. P. (1984). Conceptual level of development as it relates to student participation in gifted programs. Gifted Child Quarterly, 28, 89-91. Kelly, G. A . (1955). The psychology of personal constructs (2 vols.). New York: Norton. Kirk, R. E . (1968). Experimental design: Procedures for the behavioral sciences. Belmont, C A : Brooks/Cole. Kranda, K . (1982). Factors influencing flicker sensitivity. Pharmacopsychiatria, 15, 9-15. Krohne, H . W. (1978). Individual differences in coping with stress and anxiety. In C. D. Spielberger & I. G. Sarason (Eds.), Stress and anxiety (Vol.5) (pp 233-261). Washington, D. C : Hemisphere. Krohne, H . W. & Rogner, J . (1982). Repression-Sensitization as a central construct in coping research. In H . W. Krohne & L . Laux (Eds.), Achievement, stress, and  anxiety (pp 167-193). Washington, D . C : Hemisphere. Krohne, H . W. & Schroder, H . M . (1972). Anxiety defense and complex information processing. Archiv fur Psychologie, 124, 50-61. Kruskal, J . B. & Wish, M (1978). Multidimensional scaling. Beverley Hills: S A G E . Kryter, K . D. (1985). The effects of noise on man (2nd ed.). New York: Academic Press. Kryter, K . D. (1970). The effects of noise on man. New York: Academic Press. Lacey, J . I. (1967). Somatic response patterning and stress: Some revisions of activation theory. In M . H . Appley, & R. Turnbull (Eds.), Psychological Stress. New York: Appleton-Century-Crofts. Leary, T. (1957). Interpersonal diagnosis of personality. New York: Ronald Press. Linden, W. (1987). Effects of noise distraction during mental arithmetic on phasic cardiovascular activity. Psychophysiology, 24, 328-333. 63 Linville, P. W. (1982). Affective consequences of complexity regarding the self and others. In M . S. Clark & S. T. Fiske (Eds.), Affect and cognition: The seventeenth annual  Carnegie symposium on cognition (pp 79-109). Hillsdale, N . J . : Erlbaum. Loeb, M . (1981). The present state of research on the effects of noise: Are we asking the right questions ? The Journal of Auditory Research, 21, 93-104. Loeb, M . , & Alluisi, E. A . (1984). Theories of vigilance. In J . S. Warm (Ed.), Sustained  attention in human performance (pp 180-205). New York: Wiley. Mandler, G., & Sarason, S. B. (1952). A study of anxiety and learning. Journal of  Abnormal and Social Psychology, 47, 166-173. Mischel, W., Ebbesen, E. B. , & Zeiss, A. M . (1976). Determinants of selective memory about the self. Journal of Consulting and Clinical Psychology, 44, 92-103. O'Hanlon, J . F. , & Beatty, J . (1977). Concurrence of electroencephalographic and performance changes during a simulated radar watch and some implications for the arousal theory of vigilance. In R. R. Mackie (Ed.), Vigilance: theory, operational  performance and physiological correlates (pp 189-201). New York: Plenum. O'Malley, J . J . & Poplawsky, A . (1971). Noise induced arousal and breadth of attention. Perceptual and Motor Skills, 33, 887-890. Osgood, C. E. (1976). Focus on meaning. Paris: Morton. Osgood, C. E., Suci, G. J . , & Tannebaum, P. H . (1957). The measurement of meaning. Chicago: Univ of Illinois Press. Passer, M . W., Kelley, H . H . , & Michela, J . L . (1978). Multidimensional scaling of the causes for negative interpersonal behavior. Journal of Personality and Social  Psychology, 36, 951-962. Parrott, A . C. (1982). Critical flicker fusion thresholds and their relationship to other measures of alertness. Pharmacopsychiatria, 15, 39-43. Paulhus, D. L . (August, 1984). Self-deception: An information processing approach. Paper presented at the meeting of the American Psychological Association, Toronto. Paulhus, D. L. & Levitt, K . (1987). Affect triggers desirable responding: Automatic egotism ? Journal of Personality and Social Psychology, 52, 245-259. Paulhus, D. L . & Levitt, K . (1983). A revised Repression-sensitization scale: A mini-RS. Unpublished manuscript, University of British Columbia. Paulhus, D. L . & Lim, D. T. K . (August, 1985). The effects of arousal on cognitive  complexity and evaluation. Paper presented at the meeting of the American Psychological Association, Los Angeles. Paulhus, D. L . , L im, D. T. K. , Reid, D. G., & Murphy, G. (August, 1986). Exercise, Arousal and social judgments. Paper presented at the meeting of the American Psychological Association, Washington, D. C. Paulhus, D. L . , Murphy, G., & Reid, D. G. (1986). Comparison of different arousal agents  on task performance. Unpublished manuscript, University of British Columbia. Paulhus, D. L . , Reid, D. G., & Lim, D. T. K . (1987). C F F and white noise. Unpublished manuscript, University of British Columbia. Paulhus, D. L . & Suedfeld, P. (in press). Self-deception: A dynamic-complexity model. In J . S. Lockard & D. L . Paulhus (Eds.), Self-deception: An adaptive mechanism ? New York: Prentice-Hall. Porter, C. A. & Suedfeld, P. (1981). Integrative complexity in the corresponding of literary figures: Effects of personal and societal stress. Journal of Personality and Social  Psychology, 40, 331-345. Posner, M . I. (1975). Psychobiology in attention. In M . S. Gazzaniga & C. Blackmore (Eds.), Handbook of psychobiology (pp 441-480). New York: Academic Press. Ramsey, J . 0. (1982). M U L T I S C A L E II manual. McGill University. Ramsey, J . 0. (1977). Maximum likelihood estimation in multidimensional scaling. Psychometrika, 43, 241-266. Rosen, S. (1970). Noise, hearing, and cardiovascular functioning. In B. L. Welch & A. S. Welch (Eds.), Physiological effects of noise. New York: Plenum Press. Rosenberg, S. & Sedlak, A . (1972). Structural representations of implicit personality theory. In L . Berkowitz (Ed.), Advances in Experimental Social Psychology, (Vol. 6) (pp 235-297). New York: Academic Press. Roth, W. T., Dorato, K . H . , & Koppell, B. S. (1984). Intensity and task effects on evoked physiological responses to noise bursts. Psychoph3'siology, 21, 466-481. Scheier, M . & Carver, C. (1982). Self-focussed attention and the experience of emotion: Attraction, repulsion, elation, and