r_.- _. _. 15‘-‘>“‘9:V-. . - -_.. V “if n- c.‘ t 23”!“ in. ‘ :‘g 2:)". \‘9 :5I. . “"K ,_.-_,, .1 "r 5 £2: .1 3' m. _ H5): - ‘sh ”(12,"? (I ‘ . _-{; A — . n J. I"'. 9“! u $.91; ‘ 4 chiv“\ :i‘g." ‘1 gig” SEW" . ‘ J%\_‘e . \' s . H} ' J!L7V\ $33K.“ ’- I 2 luvs“ ‘ “ “ ‘ . 2.252“ mg! 1’4?" ’ . 13.;f‘;,,fi, .1 L , 4‘- 51 I . a?) 1"; Jflo'il ‘22.}: Ni 35’! {'20 l—“Kr” ‘Czl:1I', hr” :2: :gn 153:, Y Q'gejI, i" . (Nth; 1 HI; I figgi 10131:“ LI I ‘. 2%). 0' . .‘gy s‘~ :I. ,5 rmI .a.. "Lyn”,- '-'/ r . £5; ' j’l.’ .. ‘ ' ’3 n . n .IVI'J“; V - .4 55' “m." . 'r,’ ‘y . yum”, ’1". II‘ ~ :4‘1' .I")‘a,'£t:l‘a'l.I‘l. ,; 1:1, I '12:: u I .‘1’“!.'; In a II ~I.-, crib, .. "r ’::".$l'f’.;;.l\. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII III 133L339, IIIIII III ”RWY ‘ Michigan Stat E University 3 This is to certify that the dissertation entitled SOCIAL SUPPORT AND ATTENTION TO SOCIAL CUES IN TYPE A INDIVIDUALS presented by Lance Edward Schumacher has been accepted towards fulfillment of the requirements for ’ Ph.D. Psychoiogy degree in IIIOWM Ma ajor professor Norman Abeles Date 12/16/87 MS U is an Affirmative Action/Equal Opportunity lutitim'ou 0-12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or betore dete due. DATE DUE DATE DUE DATE DUE I'-'. 73 a ‘4\7r-‘ ""e¢'" 001 MSU Is An Affirmative Action/Equal Opportunity Institution ——-———— I 'he-lu. SOCIAL SUPPORT AND ATTENTION TO SOCIAL CUES IN TYPE A INDIVIDUALS BY Lance Edward Schumacher AN ABSTRACT OF A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology 1 988 ABSTRACT SOCIAL SUPPORT AND ATTENTION TO SOCIAL CUES IN TYPE A INDIVIDUALS BY Lance Edward Schumacher The Type A behavior pattern (TABP) is associated with an increased risk for coronary heart disease. This study examined how Type A individuals differ from Type B individuals in their attention to social stimuli. The relationship between TABP and social support and the relationship between social support and attention to social stimuli was also investigated. A dual-task involving the discrimination of two social stimuli (angry male face, angry female face) and neutral letters was performed by 48 undergraduate students (24 males, 24 females). Signal detection theory methods were used to distinguish between sensitivity to the angry faces and willingness to report angry faces. The results revealed that Type As and Type 83 were not different in sensitivity to primary focus male or female angry faces. Type As were less sensitive to the secondary focus male angry face than Type 83; however, there was no difference between Type As and Type 83 in sensitivity to the secondary focus female angry face. Type As exhibited a bias against reporting the primary focus male angry face but not the primary focus female face. Type As and Type 83 did not differ in response bias for the secondary focus angry faces. Type As reported receiving a greater amount of social support than Type 85. Type As also reported larger available unconf licted support networks. Type As and Type Bs did not differ in perceived quality of support or size of conflicted support networks. There was a significant negative correlation between total amount of support received and sensitivity to the secondary focus male angry face; the same correlation involving the female angry face was not significant. Willingness to report angry faces was unrelated to social support. The differences in sensitivity and response bias for the male and female angry faces were interpreted as the result of the difference in the difficulty in detecting the two faces. The results suggest that Type As inhibit their attention to peripheral cues of anger and are unwilling to acknowledge perceiving central cues of anger in order to perform well on a primary task. Recommendations TOI‘ future research are OISCUSSBO. ACKNOWLEDGMENTS This research could not have been completed without the cooperation of many individuals. The author wishes to express his sincere appreciation to all those persons who made the completion of this study possible. A special thank you is extended to Dr. Lawrence VanEgeren, who provided the initial inspiration for the research and displayed an extraordinary amount of patience and understanding during the long process of completing the study. Thanks are also extended to Dr. Thomas Carr, who provided equipment and laboratory space in addition to encouraging support, to Dr. James Zacks for his assistance with the experimental design, and to Dr. Robert Caldwell, whose seminar on stress and coping piqued my curiousity on the relationship between Type A behavior and social support. An additional thank you is expressed to Dr. Norman Abeies for his understanding and guidance throughout the course of my doctoral studies and for agreeing to chair my doctoral dissertation. His support is greatly appreciated. iv TABLE OF CONTENTS V LIST OF FIGURES ................................................ viii CHAPTER 1. iNTRODUCTION ............................................ 1 Type A Behavior and Attention ........................... 1 Attention to Social and Health Cues ........................ 6 Type A Behavior and Social Support ....................... 7 Conceptualizing Social Support ........................... 9 Theory of Signal Detection ............................... l 1 Present Study .......................................... i2 2. HYPOTHESES .............................................. 19 3. METHOD .................................................. 22 Subjects .............................................. 22 Instruments ........................................... 22 Experimental Stimuli ................................... 26 Experimental Procedure ................................. 28 Analysis .............................................. 3i 4. RESULTS ................................................. 32 Tests of Hypotheses .................................... 34 Perception .......................................... 34 Social Support ...................................... 38 Perception and Social Support ........................ 4O vi Additional Findings ..................................... 4i Sensitivity: Angry Faces and Neutral Letter ............ 4i Recognition Criteria: Angry Faces and Neutral Letter. . . . 42 Hit Rates: Angry Faces and Neutral Letter .............. 44 5. DISCUSSION ............................................... 45 Perception ............................................. 45 Social Support ......................................... 48 Perception and Social Support ........................... 50 Recommendations ...................................... S I REFERENCES .................................................... S4 APPENDICES A. PERSONAL HEALTH INFORMATION ............................. 64 B. JENKINS ACTIVITY SURVEY .................................. 65 C. INVENTORY OF SOCIALLY SUPPORTIVE BEHAVIORS (ISSB) ........ 7i D. ARIZONA SOCIAL SUPPORT INTERVIEW SCHEDULE ............... 73 E. EXPERIMENTAL INSTRUCTICNS ................................ 79 F. CONSENT FORM ............................................. 84 G. TABLES I. FORCED-CHOICE MEANS, STANDARD DEVIATIONS, AND E SCORES ............................................. 85 II. DUAL-TASK a; MEANS, STANDARD DEVIATIONS, AND E SCORES ............................................. 85 III. DUAL-TASK B MEANS, STANDARD DEVIATIONS, AND E SCORES ............................................. 86 IV. ISSB MEANS, STANDARD DEVIATIONS, AND E SCORES ....... 86 V. ASSIS MEANS, STANDARD DEVIATIONS, AND E SCORES ...... 87 VI. CORRELATIONS BETWEEN ISSB AND SECONDARY d; .......... 88 vii VI I. CORRELATIONS BETWEEN ASSIS AND SECONDARY d: ......... 88 VIII. CORRELATIONS BETWEEN ISSB AND PRIMARY B ............. 90 IX. CORRELATIONS BETWEEN ASSIS AND PRIMARY B ............ 90 X. CORRELATIONS BETWEEN ISSB AND SECONDARY B ........... 9I XI. CORRELATIONS BETWEEN ASSIS AND SECONDARY B .......... 92 Figure I. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Figure 9. Figure IO. Figure l I. Figure l2. viii LIST OF FIGURES Probability density functions of signal plus noise and no signal ............................................. 14 Selected Ekman & Friesen (I976) faces as adapted for use in experiment, labeled according to intended affect .. . . . 27 Mean sensitivity (£11) for all subjects .................... 33 Mean recognition criteria (8) for all subjects ............ 33 Difference between Type As and Type BS in sensitivity (d) -to the neutral letter ............................... 35 Difference between Type As and Type 85 in sensitivity ($11) to the male face ................................... 35 Difference between Type As and Type 83 in sensitivity (£11) to the female face ................................. 35 Difference between Type As and Type 83 in recognition criteria (11) for the male face . .‘ ........................ 35 Amount of social support received by Type As and Type Bs .............................................. 38 Available social network size for Type As and Type 83 . . . . 38 Enacted social network size for Type As and Type 83 ...... 39 Satisfaction with social support for Type As and Type 85 .............................................. 39 ix Figure l3. Difference in sensitivity (d) between male face and letter as a function of behavior type and attentional focus ................................................ 43 Figure 14. Difference in recognition criteria (11) between male face and letter as a function of behavior type ............ 43 Figure IS. Difference in recognition criteria (13) between male face and letter as a function of behavior type and attentional focus .................... ' ................. 