CONJUGATE LATERAL EYE MOVEMENT AND SUBJECTIVE REPORTS OF ANXIETY SYMPTOMS Thesis for the Degree of M. A. .MkCHIGAN STATE UNIVERSITY ARTHUR ROBB BUSS 1971 \HF‘U. LIBRARY Michigan Sm: University . ABSTRACT CONJUGATE LATERAL EYE MOVEMENT AND SUBJECTIVE REPORTS OF ANXIETY SYMPTOMS By Arthur Robb Bliss Correlates of a conjugate lateral eye movement phenomenon are examined. The phenomenon occurs in a dyadic interaction when a personis asked a question requiring some reflection to answer. The initial movement of the eyes, eitherright or-left, from the questioner is the movement referred to; it tends to be consistent for individuals. A list of 51 bodily reactions of anxiety to stressful situations was given to 148 undergraduate students to rate on a frequency of occurrence scale. The frequence of occurrence of each-of these-symptoms was correlated with the variables sex and direction of eye movement. Previous authors have linked this eye movement phenomenon to anxiety reactions, reporting that moving the eyes to the left is associated with increased heart rate and breathing rate, flaccid large postural muscles, a tendency to be alcoholic, and general Arthur Robb Bliss sympathetic nervous system reactivity. Moving the eyes to the right is associated with a slowing and strengthening of both heart rate and breathing rate, an increase in underarm perspiration, high tonus in the large postural muscles, more muscle tics and twitches, more speech disturbance than left eye movers, and general para - sympathetic nervous system reactivity. The general finding is that the variables of sex and eye movement are relatively weak predictors over the set of anxiety symptoms; however, most all of the previously reported findings were, if not significant, at least in the predicted direction. The main finding not supported was the attribution of sympathetic reactivity to left eye movers and of parasympathetic reactivity to right eye movers. A more striking finding was the report of symptoms concerned with interior body functioning by left eye movers, especially for males; and the report of symptoms con— cerned with the body' 3 surface layers by right movers, especially for males. This pattern did not hold for females considered separately. The primary sex difference found was that females report more symptoms more often than do males. Clusterand factor analyses of the anxiety symptoms are presented. The symptoms cluster around particular areas and functions of the body, suggesting an inner vs. outer or body layer Arthur Robb Bliss scheme for investigating anxiety symptoms instead of a sympathetic- parasympathetic dichotomy. A discussion of the problems with the data and suggestions for further research are offered. Approved May 13, 1971 Paul Bakan, Chairman John E. Hunter Arthur Seagull CONJUGATE LATERAL EYE MOVEMENT AND SUBJECTIVE REPORTS OF ANXIETY SYMPTOMS By Arthur Robb Bliss A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1971 ACKNOWLEDGMENTS I give sincere appreciation and thanks to the following persons for their close help and support in this project: Paul Bakan for his inspiration and the data; John Hunter for his patience and statistical know -how; Kowit Pravalpruk for his friendship and assistance‘with the factor analysis; and Bonnie Bliss, who didn' t bug me about finishing. ii TABLE OF CONTENTS Page LISTOFTABLES....................v INTRODUC TION The Conjugate Lateral Eye Movement Phenomenon Correlates to CLEM A Conceptual Model Sex Differences Differences in Reaction to Anxiety Objectives of this Study PROCEDURE 1 0 Determination of CLEM Subjective Report of Reactions to Anxiety RESULTS 12 General Response Tendencies Results of Analysis Over All 51 Variables LM-RM Differences Sex Differences Results of Cluster and Factor Analyses An Oblique Multiple Groups Solution A Varimax Rotation to Five Factors Solution LM-RM Differences Sex Differences DISCUSSION . 3 1 General Response Tendencies Discussion of Analysis Over All 51 Variables iii Page LM-RM Differences for Males and Females Combined LM-RM Differences for Males Considered Separately LM-RM Differences for Females Considered Separately Sex Differences Discussion of Cluster and Factor Analyses The Oblique Multiple Groups Solution The Varimax Rotation to Five Factors Solution RM - LM Differences Sex Differences Problems and Suggestions for Further Study Mmary LIST OF REFERENCES . . . . . . . . . . . . . . . . . . 44 APPENDIX A. PROVERBS USED TO DETERMINE CLEM . . . 46 B. BODY FEELING INVENTORY . . . . . . . . . 47 C. AVERAGE RESPONSE SCORES OVER ALL 51 VARIABLES FOR THE FOUR MAIN SUBJECT GROUPS . . . . . . . . . . . . . . 