depression. Journal of Personality and Social  Psychology, 35, 625-636. Schiffman, S. S., Reynolds, M . L. , & Young, F. W. (1982). Introduction to multidimensional scaling: Theory, methods, and applications. New York: Academic Press. Schroder, H . M . & Suedfeld, P. (1971). Personality theory and information processing. New York: Ronald. Schroder, H. M . , Driver, M . J . , & Streufert, S. (1967). Human information processing:  Individuals and groups functioning in complex social situations. New York: Holt, Rinehart, & Winston. Scott, W. A. (1966). Measures of cognitive structure. Multivariate Behavioral Research, 1, 391-395. Scott, W. A. , Osgood, D. W., & Peterson, C. (1979). Cognitive structure: Theory and  measurement of individual differences. Washington, D. C : Winston/Wiley. 65 Shepard, R. N . , Romney, A. K. , & Nerlove, S. B. (1972). Multidimensional scaling: Theory  and applications in the behavioral sciences (2 vols.). New York: Seminar Press. Shoeben, E. J . (1983). Applications of multidimensional scaling in cognitive psychology. Applied Psychological Measurement, 7, 473-490. Slugoski, B. R., Marcia, J . E. , & Koopman, R. (1984). Cognitive and social interactional characteristics of ego identity statuses in college males. Journal of Personality and  Social Psychology, 47, 646-661. Smith, A . P., Jones, D. M . , & Broadbent, D. E. (1981). The effects of noise on recall of categorized lists. British Journal of Psychology, 72, 299-316. Smith, J . M . , & Misiak, H . (1976). Critical flicker frequency (CFF) and psychotropic drugs on normal human subjects- A review. Psychopharmacology, 47, 175-182. Streufert, S. & Streufert, S. C. (1978). Behavior in the complex environment. Washington, D. C : V . H . Winston. Suedfeld, P. (1981). Indices of world tension in the Bulletin of the Atomic Scientists. Political Psychology, 2, 114-123. Suedfeld, P., & Rank, D. A. (1976). Revolutionary leaders: long-term success as a function of changes in conceptual complexity. Journal of Personality and Social Psychology, 34, 169-178. Suedfeld, P. Tetlock, P. E . (1977). Integrative complexity of communication in international crises. Journal of Conflict Resolution, 21, 169-184. Suedfeld, P., Tetlock, P. E. , & Ramirez, C. (1977). War, peace, and integrative complexity. Journal of Conflict Resolution, 21, 427-442. Sullivan, H . S. (1953). The interpersonal theory of psychiatry. New York: Norton. Swenson, R. A . , Tucker, D. M . (1983). Multivariate analysis of E E G coherence: Stability of the metric, individual differences in patterning and response to arousal. Biological Psychology, 17, 59-75. Tetlock, P. E . (1979). Identifying victims of groupthink from public statements of decision makers. Journal of Personality and Social Psychology, 37, 1314-1324. Thayer, R. E . (1978). Toward a psychological theory of multidimensional activation (arousal). Motivation and Emotion, 2, 1-34. Thayer, R. E . & Carey, D. (1974). Spatial stimulus generalization as a function of white noise and activation level. Journal of Experimental Psychology, 102, 539-542. Tomkins, S. S. (1980). Affect as amplification: Some modifications in theory. In R. Plutchik, & H . Kellerman (Eds.), Emotion: Theory, research, and experience (vol. 1) (pp. 141-164). N Y : Academic Press. Vannoy, J . S. (1965). Generality of cognitive complexity-simplicity as a personal construct. Journal of Personality and Social Psychology, 2, 385-396. 