43 Chapter I INTRODUCTION A strong association has been shown to exist between coronary heart - disease (CHD) and the impatient, competitive, and aggressive behavioral style known as the Type A behavior pattern (Dembroski, MacDougall, Herd, & Shields, 1983; Dembroski, Weiss, Shields, Haynes, & Feinleib, I978,- Krantz, Glass, Schaeffer, 8: Davia, I982). Research on Type A has taken three major directions. One has been to examine the physiological correlates of the behavior pattern in order to understand the physiological mechanisms that may mediate between the behavior pattern and CH0 (Herd, I978). A second has been to study the environmental conditions under which Type A individuals experience exaggerated physiological activity (e.g., Dembroski, MacDougall, & Shields, I977; Manuck, Craft, & Gold, I978,- Van Egeren, I979). A third has been to focus on the cognitive functioning of Type A individuals, on the assumption that specific cognitive processes may underlie the behavioral and physiological responses associated with the Type A behavior pattern (Jennings, I983). The latter approach provided the impetus for the present research. What: Type As have been described as extremely alert and hypervigilant but inattentive to task-irrelevant inf ormation (e.g., Bortner & Rosenman, I967). This description led Weidner and Matthews ( I978) to suggest that, I 2 in the face of environmental stresses, Type As tend to focus their attention on doing well on the task at hand. As a result, they do not notice peripheral stimuli that are not immediately relevantto task performance, such as physical symptoms or subtle social cues. Matthews and Brunson ( I979) conducted a series of laboratory experiments to test this hypothesis. The first study examined Type As’ and Type 83' performances on two concurrent tasks, in which one was defined as central or primary and one was defined as peripheral or secondary. Assumptions of the dual-task paradigm are that individuals have a limited attentional capacity and that relative performance on the two tasks varies according to the amount of attention allocated to them, provided the individual's attentional capacity is fully utilized. Focusing of attention is reflected by deterioration of performance on the secondary task, and enhanced performance on the primary task (e.g., Bahrick, F itts, & Rankin, I952; Hockey, l970). The primary task in the Matthews and Brunson research was the Stroop Color-Naming Task and the secondary task was to depress a telegraph key upon the onset of a light. In this experiment Type As performed poorer on the secondary task and better on the primary task, compared to Type Bs. In another experiment distracting noise was combined with the Stroop task. Noise has been previously shown to improve performance on the color-naming task (Hartley 8: Adams, I974). This apparently occurs because subjects exposed to noise actively suppress their attention to the noise; in doing so, they also suppress their attention to other task-irrelevant cues, including the name of the color of the stimulus word (Houston, l969). If Type As attend less to task-irrelevant peripheral stimuli than Type 85 do simply because they are attending to the central task, then the Stroop performance of As should not be affected by the 3 presence of a distractor. However, if As actively suppress their attention to distractors, the performance of As should be facilitated by the presence of a distractor. In fact, the performance of Type As improved in the presence of the distracting noise, whereas the performance of Type 83 was unaffected by the distracting noise. The findings of Matthews and Brunson (l979) suggest that Type As allocate more attention to central events and less to peripheral events than do Type 83. Moreover, Type As seem to actively suppress their attention to peripheral events that might distract them from performing well on a task. Subsequent research has further supported the contention that Type As focus more attention on central events and less attention on peripheral events than do Type Bs. Stern, Harris, and Elverum (l98l ) compared the performance of Type As and Type 83 on arithmetic tasks with an embedded recall task and vice versa (arithmetic embedded in a recall task). In both conditions, Type As performed better than Type Bs on the task defined as primary and worse than Type Bs on the task defined as secondary. Humphries, Carver, and Neumann (I983) compared Type As and Type 83 in a study of category formation. With moderate and high but not low incentive for performances, Type As formed categories characterized by relatively restrictive def intions. The investigators suggest that Type As focus on the central tendencies in an evolving category and correspondingly ignore less frequently occurring attributes. Type 83, who presumably are less likely to ignore peripheral attributes, formed categories that had broader definitions, encompassing both frequent and infrequent attributes. Strube, Turner, and Perrillo (I983) examined the effects of pleasant background music on task performance of Type As and Type 83. Simple melodies have been shown to have a soothing effect, which enhances performance. Type 4 83 performed better when background music was present. In contrast, performance of Type As was unaffected by the presence of background music, presumably because the As suppressed paying attention to it. Taken together, these studies support the notion that Type As and Type Bs differ in how they allocate attention. Specifically, Type As focus their attention more on the task at hand and attend less to peripheral events than Type 8s do. Differences in how Type As and Type 83 allocate attention may be explained in terms of Easterbrook's (I959) hypothesis that arousal narrows the span of attention, leading to a decrease in the range of cue utilization. In other words, increases in arousal will presumably result in the ignoring of task-irrelevant or incidental cues. Despite the psychophysiological complexity of arousal (Humphries & Revelle, I984), experimental conditions considered to be “arousing“ do seem to increase attention to central events and decrease sensitivity of peripheral events (e.g., Hockey, i979). Evidence shows that Type As have greater increases in sympathetic nervous system arousal than Type 8s do while performing challenging tasks, although no differences have been reported in resting levels (Dembroski, et al., I977; Manuck, et al., I978). It may be argued that in earlier cited dual-task experiments (e.g., Matthews & Brunson, i979; Stern et al., l98| ), Type As were more aroused than Type Bs as a result of the challenging experimental conditions. The increase in arousal may have led to a decrease in the range of cue utilization among Type As which, in turn, impaired their performance on peripheral tasks. Thus it is possible that differences in arousal underlie the differences in Type As' and Type 83' strategies of attention allocation. Cohen (I978) suggests that attentional focusing is an adaptive 5 response to information overload under conditions of environmental stress. This interpretation is based on a limited processing capacity model of attention (Kahneman, I973). According to this model, an individual has a limited amount of attention that can be allocated at any one time. Information overload is said to occur when this attentional capacity is exceeded, for instance, when there are too many inputs to be processed or when successive inputs come too fast. Attention can be divided across several information sources depending on how much capacity each source requires. During a task, an individual must allocate attention among competing information sources including task-relevant cues and task-irrelevant cues (e.g., negative affect, symptoms of fatigue, or background noise). When an individual allocates more attention to one source, the allocation of attention to other sources must be reduced because the total capacity is limited. As indicated earlier, the limited processing capacity model serves as the conceptual basis for dual-task experiments (e.g., Cohen & Lezak, I977; Matthews & Brunson, I979; Stern et al., I98 I ). It also provides a framework for understanding Type A and Type B differences in attention allocation. Jenkins (I975) described Type As as individuals who become more upset than Type Bs by excessive task demands. Therefore, a dual-task situation might be expected to produce greater stress responses in Type As than Type 83. Environmental stressors such as noise (Finkleman & Glass, I970), crowding (Saegert, Mackintosh, & West, I975) and task overload (Kanarick & Petersen, I969) reportedly result in a focusing of attention on aspects of the environment most relevant to task performance at the expense of less relevant cues. The tendency for Type As more than Type ES to ignore peripheral cues while focusing their attention on central 6 events may therefore be due to the greater stress responses in Type As as a result of the dual-task situation. W The tendency of Type As to inhibit their attention to distracting, peripheral events should affect the likelihood of their noting and acting upon social and health cues. Clinical as well as experimental observations concur that Type A individuals are less likely to report health relevant cues than Type 8 individuals. For instance, Type As report less severe physical symptoms than Type 83 do while performing a stressful task (Weidner & Matthews, I978). The selective attention of Type A individuals may lead them to ignore early symptoms of CHD and not seek medical attention. This notion is consistent with the observations by Greene, Moss, and Goldstein (I974) that myocardial infarction patients, who more often than not are Type As, often ignore symptoms and delay seeking medical care. - ‘ Another health implication of selective attention is the ignoring of subtle, but important, social cues. Social cues are events, such as facial expressions, which provide information necessary for the maintenance of adequate relationships. Important social cues include information about the emotions and subtly expressed needs of others. The neglect of such cues may decrease the likelihood of helping another, expressing concern for another, or responding appropriately to another's needs, which, in turn, may impair one's social relationships (Berkowitz, I972). The effect of restricted attention on awareness of social cues has been demonstrated by Cohen and Lezak ( I977). Subjects were asked to memorize nonsense syllables with a secondary task of noting slides of persons engaged In an activity. When noise was added to the task, memory for the incidental social information decreased. Noise also causes 7 decreases in helping as reported by Matthews and Canon (I975) and reduces the amount of information subjects use in forming personality judgments as reported by Siegal and Steele (I979). These studies show a reduction in awareness of social cues under conditions (e.g., noise) designed to exceed attentional capacity (Cohen, I978). WW Research has determined that the support provided by social relationships can promote health (Cobb, I976), speed recovery from illness (Berle, Pinsky, Wolf, & Wolf, I952), and extend life (Weisman & Worden, I975). Lack of social support has been associated with chronic disease (Wolf, I975), distress (Caplan, I97I ), and higher mortality (Berkman & Syme, I979). Bruhn (Bruhn, I980; Bruhn, McCrady, & duPlessis, I968) found that individuals experiencing conflict in their marriage and at work and difficulty communicating effectively with others were more likely to develop CHD than individuals not reporting these characteristics. Theil, Parker, and Bruce ( I973) reported a significantly higher frequency of divorce among men with myocardial infarction. Weiss (I973) found significant differences in CHD mortality rates between married and never-married, divorced, and widowed subjects, despite no significant differences in serum cholesterol levels and blood pressure. Price (I982) has suggested that the hard-driving, competitive, and hostile behavior of the Type A individual may lead to conflict within social relationships, resulting in poor quality social support systems and a lack of close, intimate friendships. There is some evidence that Type As have inadequate social support networks. Type As have been found to devote less time to Interpersonal activities (Becker & Byme, I984), to devalue interpersonal relationships (Suls, Becker, & Mullen, I98 I ), to be 8 uncomfortable in social settings (Jenkins, Zyzanski, & Ryan, I977), and to become nervous, distracted, and impatient when working with others (Gastorf, Suls, & Sanders, I980). Type As obtain lower scores on an index of social support, indicating that they have fewer and less satisfactory social relationships (Waldron et al., I980). This finding was partially confirmed by General (I983), who reported low perceived quality of social support among Type A coronary patients. Type A coronary patients are also likely to view their families as less cohesive and supportive and report fewer contacts with friends and relatives (Black, I980). Margolis, McLeroy, Runyan, and Kaplan (I983) suggested that Type As are more likely to define their intimate relationships as competitive. This was supported in a study by Orth-Gomer (I979), who found that a large percentage of Type As who have had heart - _ attacks report having a predominantly adversarial relationship with their spouses. Burke, Weir, and DuWors (I979) report low levels of marital satisfaction among the wives of Type A men. Their findings suggest that Type A husbands may be less likely than Type B husbands to confide in their wives. Support for this Interpretation comes from Jennings and Pilkonis (cited in Jennings, I983), who found that half of Type As in their study reported no confidant or only a single confidant, while Type 83 all reported two or more confldants. These findings indicate that the quality as well as quantity of social support available to Type As is less than that available to Type Bs. Type As may have less adequate social support systems than Type 85 because Type As place relatively little importance on social relationships. Type As generally place high value on achieving and derive satisfaction primarily from activities that produce tangible accomplishment (Price, 9 I982). This reliance is often at the expense of personal relationships, since neither family nor friends are generally seen as contributing to material success. Price (I982) argues that personal relationships therefore merit less attention from Type A individuals. Type As may also have less social support because of their attitudinal set reflecting hostility. Type As tend to view others as bad, selfish, and exploitative (Williams et al., I980). They are inclined to project excessively high standards onto others and feel justified in acting highly critical and intolerant of others (Price, l982). Price (I982) also points out that the hostile behavior of Type As tends to elicit hostile and aggressive reactions from others. Type As, without realizing that they have created a great deal of the hostile reactions they perceive in others, may come to believe that they live in a hostile world. As a result of this belief, Type As, compared to Type Bs, may be more likely to misperceive hostility in others. That is, Type As may be more inclined to report that others are angry when this is not the case. This perceptual set toward hostility may, in turn, have a negative impact on Type As' social relationships; Type As may find that they have less social support as a result of their exaggerated tendency to 'see“ anger in others. WW Researchers disagree on what constitutes social support (Barrera & Ainlay, I983; Thoits, l982). As Kahn and Antonucci (I980) point out, 'social support is one of those terms that carries considerable colloquial meaning and has therefore been more often used than def ined" (p. 392). This lack of agreement concerning the conceptualization of social support has led to inconsistencies and inadequacies in measurement. Analyses of the social support literature (Kaplan, Cassel, 8. Gore, I977; Mitchell 8. I0 Trickett, l980; Thoits, I982; Wilcox, l98l) indicates that social support has both quantitative and qualitative dimensions. The quantitative dimension of social support describes the structure of individuals' social networks. Thus, some researchers have derived social support from specific demographic information, such as marital status (Eaton, I978), access to a confidant (Brown, Bhrolchain, & Harris, I975), and participation in community organizations (Lin, Simeone, Ensel, & Kuo, l979), while others have used indices such as numbers of sources and frequency of social contacts (Berkman & Syme, I979). Such structural measures often assume that the availability of a large social network or frequent social contacts are equivalent to a high degree of support. These are questionable assumptions since available support does not necessarily translate into enacted support, and not all social contacts are necessarily supportive in nature (Wellman, l98l ). According to Cobb (I976), social support Is essentially information that one is loved, esteemed, and part of a network of mutual obligation. He regards the cognitive appraisal of support as primarily important in the measurement of social support. Other researchers interested in individuals' subjective evaluation of support have looked at satisfaction with support (Hirsch, I979), happiness with key relationships (N0ckolls, Cassel, & Kaplan, I972), and adequacy of social attachments (Henderson et al., l978). Schaefer, Coyne, and Lazarus (I98I ) have emphasized the importance of measuring both the quality of support as perceived by the individual, and the quantity of support as measured in terms of network size and frequency of contacts, since different indices may relate differently to health. Wilcox (l98l ), for example, in one of the few studies which compared the qualitative and quantitative dimensions of social I I support, found that emotional support was more predictive of mental health than social network figures. The social network represents what is available to the individual from an objective viewpoint, and emotional support represents the way in which the individual perceives and feels about his or her social support system (Kaplan, Cassel, & Gore, I977). The majority of studies of social support have examined either the qualitative or the quantitative aspects of social support but have not combined them into a comprehensive assessment (Bruhn & Phillips, I984). Some recently developed instruments, such as the Arizona Social Support Interview Schedule (ASSIS), (Barrera, l 98 I ), incorporate a multidimensional approach to assessing social support. In addition to describing social networks, the ASSIS measures available and enacted social support and satisfaction with support. The ASSIS also makes a distinction between strictly supportivenetwork members and those who are sources of interpersonal conflict; thus, it is possible to calculate conflicted as well as unconf licted network size. WWII Research on perceptual differences between Type As and Type Bs has presupposed that reported detection of stimuli depends solely on sensitivity, which is a joint function of individual ability and stimulus characteristics. This assumption is false. An observer's response bias affects the obtained value of a sensory threshold (Price, l966). Response bias is a function of motivational, attitudinal and strategic factors; it influences an observer's subjective standard or criterion for reporting the presence or absence of a stimulus. Conventional methods of studying perception, which have used the classical threshold as an indicator of sensitivity, are inadequate because they confound sensitivity and response I2 bias. Hake and Rodwan (I966) point out that reluctance to report the presence of a stimulus until the evidence is great results in few hits (true positives) and tends to raise the traditional estimate of the sensory threshold. On the other hand, willingness to report the presence of a stimulus on the basis of minimal evidence results in an increased number of hits and tends to lower the estimate of the. threshold. The apparent changes in sensitivity stem from the fact that false alarms (false positives) are not adequately assessed by traditional threshold measurement procedures. Both hit rate and false alarm rate must be accounted for in any adequate description of an observer's perceptual performance (Price, I966). In studies (which used traditional psychophysical methods) indicating that Type As are less attentive than Type Bs to peripheral stimuli (e.g., ‘ Matthews & Brunson, I979; Stern et al., I98I ), it may be that Type As adopt stricter criteria for reporting the presence of peripheral stimuli and, as a result, only appear to be less sensitive to the stimulus. In studies of personality and perception, a sub ject's criterion and its attitudinal and motivational determinants are of interest as well as his perceptual sensitivity. To handle the problem of independently assessing these two elements of perceptual performance a model known as the theory of signal detection (Green & Swets, I966) has been developed. In the general signal detection problem, an observation is made of events within a fixed interval of time, and a decision is made, based on this observation, whether the interval contained only background interference or a signal as well. The interference is referred to as noise (N), and the alternative Is termed signal plus noise (SN). It is assumed that noise is l3 constant, whereas the signal may or may not be present during a specified interval. Thus, the signal is always observed in a background of noise. Noise is inherent in sensory processes, but it also may be introduced by the external (i.e., experimental) situation. Noise may be any stimulus not designated as a signal, but which may be confused with it. The term observation refers to the sensory datum on which a decision is based. It is assumed that an observation may vary continuously along a single dimension. A given observation is further assumed to arise, with specific a priori probabilities, either from noise alone or from signal plus noise. These assumptions are portrayed graphically in Figure I. The observation is labeled x and is plotted on the abscissa. Repeated presentations of N and SN form overlapping normal distributions, labeled F n(x) and an(x), respectively. In Figure I, Fn(x) represents the probability density that x will result given the occurrence of noise alone, and F sn(x) represents the probability density that x will result given the occurrence of signal plus noise. The distributions overlap to the extent there is confusion between noise alone and signal plus noise. During a given interval an observer is required to make a decision indicating whether an observation is more representative of the signal plus noise distribution or the noise distribution. In effect, the observer is testing statistical hypotheses and taking risks associated with Type I and Type II errors. The theory of signal detection assumes that subjects make decisions about stimulus events by comparing any given observation with some subjective standard. Subjects are assumed to adopt a cutoff point (the criterion) on the common dimension of observations associated with noise and signal plus noise. An observation that matches or exceeds the criterion will be considered more representative of signal plus noise than noise I4 Probability Density Fn (x) F sn (x) I I I I I I I n Observation ()5) Figure I. Probability density functions of signal plus noise (SD) and no signal (h). After Swets, Tanner, and Birdsall, I96 I. IS alone and the response will be “yes.“ Conversely, an observation that is less than the criterion will be considered more representative of noise alone than signal plus noise and the response will be ”no.” The particular criterion adopted by an observer depends on the consequences associated with correct and incorrect responses as well as the a priori probabilities of signal plus noise and noise alone occurring within a specified interval. In Figure I, the area under the signal plus noise curve to the right of the criterion (8) corresponds to the hit rate (HR), and the area under the noise curve to the right of 8 corresponds to the false alarm rate (FAR). If a subject establishes a cutoff point far to the right, few hits and few false alarms will be obtained; the subject has adopted a “strict” criterion for reporting SN. If a subject establishes a cutoff point far to the left, both HR and FAR will be increased; the subject has adopted a "lax” criterion for reporting SN. The accuracy of an observer's decision is a function of sensitivity (d’). The sensitivity measure d: is an indication of the extent to which the signal plus noise and noise distributions overlap. The value of d; varies directly with hit rate and inversely with false alarm rate. If hit rate and false alarm rate are equivalent, then d; is equal to zero; this indicates an observer is responding at a chance level of accuracy. Hit rates greater than the false alarm rate result in values of d: greater than zero (accuracy above chance); hit rates less than the false alarm rate result in values of d“. less than zero (accuracy below chance). Values of d; for any combination of hit and false alarm rates can be obtained from a table developed by Elliott (I964). Like the measure of sensitivity, the value of an observer's criterion (11) can be calculated from hit and false alarm rates. Values of B can be I6 obtained from a table developed by Gardner, Dalsing, Reyes, and Brake (I984). An observer who is maximizing hits while minimizing false alarms will have a I} that Is equal to I00. A value of 11 below I.00 represents a lax criterion, such that the observer will be liberal about reporting signals; a value of 11 above I.00 represents a strict criterion, such that the observer will be conservative about reporting signals. By distinguishing between sensitivity and criterion, the theory of signal detection may clarify differences between Type As and Type 83 in the perception of central and peripheral stimuli. Given the evidence that Type As focus their attention more on the task at hand and attend less to peripheral events than Type 83 do (e.g., Matthews 8: Brunson, I979), Type As should be more sensitive than Type Bs in recognizing stimuli when this task is identified as the primary task in a dual-task experiment. However, Type As should be less sensitive than Type 83 when this task is identified as the secondary task. Differences in response bias between Type As and Type 83 have been unexplored in previous research. It may be of interest to know whether Type As and Type 83 adopt different criteria for reporting the presence of certain social stimuli. As stated earlier, as a result of their own hostility, Type As may come to believe that they live in a hostile world, and this belief may cause them to misperceive hostility in others. Type As, therefore, may be more inclined to report hostility in others on the basis of little evidence. To the degree this is so, Type As will adopt lax criteria when they are required to decide whether or not a given facial expression is angry; they will be lenient about reporting the presence of an angry OXDI‘BSSIOI'). Type As' lax criteria in the detection of angry expressions may be I? viewed, in part, as a function of the perceived benefits of hits balanced against the costs of false alarms. It is conceivable that, for Type As, the benefit of correctly identifying a true angry expression (a hit) is given more weight than the possibility of incorrectly identifying a nonangry expression as angry (a false alarm). This may be explained by the aggressive attitude of Type As and their general belief that "to be forewarned is to be f creamed” in the competitive battle with others for valued achievements. Thus, it may be more important to Type As to correctly identify an angry foe than to misidentify a friend as a foe. Ecesentfiludy The present study was designed to clarify the differences between Type As and Type Bs in the perception of social cues through the use of signal detection theory (Green 8: Swets, l966). Levels of sensitivity and decision criteria were assessed for Type As and Type Bs on a dual-task involving discrimination between angry and neutral faces and discrimination between neutral letters. If Type As are less attentive than Type 83 to personal relationships, then Type As might be expected to show lower sensitivity to angry faces when recognizing the faces is subordinate to some other task (e.g., identifying a neutral letter). The notion that Type As are predisposed to perceive hositility in others was tested by comparing Type As' and Type 83' decision criteria for reporting that a face was angry rather than neutral. The present study was also designed to extend the previous findings of inadequate social support among Type As with a comprehensive assessment of social support. In addition, sensitivity to secondary angry faces was correlated with social support, based on the assumption that attention to peripheral social cues may have a positive effect on social l8 relationships. Finally, decision criteria for reporting angry faces were correlated with social support, based on the assumption that greater willingness to report that a face is angry reflects a predisposition to perceive hostility In others, which may have a negative effect on social relationships. Chapter 2 HYPOTHESES . W I. Type As will be more sensitive in discriminating between two neutral letters than Type Bs when recognition of the letters is the primary task. ‘ 2. Type As will be more sensitive in discriminating between an angry face and a neutral face than Type 83 when recognition of the faces is the primary task. Hypotheses I and 2 follow from evidence that Type As focus more attention on a primary task than Type 88 (e.g., Matthews 8: Brunson, I979). I 3. Type As will be less sensitive in discriminating between two neutral letters than Type 83 when recognition of the letters is the secondary task. 4. Type As will be less sensitive in discriminating between an angry face and a neutral face than Type Bs when recognition of the faces is the secondary task. Hypotheses 3 and 4 follow from evidence that Type As focus less attention on a secondary task than Type Bs (e.g., Matthews 8: Brunson, I979). 5. Type As will not differ from Type 83 in their criteria for reporting one neutral letter rather than another neutral letter when recognition of the letters is the primary task. I9 20 6. Type As will not differ from Type Bs in their criteria for reporting one neutral letter rather than another neutral letter when recognition of the letters is the secondary task. 7. Type As will be more lenient in reporting that a face is angry rather than neutral than Type 8s when recognition of the faces is the primary task. 8. Type As will be more lenient in reporting that a face is angry rather than neutral than Type 83 when recognition of the faces is the secondary task. Hypotheses 7 and 8 are based on the assumption that Type As are predisposed to report anger in others on the basis of little evidence. II. W 9. Type As will report less social support than Type Bs. I0. Type As will report less unconflicted social support than Type II. I2. 8s. Hypotheses 9 and ID are based on research reporting that Type As have fewer social relationships (e.g., Jennings, I983; Waldron et al., I980). Type As will report more conflicted social support than Type 8s. The social relationships of Type As are characterized by aggression and conflict (e.g., Orth-Gomer, I979; Price, l982). Type As will report less satisfaction with social support than Type Bs. Type As report less satisfactory social relationships (e.g., Black, I980; General, I983; Waldron et al, I980). 