50 D. TABLE OF CORRELATIONS OVER ALL 148 Ss BETWEEN THE 51 ANXIETY VARIABLES, THE 8 GROUP FACTORS FROM THE FACTOR ANALYSIS SOLUTION AND THE VARIABLES SEX, EYE MOVE - MENT AND THE PRODUCT INTERACTION OF SEX AND EYE MOVEMENT . . . . . . . . 53 E. TABLE OF INTERCORRELATIONS BETWEEN THE 8 GROUP FACTORS FROM THE FACTOR ANALYSIS SOLUTION AND THE VARIABLES SEX, EYE MOVEMENT AND THE PRODUCT INTERACTION OF SEX AND EYE MOVEMENT; THE STANDARD SCORE COEFFICIENT ALPHAS OR THE RELIABILITIES OF THE 8FACTORGROUPS 55 iv TABLE 1a. 1b. 3a. 3b. LIST OF TABLE S Page Distribution of average response scores over all 51 variables for all 148 _§_s and group average scores . . . . . . . . . . . . . . . 13 Number of _S_s by group below and above the overall average and Chi -square tests for-significance . . . . . . . . . . . . . . . . . 13 Eye movement effects in order of significance up to p < . 20 for all 148 _S_s; point—biserial correlations are given . . . . . . . . . . . . . . 15 Eye movement effects in order of significance up to p < . 20 for the 78 male Es; point- biserial correlations are given . . . . . . . . . . 18 Eye-movement effects in order of significance up to p < . 20 for the 70 female _S_s; point— biserial correlations are given . . . . . . . . . . 18 Sex effects over all 51 variables for all 148 _§_s, in order of significance up to p < . 20; point-biserial correlations are given . . . . . . . . 20 Final oblique multiple groups solution after positive manifold has been partialed out . . . . . . 22 Factor analysis, Varimax rotation to five factors, Kiel -Wrigley criterion, factors VI- VIII from previous multiple groups solution; factor loadings are given . . . . . . . . . . . . . 25 TAB LE 8a. 8b. Page Eye movement effects in order of significance up to p < . 50 for all 148 _S_s; point-biserial correlations are given . . . . . . . . . . . . . . 27 Eye movement effects in order of significance up to p < . 50 for the 78 male _S_s; point- biserial correlations are given . . . . . . . . . . 29 Eye movement effects in order of significance up to p < . 50 for the 70 female _S_s; point- biserial correlations are given . . . . . . . . . . 29 Sex effects in order of significance up to p < .50 for all 148 _S_s; point-biserial correlationsaregiven . . . . . . . . . . . . . . 30 vi INTRODUC TION The Conjgate Lateral Eye Movement Phenomenon The phenomenon of conjugate lateral eye movement, CLEM, has been investigated by Day [ 1 ] and later and independently by Libby [2 ] . When a person is asked a question that requires some reflection while he is in eye contact with the questioner, he tends to move his eyes to either the right or the left; up and down move- ~ments may also occur at the same time. Day noted that movement to the right or left is consistent for a given individual. Day [3 ] later reported that the direction of CLEM was not related to handed- ness or to eye dominance. Duke [4] showed that CLEM was a relatively reliable phenomenon; that there was a strong association with reflective thought processes; and that there was a lack of asso- ciation with sex, handedness, and with eyedominance factors. Libby [2 ] has also shown these relations. Correlates to CLEM Day'[ 1 ] observed that persons with left directed CLEM, left movers or LMS, tended to have a passive internalized mode of attention and to emphasize subjective internal experience and sensations; persons with right directed CLEM, right movers or RMS, tended to have a passive externalized mode of attention and to emphasize externally perceived causes for anxiety and tension. In a. later paper Day [ 5] showed that LMS and RMS show systematic differences in EEG and ECG reactivity, e. g. , EEG records of LMS showed a greater amplitude and lower frequency than those of RMS. Day states that these differences appear to relate to individual dif- ferences in higher verbal processes, reaction to anxiety, and to characteristic functional differences in attentional processes. Day' 8 description of LMS as tending to focus more on internal, subjective experiences, and Hilgard' s nearly parallel description of the good hypnotic S, led Bakan [6] to show that indeed LMS had the highest scores on the Stanford Hypnotic Susceptibility Scale (Form C) and RMS had the lowest scores. Day [5] found that the EEG records of LMS show greater amplitude and lower frequency than those of RMS. Bakan [7 ] cites a number of results showing that high alpha -wave (8- 13 cycle -per- second) production on EEGs was correlated with high susceptibility to hypnosis. Combining these findings, Bakan and Svorad [8] went on to show that LMS produce alpha -waves 52% of the time and RMS only 20% of the time. A replication of this study by Strayer [9] showed that left CLEM was more frequent in persons with high basal alpha -wave levels. However, this