66 Voicu, C. (1970). Inference of the level of bioelectric arousal on the mnemonic efficiency of verbal and non-verbal stimuli. Revue Roumaine de Sciences Socialis: Serie de  Psychologie, 14, 151-162. Wallsten, T. S. & Barton, C. (1982). Processing probabilistic multidimensional information for decisions. Journal of Experimental Psychology: Learning, Memory, and  Cognition, 8, 361-384. Wiggins, J . S. (1979). A psychological taxonomy of trait-descriptive terms: The Interpersonal domain. Journal of Personality and Social Psychology, 37, 395-412. Winter, L . , Uleman, J . S., & Cunniff, C. (1985). How automatic are social judgments ? Journal of Personality and Social Psychology, 49, 904-917. Yerkes, R. M . & Dodson, J . D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal of Comparative and Neurological Psychology, 18, 459-482. ~ 67 Appendix A Behavioral Descriptions of the Eight Social Stimuli Person A : -- Instead of doing his own assignments, A L A N demands that others let him copy theirs. (Unlikable, _Dominant,_Lazy) * Person B: - At the barbeque, BOB's jokes about how lazy he is effectively entertained the group. (Likable, _Dominant,_Lazy) Person C: - When a friend falls behind in his classwork, CHRIS usually takes over the situation and does some of his work for him. (Likable, Dominant, Industrious) Person D: —At dinner-study sessions with his friends, D A V E always gets talked into cleaning up. ^ikable,J3ubmissive,_Industrious) Person E: - Often in the middle of exam period, ED would cheerfully let his friends talk him into going for a movie. (Likable,_Submissive,_Lazy) Person F: - At study groups in his flat, F R A N K usually gives advice even when none is requested. (Unlikable, rJominant,_Industrious) Person G: - After geography class, when his friends want to borrow his notes, G A R Y whines about it. (TJnlikable,^ubmissive,_Industrious) Person H : - At home, H A R R Y constantly grumbles about other students working so hard. (Unlikable,_Submissive,_Lazy) "^Character-type descriptions in parentheses are not presented to subjects. 68 Appendix B Please rate each person on each of the following characteristics using the 7-point scale below. 