2I III. WW I3. I4. IS. Sensitivity in discriminating between secondary angry and neutral faces will be positively correlated with social support. Inattention to incidental social cues may impair social relationships (Berkowitz, I972). Criteria for reporting that a primary focus face is angry rather than neutral will be positively correlated with social support. Criteria for reporting that a secondary focus face is angry . rather than neutral will be positively correlated with social support. Hypotheses I4 and IS assume that greater willingness to report angry faces reflects the predisposition to “see” hostility in others, which, in turn, may impair social relationships. Note that more strict criteria will be indicated by larger values of 8, hence the expected positive correlation. Chapter 3 METHOD Subjects _ Subjects were 24 male and 24 female undergraduate students recruited from introductory psychology classes at Michigan State University. Subjects were offered either five dollars or research credit in return for their participation in this study. The 48 experimental subjects were chosen from a pool of approximately two hundred subjects who were screened for Type A behavior, using the student version of the Jenkins Activity Survey (JAS) (Krantz, Glass, 8: Snyder, l974). Separate distributions of Type A behavior scores were formed for males and females, and experimental subjects were selected from the top and bottom 25% of these distributions. During screening sessions subjects also completed the Inventory of Socially Supportive Behaviors (ISSB) (Barrera, I98l; Barrera et al., I98 I ), and a brief health questionnaire, which included a question about vision. Only subjects reporting vision corrected to 20/20 were included in the experiment. The Arizona Social Support Interview Schedule (ASSIS) (Barrera, I98I) was administered at the end of the laboratory experiment. Insmlmems W The Jenkins Activity Survey (JAS) (Jenkins et al., I97I) is a self -report measure that was used to assess degree of 22 23 Type A characteristics. This scale contains questions that are similar to those used in a standardized stress interview (Structured Interview) developed by Friedman and Rosenman to assess Type A behavior (Rosenman, I 978). Scoring of the items on the JAS is based on a series of optimal weights derived from discriminant function analyses predicting the Structured Interview classification of large groups of men enrolled in the Western Collaborative Group Study (Jenkins et al., I97 I ). The JAS has been shown to agree with the Structured Interview assessment 73% of the time, and with the assessment of extreme A's and 8'5 90% of the time (Jenkins et al., I97I ). The test-retest reliability of the JAS has ranged between .60 and .70 across I-4 year intervals (Jenkins, I978). Validity for the JAS as a predictor of coronary heart disease has been well-established through prospective and retrospective studies on the risk factors in the etiology of CHD (Jenkins, Rosenman, 8: Zyzanski, I974; Jenkins et al., I97l ; Kenigsberg, Zyzanski, Jenkins, Wardell, 8: Licciardello, I974; Shekelle, Schoenberger, 8: Stamler, I976). Because the original JAS was developed for use with working adult males, it is not appropriate for male and female college students. A modified, student version of the JAS (Form T) was consequently developed by Krantz, Glass, and Snyder ( l 974). It is essentially the same as the original version, except that items concerning income, job involvement, and job responsibility have been deleted. In large samples of undergraduate students the median of the JAS falls between 7 and 8 (Glass, I977). In the present research, Type As were those who scored well above the median (I9 and above), whereas Type 85 were those who scored below the median (7 and below). MW The Inventory of Socially 24 Supportive Behaviors (ISSB) (Barrera, I98I; Barrera et al., I98 I) is a self -report measure that was used to assess type and amount of social support. The questionnaire consists of 40 statements describing the enactment of various types of support during the preceding four weeks. Ratings are made on a 5-point Likert-type scale that ranges from "not at all" to ”about every day.“ The ISSB has shown a high degree of internal consistency and test-retest reliability as indicated by a coefficient alpha of .93 and a test-retest correlation of .88. The measure also showed a modest, but significant correlation with an index of perceived supportiveness (the Moos Family Environment Scale) (Barrera et al, I98I ). The large coefficient alpha reported by Barrera et al. 098” as well as the results of a factor analysis by Stokes and Wilson ( I984) suggest that. it is appropriate to use the ISSB as a global measure of social support. However, four interpretable components emerged from Stokes and Wilson's (I984) analysis that are consistent with the types of social support that have been cited in the literature (e.g., Mitchell 8: Trickett, I980). Stokes and Wilson (I984) describe the four principle components as follows: emotional support or intimate interaction; tangible assistance and material aid; cognitive information, feedback, and clarification; guidance with a parental or directive quality. Coefficient alphas computed for each component were .85, .7I, .83, and .77, respectively. The variance accounted for by each of the four components in Stokes and Wilson's (I984) study is as follows: emotional support, l8.46%; tangible assistance, I2.30%; cognitive information, I8.65%; guidance, I4.29%. In the present study, scores pertaining to these four principle components along with total ISSB scores were derived for each subject. 25 WW1: The Arizona Social Support Interview Schedule (ASSIS) (Barrera, I98I) is a structured interview designed to measure support satisfaction, type of support, and source of support. Six different types of social support are assessed: intimate interaction, material aid, guidance, positive feedback, physical assistance, and social participation. These categories appear to capture the range of activities that a number of social support researchers (e.g., Caplan, I976; Hirsch, I980) have described in their discussions of social support. The ASSIS assesses actual utilization of support resources in addition to availability of support. It is particularly well-suited for this study as it distinguishes between strictly supportive network members and those who are sources of conflict or unpleasant interactions. Thus, differences between conflicted and unconflicted network size may be assessed. The ASSIS is also unique in that it measures satisfaction with social support. Barrera ( l98l) reported that the test-retest reliability for total network size was .88. Seventy-f our percent of the sources named appeared on both administrations of the interview. Conflicted network size had a significant but much lower test-retest correlation coefficient of .54. The support satisfaction measure was positively skewed toward high satisfaction scores. The test-retest reliability for support satisfaction was fairly low (correlation coefficient - .69), while the internal consistency was much lower (coefficient alpha - .33), indicating that this measure varies considerably internally but is more stable over time at the aggregate level. When the ASSIS was administered with the ISSB, the ISSB showed a modest, but significant correlation with total network size. In addition, 26 the support satisfaction measure of the ASSIS did not correlate significantly with the ISSB. Thus, the ISSB and the ASSIS appear to be measuring unique features of social support (Barrera, I98l ). E . l I Sl' II The experimental stimuli were two different sets of 35 mm slides. One set contained four photographs of the faces of two models (one male and one female). Each model projected an angry and neutral closed mouth facial expression (see Figure 2). The emotional expressions were chosen from a set of photographs developed and tested by Ekman and F riesen (I976). The photographs were modified to achieve greater uniformity. Modifications included removing identifying numbers and matching hairlines within male and female subsets. The faces were I2.6 cm in width and I8.2 cm in height, corresponding to 3.0 degrees and 4.3 degrees of horizontal and vertical subtense, respectively, when projected onto a screen 240 cm from the subject. Each face was presented in the center of an illuminated pre-exposure field 58.9 cm in width and 38.9 cm in height. A black dot, positioned in the center of the pre-exposure field, served as a fixation target. The second set of slides contained two letters (P and R). The letters were 2.2 cm in height and subtended an angle of 0.5 degrees. The near- and far-points from fixation equalled 3.7 degrees and 4.2 degrees, respectively. Stimuli were presented by means of a three field projection tachistoscope focused on a white screen. One projector presented faces and a second projector presented letters. A face and a letter were presented together during each trial. In half of the trials the letters were below the faces, and on the other half of the trials the letters were above the faces. There were an equal number of upper and lower Ps and Rs, as STIMULUS FACES N EUTRAL ANGRY NEUTRAL Figure 2. Selected Ekman 8: Friesen (1976) faces as adapted for use In experiment, labeled according to Intended affect. 28 well as an equal number of angry and neutral and male and female faces. Sixteen different combinations of the four faces and two letters were possible when letter position was introduced as a factor. ExperimentaLEcoceduce After being introduced to the experimental situation, subjects were read general instructions describing the experimental procedure. Following this, subjects previewed the experimental stimuli. Sixteen slides were presented for 3 seconds each. Subjects were instructed to identify the facial expression and the letter presented in each ‘of the slides. Immediate feedback was given following the preview of each of the sixteen slides. Mean accuracy levels were assessed for each expression and each letter. There were four experimental conditions as follows: (a) social cue detection task alone; (b) letter detection task alone; (c) dual-task condition with the social cue detection task as the primary task and the letter detection task as the secondary task; and (d) dual-task condition with the letter detection task as the primary task and the social cue detection task as the secondary task. The order of presentation of experimental conditions was counterbalanced by randomly assigning subjects to one of four task sequences: abcd, abdc, bacd, or badc. Prior to each condition, subjects read a set of instructions detailing the procedure for that condition. Subjects were instructed to f ixate a central point when the experimenter gave a verbal “ready” signal. Throughout the experiment subjects were reminded to f ixate the central point when given this 'ready' signal. Following the “ready“ signal a slide containing a face and a slide containing a letter were presented simultaneously. Onset of the stimuli was triggered manually by the experimenter. lntertrial intervals were approximately 5 seconds in each of 29 the four conditions. The tasks in conditions (a) and (b) were f orced-choice tasks, while the tasks in conditions (c) and (d) were yes-no tasks. In the two forced-choice conditions each trial consisted of two stimulus intervals, separated by approximately 2 seconds. Subjects were instructed to identify the interval (first or second) in which the target stimulus was presented. The target stimulus was either an angry face or the letter R, depending on the task. Subjects were told to ignore the letters during the social cue detection task and to ignore the faces during the letter detection task. In the two yes-no (i.e., dual-task) conditions each trial consisted of a single stimulus interval, during which a target face and a target letter were presented. Subjects were told to do their very best on the primary task, while also trying to do as well as they could on the secondary task. They were instructed to first report the primary task stimulus, followed by the secondary task stimulus. For example, if the social cue recognition task was primary and the letter recognition task was secondary, the response might have been “neutral, P.