relation was true for males but not for females. Also LMS were able to increase their alpha -wave production and RMS were not. Bakan hypothesizes that since relaxa - tion facilitates alpha -wave production, the LMS were more able to relax. Day observed that RMS have greater tonus in their large postural muscles indicating less ability to relax and thus supporting the relaxation hypothesis. Some additional differences between RMS and LMS noted by Bakan [10] are that RMS have more muscular tics and twitches than LMS and male LMS are more likely to be alcoholic. A Conceptual Model Since the direction of CLEM is correlated with these kinds of variables, the question arises as to how this data might be brought into a conceptual model of nervous system functioning. According to Bakan [7], One of the most frequent motor responses to electrical stimula- tion of the cerebral cortex is eye -movement. When parts of the left cerebral hemisphere are stimulated, the eyes will move to the right. When the [same] parts of the right hemisphere are stimulated, the eyes shift to the left. Bakan' s thesis is that CLEM is related to the two asymmetrical hemispheres of the brain and that the personalities of persons who look either right, RMS, or who look left, LMS, are more likely to reflect the style of left or right hemisphere functioning respectively. Bakan proposes, in this regard, a double cerebral dominance model for brain functioning, i. e. , that dominance shifts between the two hemispheres of the brain depending upon the particular behavior or function involved. To quote Bakan [7 ] : Scientists have long believed that the left hemisphere is dominant for speech, writing, and other symbolic functions. There is also evidence that the left hemisphere dominates in such functions as arithmetic, analytical activity, . logical and abstract thinking, time discrimination and euphoria. There is rapidly accumulating evidence that the right hemisphere is dominant for such nonverbal functions as spatial perception, body -image functions, imagery, recogni- tion of melodies, recognition of faces and .for such perceptual- motor constructional functions as drawing and block design. There are also data that support association of the right hemi- sphere with creativity, depression and certain language functions. Bakan also notes recent evidence that the left hemisphere functions at a higher rate of arousal than the right hemisphere. This may explain the finding of greater amounts of EEG alpha wave production associated with a low arousal or more relaxed state occurring in the right hemisphere. Bakan suggests that other states of consciousness associated with relaxation or-low arousal, e. g. , hypnosis, meditation, daydreaming, psychedelic intoxication and anesthesia, may be associated to a greater extent with right hemi- sphere functioning; he notes in addition that these states are characterized by the absence of the left hemisphere function of . logical speech. Sperry [ 11] has studied a number of persons who have had their right and left hemispheres surgically separated. Thus able to test very directly the differential functioning of the two hemispheres in adults, he characterized the two hemispheres as follows: the left hemisphere tends to be verbal, rational, digital and objective while the‘right hemisphere tends to be nonverbal, emotional, analogical and subjective. Again Bakan proposes that RMS will be better described by the left hemisphere functions and LMS by the right hemisphere functions. Of course in studies of persons with the two hemispheres intact and connected, the communication between the two hemispheres allows for a wide range of possibilities of integra- tion of many functions. Thus differences based upon the criterion of direction of CLEM may not be expected to be in many instances clearly significant. Sex Differences In several studies of CLEM it has been shown that females are less likely to move their eyes consistently either left or right and are‘thus less frequently consistent LMS or RMS. Day [3 ] notes a tendency for males to be RMS and for females to be LMS. Differences in the EEG alpha -wave production in right and left hemispheres are not as clear for females as for males. Brain surgery data indicate that female functioning is less impaired than male functioning after removal of comparable parts of the left or right hemisphere. This data and additional data showing differences occurring in cortical evoked potentials all seem to indicate that there is more hemispheric integration in females and that there are dif- ferences in asymmetrical organization of the brain for males and females. Bakan [7 ] suggests that many observed cognitive and affective differences