1 2 3 4 5 6 7 not at very all much so Likable Assertive Industrious Person: A L A N (ASSIGNMENTS) BOB (BARBEQUE) CHRIS (CLASSWORK) D A V E (DINNER) E D ( E X A M ) F R A N K (FLAT) G A R Y (GEOGRAPHY) H A R R Y (HOME) 69 Appendix C The 2 blocks of the 28 pairs for the similarity judgments of the eight social stimuli- Block A Block B "Crucial" Pairs CHRIS (LDI) w F R A N K (UDI) 1 A L A N (UDL) w BOB ( L D L ) 1 E D (LSL) w H A R R Y ( U S D 1 D A V E (LSI) w G A R Y (USI) 1 BOB (LDL) w ED ( L S L ) 2 A L A N (UDL) w H A R R Y (USL) 2 , F R A N K (UDI) w G A R Y ( U S D 2 CHRIS (UDI) w D A V E (LSI) 2 G A R Y (USI) w H A R R Y ( U S D 3 A L A N (UDL) w F R A N K (UDI) 3 BOB (LDL) w CHRIS (LDI) 3 D A V E (LSI) w ED (LSL3) Other pairs E D (LSL) w G A R Y (USI) CHRIS (LDI) w G A R Y (USI) A L A N (UDL) w D A V E (LSD CHRIS (LDI) w ED ( L S D BOB (LDL) w G A R Y (USI) D A V E (LSI) w H A R R Y (USL) BOB (LDL) w D A V E (LSI) ED (LSL) w F R A N K (UDI) A L A N (UDL) w ED (LSL) BOB (LDL) w H A R R Y (USL) F R A N K (UDI) w H A R R Y (USD BOB (LDL) w F R A N K (UDI) D A V E (LSI) w F R A N K (UDI) A L A N (UDL) w G A R Y (USI) A L A N (UDL) w CHRIS (LDI) CHRIS (LDI) w H A R R Y (USL) 1 "crucial" pair for Evaluation. 2 "crucial" pair for Dominance. 3 "crucial" pair for Industrious. Appendix Using the 7-point scale below, please answer the following questions: 1 2 3 4 5 6 7 not at all very much so 1. How difficult did you find the memory task ? 2. How difficult did you find the similarity judgments task ? 3. How distracting did you find the white noise ? 4. How loud did you find the white noise ? 5. How irritating did you find the white noise ? 6. How physically arousing did you find the white noise ? Thank you for your participation. C o r r e l a t i o n among the Independent and dependent measures i n t h e s o f t n o i s e c o n d i t i o n (N=20) (1) (2 ) (3) 14) (5) (6) ( ? ) (6) (9) DO) (11) (12) ( 13) (14) (15) ( 16) DPCT ( 1) PCT (2 ) 0 . 79 PCT2 (3) 0. .87 0 .80 PCT6 (4) 0 86 0 . 90 0 . 84 RS (5) 0. . 44 0 . 4 1 0 . 53 0 . 40 MEMV (6) -0 43 -0 . 23 -0 . 37 -o . 36 -0 . 32 RATG (7) -0 . 30 -0 .31 -0 . 36 -0 . 34 -0 08 0 . 62 DIST (8) -0 .01 -0 . 10 -0 .01 -0 .05 -0 . 10 0 . 34 0 . 256 LOUD (9) 0 .07 o . 10 -0 .07 0 . 19 -0 .OO 0 .09 -0 . 22 0 . 35 IRRG( 10) 0. 13 0 . 13 0 .09 0 .07 0 03 0 .31 0 . 18 0 .64 0 .51 ARS ( 1 1 ) 0. 15 o . 27 0 op 0 .07 0 . 27 -o .06 0 .06 0 . 26 0 . 25 0 35 DEV ( 12) 0. 05 -0 . 07 -0 •01, -0 .07 0 .07 -0 . 19 -0 . 15 -0 . 17 -0 . 27 -0 13 -o . 14 DD0 ( 13) 0. 04 o 34 0 . 24 o . 13 0 . 29 O . 36 0 . 38 0 . 1 1 -0 . 10 0. 33 0 38 -0 2 1 DIN ( 14) -o. 3 1 -o 18 -o .06 -o . 27 0 28 0 . 13 -0 .06 -0 .02 -0 .03 -0 02 o 14 -O 27 o 16 UEI ( 15) 0. 08 -0 . 17 -0 • 1.5 -0 .06 -o . 12 -0 .31 -0 . 17 -0 .09 -0 . 14 -o 25 -0. 22 0. 81 -0 58 -0.59 MWE ( 16) 0 06 -o 25 -o 09 -0 . 1 1 -0 .07 -0 . 29 -0 . 17 -o . 15 -0 .07 -o. 16 -0 24 0 73 -o 37 -0 . 47 0.77 MWD ( 17) -o. 12 -o .08 -o . 19 -0 . 1 1 -0 .06 0 .08 -0 .08 -0 .09 0 . 23 -o. 04 0. 2 1 -0. 40 -0 30 0.57 -0. 33 -O. 55 MWI ( 18) 0. 25 0 . 52 0 . 35 0 . 37 0 . 13 0 . 39 0 .31 0 . 36 0 . 14 0. 47 0. 29 -o. 40 0. 78 -0.04 -0.57 -0. 46 SEI ( 19) 0. 00 -o . 22 -o . 14 -0 . 10 -0 .09 -0 . 32 -0 . 26 -0 . 24 0 .01 -0. 17 -0. 30 0. 76 -0. 44 -0. 50 0.82 0. 