“ According to McNicol (I972), the f orced-choice task in signal detection theory provides a relatively pure measure of sensitivity that corresponds closely to the measure of sensitivity estimated by the yes-no task. Therefore, stimulus durations for each task in the dual-task conditions were individually determined for each subject, based on their performance on the corresponding f orced-choice tasks. Different stimulus durations were generally used for the two tasks in the dual-task conditions; this was possible since faces and letters were presented by means of separate slide projectors. Stimulus durations for the dual-task conditions corresponded to 30 approximately 79% correct responses for each task in the f orced-choice condition. These durations were determined by means of a transformed up-down procedure described by Levitt (I97 I ). According to this procedure, a stimulus having a high probability of a positive response is presented to the subject. Three consecutive correct responses are followed by a step decrease on the next trial. A single wrong response is followed by a step increase on the next trial. This up-down procedure is repeated for several runs in each direction. Levitt (I97l) recommends using mid-run averages to analyze the data obtained with this procedure. The peaks and valleys of several runs are averaged to obtain a robust estimate of the stimulus duration corresponding to 79% correct responses for each task. In the present experiment, each f orced-choice staircase involved I7 runs. Only the last I0 runs were used to compute a reliable stimulus duration for each subject. Since the length of each run varied between subjects, the number of trials also varied between subjects. Levitt ( I970) recommends starting with a large step size which is gradually decreased during the course of the experiment in order to maximize efficiency of estimation. The initial value in the forced-choice staircase was 290 msec, with subsequent exposures being x/ I.2, where x was the previous exposure. Steps decreased or increased by increments of one exposure level, except in the first and second runs, where every third exposure level was presented, and in the third run, where every other exposure level was presented. Each dual-task condition consisted of 2I6 trials. Out of I6 possible dual-task combinations of faces and letters, I2 combinations were presented I2 times and 4 combinations were presented I8 times. There was a one minute break after each block of 72 trials in all four conditions. 3i There also was a five minute break between conditions, during which instructions for the upcoming set of trials were given to subjects. Following the experiment, subjects were administered the ASSIS. Before leaving, subjects were told how they scored on the JAS, along with a verbal explanation of their score. Subjects also were debrief ed about the nature of the study. The total time required to complete the experiment, including the ASSIS, was approximately 3 hours. Analxsis w Using 11'. and B as dependent measures, 2 (Type A-Type B) X 2 (Male-Female) X 2 (Task Order) X 2 (Attentional Focus) analyses of variance were performed to test the hypotheses pertaining to sensitivity and decision criteria in the perception of angry faces and neutral letters. For each of the dual-task conditions, hit and false alarm rates were assessed for each subject in order to obtain d’ and 11(McNicol, I972). Values of d“ corresponding to obtained hit and false alarm rates were determined by using a table developed by Elliot (I964). A similar table developed by Gardner, Dalsing, Reyes, and Drake (l984) was used to determine values of 11 based on obtained hit and false alarm rates. Two (Type A-Type B) X 2 (Male-Female) analyses of variance were performed to test the hypotheses pertaining to social support. Correlation coefficients were computed to determine the relationship between sensitivity to social cues in the dual-task experiment and measures of social support. Correlation coefficients were also computed to assess the relationship between recognition criteria and measures of social support. Chapter 3 RESULTS A significant Stimulus effect on sensitivity was found when the male angry face, female angry face, and neutral letter were compared in the dual-task experiment, E (2,80) = 63.86, p < .OOI. Figure 3 shows that the male face was easier to discriminate than the letter, which, in turn, was easier to discriminate than the female face. The mean d: for the male angry face was I50, the mean i for the female angry face was .70, and the mean d'. for the neutral letter was .95. Post-hoc comparisons using the Tukey method revealed that each mean was significantly different from the others (p < .0l ). In addition, a comparison of recognition criteria for the male face and the female face revealed an overall significant difference, E (I,40) - 4.70, p - .036. Figure 4 shows that subjects were more conservative in reporting that the male face was angry (the larger the value, the more conservative the criterion). The mean 11 for the male angry face was I29 and the mean 8 for the female angry face was l.08. In view of the differences in overall sensitivity and recognition criteria, the male and female faces were treated separately in the following analyses. Because there were no significant effects for sex of subject or task order in any of the analyses, only analyses by type are reported. 32 33 Letter . 2 Female Face Male Face 2 d' Figure 3. Mean sensitivity (d') for all subjects. Letter Female Face Male Face 1.29 O I 2 D Figure 4. Mean recognition criteria (8) for all subjects. 34 135W Remenlmn Hypothesiit; Type As will be more sensitive in discriminating between two neutral letters than Type Bs when recognition of the letters is the primary task. Beams; Figure 5 shows the sensitivity of Type As and Type 85 to the neutral letter. The two groups did not differ in primary focus sensitivity to the letter, E ( l,40) Please read each item carefully and select the rating that you think is the most accurate. During the nasljnumeeks, how often did other people do these activities for you, to you, or with you: I. Looked after a family member when you were away. 2. Was right there with you (physically) In a stressful situation. 3. Provided you with a place where you could get away for awhile. 4. Watched after your possessions when you were away (pets, plants, home, apartment, etc. ). Told you what he/she did in a situation that was similar to yours. Did some activity together to help you get your mind off things. Talked with you about some interests of yours. Let you know that you did something well. Went with you to someone who could take action. . Told you that you are OK just the way you are. . Told you that he/she would keep things that you talk about private-- just between the two of you. l2. Assisted you in setting a goal for yourself. l3. Made it clear what was expected of you. --os0.°°.\|.0‘.0‘ .‘ _ 20. 2I. __ 26. 27. I4. 15. I6. 17. 18. I9. 22. 23. 24. 25. 28. 29. 30. 3 I. 32. 33. 34. 35. 36. 37. 38. 39. 72 Expressed esteem or respect for a competency or personal quality of yours. Gave you some information on how to do something. Suggested some action that you should take. Gave you over $25. Comforted you by showing you some physical affection. Gave you some information to help you understand a situation you were in. Provided you with some transportation. Checked back with to see if you followed the advice you were given. Gave you under 325. Helped you understand why you didn t do something well. Listened to you talk about your private feelings Loaned you or gave you something (a physical object other than money) that you needed. Agreed that what you wanted to do was right. Said things that made your situation clearer and easier to understand. Told you how he/she felt in a situation that was similar to yours. Let you know that he/she will always be around if you need assistance. Expressed interest and concern in your well-being. Told you that he/she feels very close to you. Told you who you should see for assistance. Told you what to expect in a situation that was about to happen. Loaned you over $25. Taught you how to do something. Gave you feedback on how you were doing without saying it was for good or bad. Joked and kidded to try to cheer you up. Provided you with a place to stay. Pitched in to help you do something that needed to get done. Loaned you under $25. APPENDIX D ARIZONA SOCIAL SUPPORT INTERVIEW SCHEDULE (ASSIS) 73 ARIZONA SOCIAL SUPPORT INTERVIEW SCHEDULE (ASSIS) I would like to get an idea of the people who are important to you In a number of different ways. I will be reading descriptions of ways that people are often important to us. After I read each description I will be asking you to give me the first names, initials, or nicknames of the people who fit the description. These people might be friends, family members, teachers, ministers, doctors, or other people you might know. If you have any questions about the descriptions after I read each one, please ask me to try and make it clearer. AEBIMAILEEELLNQS I. If you wanted to talk to someone about things that are very personal and private, who would you talk to? Give me the first names, Initials, or‘nicknames of the people that you would talk to about things that are very personal and private. EBQBE; Is there anyone else that you can think of ? 