96 DPCT» s c o r e on d i f f e r e n t i a t i o n PCT; PCT* s c o r e on i n t e g r a t i v e PCT; PCT2= s c o r e on top 2 stems; PCT6= s c o r e on a l l s i x stems; RS" r e p r e s s 1 on-sensi11za11 on; MEMY* d i f f i c u l t y of memory t a s k ; C0MP = d i f f i c u l t y of computer d i s s i m i l a r i t y task DIST» d i s t r a c t i o n r a t i n g of n o i s e : LOUD* loudness r a t i n g of n o i s e ; IRRG= i r r i t a t i n g r a t i n g of n o i s e ; ARS = p h y s i c a l a r o u s a l o f n o i s e ; DEV = D - s t r e n g t h of e v a l u a t i o n ; DD0= D - s t r e n g t h of dominance; DIN" D - s t r e n g t h of i n d u s t r i o u s n e s s ; UEI= u n s e a l e d E l - r a t i o ; MWE = MDS weight f o r e v a l u a t i o n ; MWD" MOS weight f o r dominance; MWI" MDS w e i g h t f o r i n d u s t r i o u s n e s s ; S E I * l o g s c a l e d E l - r a t i o . 44 < | r < .54, p <.05 | r > .55. p < .OI C o r r e l a t i o n among Independent and dependent measures i n the l o u d n o i s e c o n d i t i o n (N=20) ( D (.2) (3) (4) (5) (6) (8) (9) ( 10) (11) (12) ( 13) (14) ( 15) ( 16) DPCT ( 1 ) PCT (2) 0 56 PCT2 (3) 0 87 0 .68 PCT6 (4) 0 84 0 . 78 O .91 RS (5) -0 . 12 -0 . 44 -0 . 16 -0 . 24 MEMY (6) -0 . iO -0 . 25 -o .03 -0 .05 -0 .03 RATG (7) -0 . 2 1 -0 . 2 1 -o . 19 -o .07 0 . 49 0 .40 DIST (8) -0. . 27 -0 . 32 -0 . 29 -0 . 29 0 .47 -0 .07 0 . 4 1 LOUD (9 ) -0. 10 -o . 14 0 OO -o . 12 o 31 0 . 28 O . 44 O . 27 I RRG( 10) -o. .42 -o . 29 -o . 34 -0 . 26 0 . 43 -o .09 0 . 50 0 . 77 O . 38 ARS ( 1 1 ) 0 30 0 .00 o . 14 0 . 10 0 . 38 -0 . 37 0 . 12 0 . 43 O . 40 0 35 DE V ( 12) 0. 12 0 . 36 0 . 2 1 0 . 15 -0 . 43 -0 .00 -o .37 -o . 12 0 . 37 -0 .03 0 .01 000 ( 13) -0. 07 0 . 14 -0 .01 0 .06 -0 .09 0 . 22 0 . 29 0 .02 0 . 1 1 0. 1 1 -0 08 0 01 DIN ( 14) -o 13 -o . 38 -o . 15 -0 . 26 o . 16 0 . 42 0 . 14 -0 . 29 0 . 15 -0. 21 -o 1 1 -o .21 -0. . 14 UEI ( 15) 0. 12 0 . 24 0 .07 0 . 1 1 -o . 2 1 -0 . 43 -0 . 44 0 . 14 -0 .08 0. 07 0 14 0 53 -0. 42 -O. 73 MWE ( 16) 0. 28 0 . 1 1 o . 18 0 .09 -0 . 20 -0 . 35 -0 . 55 -0 .01 0 .07 -o. 20 o. 15 o 54 -O 50 -O. 26 0.67 MWD ( 17 ) -0. 27 -0 . 36 -0 . 28 -0 . 27 0 . 13 0 . 40 0 . 33 -0 . 19 O .02 -o 10 -0. 25 -o 30 -O. 32 0.62 -0. 40 -O. 34 MWI ( 18) -0 00 0 . 27 0 . 15 0 . 2 1 -0 00 -0 .03 0 . 10 0 . 25 -0 . 10 0. 34 -0. 08 -o. 02 0. 66 -0.42 -0.08 -0 40 SEI ( 19) 0 36 0 . 13 0 . 2 1 0 . 10 -o . 2 1 -o . 35 -o .56 -o .03 0 .09 -o. 28 0. 27 0. 51 -o. 47 -O. 22 0.62 0.97 -0.38 -0.42 > DPCT» s c o r e on d i f f e r e n t i a t i o n PCT; PCT« s c o r e on i n t e g r a t i v e PCT; PCT2- s c o r e on top 2 stems; PCT6- s c o r e on a l l s i x fp stems; RS" r e p r e s s i o n - s e n s i t i z a t i o n ; MEMY» d i f f i c u l t y of memory t a s k ; COMP- d i f f i c u l t y o f computer d i s s i m i l a r i t y t a s k ; DIST- d i s t r a c t i o n r a t i n g of n o i s e ; LOUD- loudness r a t i n g of n o i s e ; IRRG- i r r i t a t i n g r a t i n g o f n o i s e ; ARS* p h y s i c a l & a r o u s a l o f n o i s e ; DEV- D - s t r e n g t h of e v a l u a t i o n ; DDO= D - s t r e n g t h of dominance; DIN- D - s t r e n g t h of i n d u s t