2. Wm, which of these people did you actually talk to about things that were personal and private? EBQBE; Ask specifically about people who were listed in response to ‘I but not listed in response to *2. 3. During the past month, would you have liked: I - a lot more opportunities to talk to people about your personal and private feelings 2 - a few more opportunities 3 =- or was this about right? 74 4. During the past month, how much do you think you needed pe0ple to talk to you about things that were very personal and private? I - not at all 2 - a little bit 3 = quite a bit BEIAIEBLALALD I. Who are the people you know that would lend or give you $25 or more if you needed it, or would lend or give you something (a physical object) that was valuable? You can name some of the same people that you named before if they fit this description, too, or you can name some other people. EBQBE: Is there anyone else that you can think of? 2. Dnflnginsnaslmnmn, which of these people actually loaned or gave you some money over $25 or gave or loaned you some valuable object that you needed. EBQBE Ask about people named in response to ‘I that were not named in response to *2. 3. During the past month, would you have liked people to have loaned you or to have given you: I - a lot more 2 - a little more 3 - or was it about right? 4. During the past month, how much do you think you needed people who could give or lend you things that you needed? I - not at all 2 - a little bit 3 - quite a bit 75 C. ADVICE I. Who would you go to if a situation came up when you needed some advice? Remember, you can name some of the same people that you mentioned before, or you can name some new people. EBQBE Anyone else? 2. mm, which of these people actually gave you some important advice? EBQBE: Inquire about people who were listed for *I but not for ‘2. 3. During the past month, would you have liked: I - a lot more advice 2 - a little more advice 3 = or was it about right? 4. During the past month, how much do you think you needed to get advice? I - not at all 2 - a little bit 3 - quite a bit OW I. Who are the people that you could expect to let you know when they like your ideas or the things that you do? These might be people you mentioned before or new people. BBQBE Anyone else? 76 2. WWI], which of these people actually let you know that they liked your ideas or liked the things that you did? PROBE: Ask about individuals who were listed for *l but not for ‘2. 3. WW, would you have liked pe0ple to tell you that , they liked your ideas or things that you did: I = a lot more often 2 =- a little more often 3 - or was it about right? 4. Duringlhmastmpntb, how much do you think you needed to have people let you know when they liked your ideas or things that you did? I - not at all 2 - a little bit 3 - quite a bit 6W I. Who are the people that you could call on to give up some of their time and energy to help you take care of something that you needed to do--things like driving you someplace you needed to go, helping you do some work, going to the store for you, and things like that? Remember, you might have listed these people before or they could be new people. 128085“ Anyone else you can think of? 2. DuringibuasLmnbtb, which of these people actually pitched in to help you do things that you needed some help with? 77 1280st Ask about people who were named in response to *l but who were not named in response to ‘2. 3. WOULD, would you have liked: I - a lot more help with things that you needed to do. 2 = a little more help 3 = or was this about right? 4. Wampum, how much do you feel you needed people who would pitch in to help you do things? I - not at all 2 = a little bit 3 - quite a bit ESQCIALBABILCLEAIIQN I. Who are the people that you get together with to have fun or to relax? These could be new names or ones you listed before. EBQBE' Anyone else? 2. WW, which of these people did you actually get together with to have fun or to relax? EBQBE Ask about people who were named in ‘l but not in *2. 3. WW, would you have liked: I - a lot more opportunities to get together with peOpIe for fun and relaxation 2 - a few more 3 - or was it about right? 78 4. How much do you think that you needed to get together with other people for fun and relaxation during the past month? I = not at all - 2 - a little bit 3 = quite a bit 8W I. Who are the people that you can expect to have some unpleasant disagreements with or people that you can expect to make you angry and upset? These could be new names or names you listed before. PROBE Anyone else? 2. WM, which of these pe0ple have you actually had some unpleasant disagreements with or have actually made you angry and upset? EBQBE' Ask about people listed for ‘I but not for ‘2. HW Now I would like to get some information about the people you have just listed. For each person on the list, could you tell me what this person's relationship is to you? For family members, specify the relationship (mother, father, brother, sister, grandmother, etc.). For professional people, also specify the exact profession (teacher, minister, doctor, counselor, etc.) (If sex of the person is unclear, ask for clarification). APPENDIX E EXPERIMENTAL INSTRUCTIONS 79 EXPERIMENTAL INSTRUCTIONS ' The experiment you are participating in is designed to study individual differences on a test of perception. A series of faces and letters will be projected on the screen in front of you. The faces are of two models, one male and one female, portraying two different closed mouth expressions: angry and neutral (indifferent). The faces will always be presented in the center of the screen. A letter (P or R) will be presented at the same time that a face is presented. The position of the letters may be different from one slide to the next. In half of the slides the letters will be above the faces, and in the remainder of the slides the letters will be below the faces. There are an equal number of Ps and Rs, as well as an equal number of angry and neutral and male and female faces. There will be 4 parts to the test. Before beginning Part I of the test you will be shown 2 sets of 8 slides. You will be asked to identify the facial expressions in the first set and the letters In the second set. Each slide will be presented for 3 seconds during this preview period. Following this preview we will commence with Part I of the test. 80 EXPERIMENTAL INSTRUCTIONS: PART I In Part I, your attention should be directed to the faces that will be projected on the screen. You are to ignore the letters in this part of the test. Pairs of slides will be presented with a 2 second interval between them. One of the slides will show an angry face and the other slide will show a neutral face. The sex of the model will be the same within each pair of slides. Your task is to identify which one of the pair of slides (i.e., first or second) shows the angry face. Each face will be presented in the center of the screen. Your best strategy will be to focus your attention on the black dot (central point) in the center of the screen. As the slides will be, presented for a very brief period of time, It will be necessary for you to pay close attention. You will be given a “ready“ signal just before each trial. Be sure to focus your attention on the central point as soon as you hear this “ready" signal. Following the ”ready” signal the first slide will be flashed on the screen. The second slide will be presented 2 seconds later. Please wait until both slides have been presented before responding. Call out either "first" or “second” to identify which of the two slides contains the angry face. In the beginning, the test will probably seem fairly easy to you. Gradually, the amount of time you are allowed to see the faces will be reduced and It will become rather difficult to identify the expressions. Do your very best to accurately identify which slide shows the angry face. If you are not sure of the correct response, please give an answer anyway, even if it means guessing. It is important that you respond quickly, as a different pair of slides will be presented every few seconds. 8) EXPERIMENTAL INSTRUCTIONS: PART 2 In the Part 2 of the test your attention should be directed to the letters that will be projected on the screen. You are to ignore the faces In this part of the test. Again, pairs of slides will be presented with a 2 second interval between them. One of the slides will show a letter R and the other slide will show a letter P. Your task Is to Identify which one of the pair of slides (i.e., first or second) shows the letter R. The position of the letter will be random, so that for any particular slide, the letter may be above the face or below the face. Your best strategy will be to focus your attention on the central point in the center of the screen. A ”ready" signal will be given to alert you that a trial Is _ about to begin. Be sure to focus your attention on the central point as soon . as you hear the ”ready" signal. The first slide will be flashed on the screen soon after the ”ready" signal. The second slide will be presented 2 seconds later. Please wait until both slides have been presented before responding. Call out either ”first" or ”second“ to identify which of the two slides contains the letter R. As was the case in Part I, this part of the test will probably seem rather easy in the beginning. Again, however, the amount of time you are allowed to view the letters will be gradually reduced, making the task more difficult. Do your very best to correctly Identify which slide shows the letter R. If you are not sure of the correct response, please give an answer anyway, even if it means guessing. And please do respond quickly. A different pair of slides will be flashed on the screen every few seconds. 