r i o u s n e s s ; ^" UEI- u n s e a l e d E l - r a t i o ; MWE= MDS weight f o r e v a l u a t i o n ; MWD- MDS weight f o r dominance; MWI- MDS w e i g h t f o r ^ i n d u s t r i o u s n e s s ; SEI= l o g s c a l e d E l - r a t i o . ^ 44 < | r | < .54. p < 0 5 | r | > .55. p < .01 Appendi* Appendix of ANOVA t a b l e s R e s u l t s of Repeated ANOVA on o v e r a l l u n s e a l e d E l - r a t i o . Source of v a r i a t i o n SS df MS F p Between S u b j e c t s Order 0.002 1 0.002 0.28 .60 E r r o r 0.259 39 0.007 W i t h i n S u b j e c t s Block (B) 0.005. 1 0.005 2.04 .17 B X Order 0.001 1 0.001 0.22 .64 E r r o r 0.090 38 0.002 Order has two l e v e l s (1= Blk A i n time 1; 2=Blk B i n time 1). R e s u l t s of t h r e e - f a c t o r A n a l y s i s of Covariance (ANCOVA) on unsealed E l - r a t i o Sources of v a r i a t i o n SS df MS F P Noise Condition(WN) 0.313 1 0.313 7. 24 .01 T r a i t - C o m p l e x i t y ( T C ) 0.144 1 0. 1 44 3. 32 .07 Repn-Sensn (RS) 0.031 1 0.031 0. 71 .40 TC X RS 0.007 1 0.007 0. 16 .69 WN X TC 0.074 1 0.074 1 . 72 .08 WN X RS 0.006 1 0.006 0. 1 3 .72 WN X TC X RS 0.131 1 0.131 3. 02 .09 D i f f i c u l t y of memory ( c o v a r i a t e ) 0.489 1 0.489 1 1 . 31 .002 E r r o r 1.341 31 0.043 R e s u l t s of t h r e e - f a c t o r ANCOVA on s c a l e d EI - r a t i o Sources of v a r i a t i o n SS df MS F P Noise C o n d i t i o n (WN) 1 .205 1 1 .205 4.35 .05 T r a i t Complexity(TC) 0.429 1 0.429 1 .55 . 1 1 Repn-Sensn (RS) 0.055 1 0.055 0.20 .66 TC X RS , 0.000 1 0.000 0.00 .99 WN X TC 0.811 1 0.811 2.92 .05 WN X RS 0. 194 1 0. 194 0.70 .41 WN X TC X RS 1 .276 1 1 .276 4.60 .04 D i f f i c u l t y of memory ( c o v a r i a t e ) 2.695 1 2.695 9.72 .004 E r r o r 8.594 31 0.277 75 i v . R e s u l t s of t h r e e - f a c t o r repeated ANCOVA on un s e a l e d  dimension s t r e n g t h Sources of v a r i a t i o n SS df MS F P Between S u b j e c t s Noise Condition(WN) 0.026 0.026 0. 05 .83 T r a i t Complexity (TC) 0.660 ' 0.660 1 . 19 .28 Repn-Sensn (RS) 0.614 1 0.614 1 . 10 .30 TC X RS 0.079 • 1 0.079 0. 14 .71 WN X TC 0.090 1 0.090 0. 16 .69 WN X RS 1 .875 1 1 .875 3. 37 .08 WN X TC X RS 0.032 1 0.032 0. 06 .81 D i f f i c u l t y of memory ( c o v a r i a t e ) 3.462 1 3.462 6. 22 .02 E r r o r 17.244 31 0.556 W i t h i n S u b j e c t s Dimension (D) 28.011 2 14.006 15. 89 .0001 D X WN 7.239 2 3.620 4. 1 1 .02 D X TC 1 .475 2 0.737 0. 84 .44 D X RS 0.892 2 0.446 0. 51 .61 D X TC X RS 1 .765 2 0.883 1 . 00 .37 D X WN X TC 2.653 2 1 .326 . 1 . 50 .16 D X WN X RS 0.366 2 0. 183 0. 21 .81 D X WN X TC X RS 1 .905 2 0.952 1 . 08 .35 D X D i f f i c u l t y of memory ( c o v a r i a t e ) 6.981 2 3.490 3. 96 .02 E r r o r 54.644 62 0.881 76 v. Results of three-factor repeated ANCOVA on scaled Sources of var iat ion SS df MS F P Between Subjects Noise Condition(WN) 0. 008 1 0. 008 0. 25 .62 Tra i t Complexity (TC) 0. 007 1 0. 007 0. 21 .65 Repn-Sensn (RS) 0. 013 1 0. 013 0. 41 .52 TC X RS 0. 003 1 0. 003 0. 10 .75 WN X TC 0. 001 1 0. 001 0. 02 .90 WN X RS 0. 028 1 0. 028 0. 88 .36 WN X TC X RS 0. 005 1 0. 005 0. 17 .68 D i f f i c u l t y of memory (covar iate) 0. 1 03 1 0. 1 03 3. 20 .08 Error 1 . 002 31 0. 032 Within Subjects Dimension (D) 3. 569 2 1. 785 3. 84 .03 D X WN 1 . 752 2 0. 876 1 . 88 .08 D X TC 0. 518 2 0. 518 1 . 1 1 .33 D X RS 0. 222 2 0. 111 0. 24 .79 D X TC X RS 0. 265 2 0. 1 33 0. 28 .75 D X WN X TC D. 863 2 0. 432 0. 93 .40 D X WN X RS 0. 314 2 0. 1 57 0. 34 .71 D X WN X TC X RS 1 . 578 2 0. 789 1 . 70 . 1 9 D X D i f f i c u l t y of memory (covariate) 3. 680 2 1. 840 3. 96 .02 Error 28. 840 62 0. 465 77 Appendix H Paragraph Completion Test Complete the following stems with a paragraph of 2 to 3 sentences each. Please do not spend than two minutes on any single stem. 1. When I am in doubt... 2. When I am criticized... 3. Parents.. 4. Rules... 5. Confusion... 6. Criticism... 79 Appendix H Short R-S Scale For each question below, circle the best answer - True or False. Answer all questions. 1. T F 2. T F 3. T F 4. T F 5. T F 6. T F 7. T F 8. T F 9. T F 10. T F 11. T F 12. T F 13. T F 14. T F 15. T F 16. T F 17. T F 18. T F 19. T F 20. T F 21. T F couldn't get going. 80 22. T F I wish I were not so shy. 23. T F I frequently find myself worrying about something. 24. T F I brood a great deal. 25. T F I believe I am no more nervous than most others. 26. T F It is safer to trust nobody. 27. T F When I leave home, I do not worry about whether the door is locked and the windows closed. 28. T F 1 am always disgusted with the law when a criminal is freed through the arguments of a smart lawyer. 29. T F Life is a strain for me much of the time. 30. T F I think nearly anyone would tell a lie to keep out of trouble. 31. T F I easily become impatient with people. 32. T F I forget right away what people say to me. 33. T F I often feel as if things were not real. 34. T F I have been afraid of things or people that I knew could not hurt me. -35. T F I have several times given up doing a thing because I thought too little of my ability. 36. T F Almost every day something happens to frighten me. 37. T F At periods, my mind seems to work more slowly than usual. 81 38. T F People often disappoint me. 39. T F Often, even though everything is going fine for me, I feel that I don't care about anything. 40. T F It makes me feel like a failure when I hear of the success of someone I know well. 41. T F I worry quite a bit over possible misfortunes. 42. T F I have a daydream life about which I do not tell other people. 43. T F I sometimes feel that I am about to go to pieces. Key to Short R-S Scale Scoring Give one point for a response of "False" to the following items: 1, 5, 6, 15, 16, 17, 21, 25, 27 Give one point for a response of "True" to all other items. A high score indicates a sensitizer; a low score indicates a repressor. Norms - sample of 246 university students Mean SD _n Alpha Males 28.6 6.5 94 .86 Females 27.4 8.0 152 .90 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items