82 EXPERIMENTAL INSTRUCTIONS: PART 3 In Part 3 of the test your task will be to do two things at the same time. Your primary and most important task will be to identify the expression (angry or neutral) of each face that is projected on the screen. It is important that you do your very best to correctly identify each facial expression. In addition to Identifying the expression of each face, you are to identify the letter (R or P) that will randomly appear above or below the face. 00 as well as you can to identify the letters, knowing that your most important task Is to get as many of the expressions correct as you can. The procedure In Part 3 will be essentially the same as it was In Part I and Part 2. A "ready” signal will alert you that a trial Is about to begin. Be sure to focus your attention on the central point as soon as you hear the ”ready” signal. A face and a letter will be flashed on the screen soon after the "ready” signal. First report the expression, then report the letter. Call out either ”angry,“ or ”neutral," followed by 'P" or “R.” Before beginning the test trials, you will be given l6 practice trials. The first few practice trials will probably seem fairly easy to you. However, the amount of time you are allowed to see the practice slides will be gradually reduced, and it will become increasingly difficult to see the stimuli. Following the practice trials, we will begin the test trials. Do your very best to correctly Identify each expression. At the same time, try to identify the letters. If you are not sure of the correct response, please give an answer anyway, even if means guessing. You must respond quickly, as a different slide will be presented every few seconds. 83 EXPERIMENTAL INSTRUCTIONS: PART 4 In Part 4 of the test your task again will be to do two things at the same time. However, now your primary and most important task will be to identify the letter (R or P) that is projected on the screen. It is important that you do your very best to correctly identify each letter. Again, the letters will be randomly presented above or below the faces on each trial. In addition to identifying each letter, you are to report the expression (angry or neutral) of the face that will appear with the letter. 00 as well as you can to Identify the facial expressions, knowing that your most Important task is to get as many of the letters correct as you can. The procedure in Part 4 will be essentially the same as it was in the other 3 parts of the test. A “ready" signal will alert you that a trial is about to begin. Be sure to focus your attention on the central point as soon as you hear the ”ready” signal. A letter and a face will be flashed on the screen soon after the ”ready“ signal. First report the letter, then report the expression. Call out either “P,“ or “R,“ followed by ”angry“ or ”neutral.“ Before beginning the test trials, you will be given I6 practice trials. As was the case In Part 3, the first few practice trials will probably seem fairly easy to you. Following the practice trials, we will begin the test trials. 00 your very best to accurately Identify each letter. At the same time, try to identify the expressions. If you are not sure of the correct response, please give an answer anyway, even if means guessing. Again, you must respond quickly, as a different slide will be presented every few seconds. APPENDIX F CONSENT FORM 84 RESEARCH CONSENT FORM Michigan State University Department of Psychology I have freely consented to take part in a scientific study entitled ”Social support and perception” being conducted by Lance Schumacher under the supervision of Dr. Norman Abeles, Professor of Psychology. The study has been explained to me and I understand the explanation that has been given and what my participation will Involve. I understand that I am free to discontinue my participation In the study at any time without penalty. I understand that the results of the study will be treated In strict confidence and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. I understand that my participation In the study does not guarantee any beneficial results to me. I understand that, at my request, I will receive additional explanation of the study after my participation is completed. Signed: Date: APPENDIX 6 TABLES I -XI 85 TABLE I FORCED-CHOICE MEANS, STANDARD DEVIATIONS, AND E SCORES Type!) _I¥ne_B__ 6nnnflus_______JMHMLinmncLJfl> JfiauLlamncLJiL F Fm 43.54 19.90 41.25 16.71 .18 Letters 42.79 22.53 36.54 I 2.92 I .64 M118. All tests are two-tailed, with d! = I, 46. TABLE II DUAL-TASK d'. MEANS, STANDARD DEVIATIONS, AND E SCORES IvgeA Tvge B Stimulus Mean SD 11m SD F Female Face Primary Focus .93 .45 .92 .45 .00 Seconmry Focus .44 .24 .50 .25 .56 rhfleFazz Primary Focus 1.92 .68 1.87 .46 .07 Secorkwy Focus .96 .60 1.26 .56 308* Letter Primary Focus 1.18 .30 1.24 .36 .30 My Focus .67 .24 .70 .25 .26 up): All tests are one-tailed, with d! = l ,40. *p<.05 86 TABLE III DUAL-TASK fl MEANS, STANDARD DEVIATIONS, AND E SCORES TypeA I34): 8 Stimulus Mam SD Mam SD F Female Face Primary Focus 1.02 1.35 ' 1.12 1.23 1.54 SeoondaryFocus 1.07 1.12 1.10 1.20 .40 Male Face Primary Focus 1.59 1.91 1.05 1.70 578* Swoodary Focus 1.32 1.66 1.26 1.70 .05 Letter Primary Focus .93 1.18 1.02 1.29 2.00 SeoondaryFocus .91 1.12 .91 1.12 .04 Note: All tests are one—tailed, with d! = 1 ,40. *p < .05 TABLE IV ISSB MEANS, STANDARD DEVIATIONS, AND F SCORES Type A Lime 8 times rim—J Mm SD E Emotional support 25.92 7.41 22. 13 6.77 3.38 Tangible assistance 15.46 5.04 14.08 ' 3.46 1.23 Cognitive information 29.42 6.99 26.29 5.43 2.90 Direct Guidance 17.96 5. 13 14.46 2.75 8.98“ Total 1 07.63 25.88 93.92 I 5.50 476* Note All tests are one-tailed, with d! = 1 ,46. *p<.05 ”p < .01 87 TABLE Y ASSIS MEANS, STANDARD DEVIATIONS, AND E SCORES __IxneA_ Type D We Mean SD film 30 F Private feelings Available 3.83 2.51 3.88 1.54 .01 Enacted 2.88 2.31 2.38 1.35 .85 Satisfaction 2.54 .72 2.71 .55 .81 Material Aid Available 8.67 5.28 7. 79 3.31 .46 Enacted 2.33 2.08 2.13 1.54 . 15 Satisfaction 2.88 .34 2.92 .28 .22 Advice Available 3.67 2.58 4.38 2.92 .80 Enacted 2.00 1.79 2.54 1.62 1.18 Satisfaction 2.88 .34 2.83 .38 . 17 Positive fearback Available 7.13 4.81 5.88 3.07 1.15 Enacted 4.96 2.97 4.04 2.53 1.27 Satisfaction 2.25 .94 2.25 .74 .00 Physical assistance Available 7.42 4.10 6.08 3.26 1.53 Enmted 4.17 2.41 3.42 2.34 1.16 Satisfmtion 2.79 .51 2.67 .57 .62 Social participation Available 9.25 5.68 8.42 4.47 .33 Enwted 8.71 5.68 6.96 3.74 1.63 ' Satisfaction 1.83 .82 1.96 .69 .34 Total Unconflicted Available 15.88 7.27 12.71 5.40 2.84 Enacted I 1.88 6.07 9.71 4.37 2.03 88 TABLE V (Continued) Total Conflicted . Available 3.88 2.72 3.83 1.74 .01 Enacted 2.42 1.89 2.29 1.49 .06 Note All tests are one-tailed, with d! = 1 ,46. TABLE VI CORRELATIONS BETWEEN ISSB AND SECONDARY (I: _Eamala_En_ _t1a1a.Ea:a__ _Lat!ar__ WWW Emotional Support -.28 --.09 -.07 -.41* .19 -.19 Tangible Assistance -.O4 .27 .01 -.21 .01 .16 Cognitive Information -.13 , -.36 .02 -.45* .26 -.14 Direct Guidance -.O4 -.28 -. 12 -.21 .09 -.01 Total -. 16 -.28 -.06 -.55** .15 -.05 *p < .05 **p<.01 TABLE VII CORRELATIONS BETWEEN ASSIS AND SECONDARY d1 mm. mm— Letter times 1W Private Feelings Availwle -.22 .29 -.27 .16 .49“ - .03 Enacted —.23 .16 -.27 .13 51* . 17 Satisfmtion -.I7 .09 .03 -.16 -.08 -.10 TABLE VII (Continued) 89 Material Aid Available -.12 -.08 .20 -.I9 .20 .18 Enacted -.22 -.I3 .05 -.47* .35 .27 Satisfaction .12 - .56" .07 - .31 .26 .09 Advice Available -.33 -.06 -.20 -. 18 49* - .02 Enacted -.34 .01 -.37 .11 46* -.08 Satisfaction .17 .19 .27 .33 .19 .17 PmflUveFanmwk Available -.26 .1 l .18 -.O4 .31 -.28 Enacted -.45* .06 .06 .21 .25 -.45* Satisfaction .07 .19 -.06. .10 .02 - .30 FstmahMBEmmwe Available -.15 —.19 .17 -.I3 .07 .07 Enacted -.23 -.09 -. 18 -.07 .05 .00 Satisfaction .19 .14 .24 .16 .28 -.21 Social Participation Available -.31 -.O4 .07 -.07 .20 .05 Enacted -.40 -.IO .03 -.03 .22 -.16 Satisfaction .17 .34 .18 .1 l -.07 -.I 1 Total Unconflicted Available -.33 -.06 .1 I -.16 .22 .07 Enacted -.46* -.11 -.01 -.22 .28 -.10 Total Conflicted IWmfldfle :18 -207 .15 -210 .39 -.11 Enacted -.27 -.03 -.12 -. l 9 .36 .00 *p<.05 **p < .01 90 TABLE VI!) CORRELATIONS BETWEEN ISSB AND PRIMARY 3 Emotional Support -.46* 43* .01 -.35 -.08 .35 TangibleAsaistance -.16 -.10 -.09 -.13 .14 .13 Cognitive Information .26 .04 . 16 -.40 .09 -. 14 Direct Guidance —.12 .16 -.O2 -.30 -.19 .03 Total -.31 .19 .09 -.4I‘ .00 .01 *p<.05 TABLE IX CORRELATIONS BETWEEN ASSIS mo PRIMARY B _Eemnle.E£I:e_ Elohim: Latter: Elmore Tye“ Type B Twain—M. Private Feelings Available -.30 .04 .05 -.01 -.27 .19 Enacted -.30 .39 -.04 -.23 -.30 .01 Satisfaction .14 -.02 -. 18 —.O4 .29 .07 Material Aid Available .07 .23 .20 -.24 .07 -. 18 Enacted -.25 .56“ -.02 -.31 .14 -.07 Satisfaction .17 -.II -.22 -.07 -. 18 -.05 Advice Available -.13 -.21 .06 .09 -.17 -.24 Enacted -.33 -.19 .22 .11 -.29 .01 Satisfmtion -.04 -.08 -. I 2 .30 .32 .22 91 TABLE IX (Continued) Positive Feedback Available -.34 .03 .05 .00 .41 * -. 18 Enacted -.43* -.13 .05 -.01 ,_ .38 -. 12 Satisfaction .28 -.16 -.Ol .29 -.33 -.09 Physical Assistance Available .12 -.02 .12 .17 .36 -.17 Enacted -.15 .10 -.05 .15 .38 —.O7 Satisfaction .09 .23 .00 -.04 -. I6 -. 12 Social Participation Available -.21 -.23 .14 . 12 .34 -.04 Enacted -.35 -.15 .17 -.03 .32 -.02 Satisfaction .27 -.06 .OO -.02 -.28 .09 Total Unconflicted Available. -.09 -.22 .1 I .06 .29 -. 18 Enacted -.33 -.19 .02 -.05 .35 -.21 Total Conflicted Available -.28 -.28 -.ll -.04 .41* -.O8 Enacted -.46* -.07 -.08 .02 .24 -.O4 *p<.05 **p < .01 TABLE X CORRELATIONS BETWEEN 1588 AND SECONDARY B m _t1ala.ane__ Letter: time 1W Emotional Support -.36 .07 -.06 -.25 .14 .10 Tangible Assistance .02 . 16 -.23 -.33 - .06 - .28 Comitive Information -.04 -.22 .06 -.37 .21 -. 13 Direct Guidance .03 -.28 -.14 -.20 -.06 -.15 Total --.10 -.18 -.08 -.38 .10 -.I7 92 TABLE XI CORRELATIONS BETWEEN ASSIS AND SECONDARY fl m Male [age Letter time Type!) M123 Ives A Type B Type A Type 8 Private Feelings Available -.14 .32 - .22 .29 -.06 .18 Enacted -.II .46* -.17 .14 -.08 .24 Satisfaction . 18 .07 -.15 -.O7 -.02 -.l 1 Material Aid Available .21 .05 -. 10 .08 -.14 .13 Enacted -.03 .10 .01 -.15 .14 -.37 Satisfaction .07 -.29 .21 .15 -.03 -.18 Advice Available -.07 -.33 -.08 .00 . 12 -.13 Enacted -.12 -.19 , .02 -.06 .19 .16 Satisfaction .07 .30 ’ 22 .02 . 10 .29 Positive Feedback Available .00 -.15 -.24 .07 -.02 .21 Enacted -.17 -.22 -.33 .29 .03 .38 Satisfaction .1 1 -.05 -.04 -. 18 -.04 .14 Physical Assistance Available 46* -.27 -. 12 .15 -.06 -.10 Enacted .11 -.15 -.24 .31 .25 -.03 Satisfaction .08 -.O3 .10 .13 -. 13 . 12 Social Participation Available .17 -.21 -.22 .13 -.01 -.09 Enacted -.01 -.32 -.22 .04 .04 . 12 Satisfaction . 17 .09 .20 . 15 -.I 1 .06 Total Unconflicted Available .19 -.22 -.22 . 12 -.06 -.08 Enacted .01 -.36 -.28 -.02 .01 .03 TABLE XI (Continua!) 93 Total Conflicted Awandfle .12 -224 Enacted -.08 -. 15 -.I9 -.16 -.OS -.04 -.OS .06 -.02 -.02 *p<.05