between the sexes may be related to these dif— ferences in lateral central nervous system organization. Differences in Reaction to Anxiety Day has written that RMS and LMS differ in their reactions to anxiety and/or to stress producing situations. In his earlier paper Day [1] observed that right CLEM was "perfectly" related to _S report of anxiety as a panic feeling with an externalized perception of cause. Thus for persons with what he describes as a passive externalized mode of attention, the qualitative experience of anxiety is fear and cholinergic autonomic (parasympathetic) sensations are reported somatically. Similarly left CLEM was "perfectly" related to _S_ report of anxiety as tension, as in internally perceived impulse feeling. Thus for persons with a passive, auditory internalized mode of attention, adrenergic autonomic (sympathetic) sensations are reported somatically. Day offers a hypothesis to explain how the two personality types are created. If early experience of the person is mainly externalized, as might be associated with a greater use of verbs, then any passive internalized attention would result in a panicky feeling of anxiety, as for right eye -movers, RMS. If early experience is internalized, as might be associated with greater use of adjectives, then any active externaliz ed attention would result in feelings of tension, as for left eye -movers, LMS. In a later study Day [3] made a number of observations related to CLEMs and to_S_ reactions to stress. Some of Day' 5 specific observations that apply most directly to the present study are as follows. LMS tend to emphasize hyperactivity of the upper viscera with increase in heart rate and in breathing rate; they experience an inability to maintain visual attention because of (inferred) obsessive intrusions; and the large postural muscles are often flaccid even though the S complains of generalized muscle tension. RMS tend to emphasize lower visceral hyperactivity with slowing and strengthening of heart rate and breathing rate; they have an increase in axillary armpit perspiration; and their large muscle tonus is high. Objectives of this Study To rephrase Day' 5 observations in terms of hypotheses, we would expect that, under stressful conditions, LMS will have increased heart rate and breathing rate; they should find it difficult to maintain visual attention and will be preoccupied with their thoughts and sensations; and they should have flaccid large postural muscles. RMS should have a slowing and strengthening of both heart rate and breathing rate; they should have an increase in underarm perspira - tion; and they should have high tonus in their large postural muscles. In addition to these hypotheses a number of other emotional reactions to anxiety and stress will be investigated; see Appendix B for a com - plete list of symptoms. One hypothesis to be tested is that speech disturbance, primarily a left hemisphere function, will be reported more often by RMS. Day' s data regarding differences between LMS' and RMS' reactions to anxiety come from his clinical observations. The objective of this study is to provide empirical data based on sub- jective reports of emotional reactions to anxiety provoking situations. This method was chosen as an initial investigatory procedure pri- marily because of the relative ease in gathering the data. The‘data will be analyzed for general response set tenden- cies, LM-RM differences and sex differences. A broad range of reactions was included so that various cluster and factor analyses could be made. When the variables are collapsed into group factors, an additional analysis of LM-RM differences and sex dif— ferences will be made. Following discussion of the results, a section on problems with the data will be Offered along with suggestions for further research. PROCEDURE Determination of CLEM The _S_s were 148 male and female undergraduate psychology students at Michigan State University. They were selected from a larger number of students who were presented individually with ten proverbs to interpret. See Appendix A for a list of the ten proverbs used. The Es were tested in a plain room facing a wall without dis - tracting pictures or windows. The _§ faced the male _E straight on while the _E_ gave the instructions. The instructions were not formalized but essentially the _S_ was told that this was an experiment to investigate individual differences in thought processes as the _S_ responded by interpreting a list Of ten proverbs. Before stating each proverb the Ewould make sure that the _S_was facing him directly and had direct eye contact as the proverb was given. The _S_s first directional eye movement response was recorded for each of the ten proverbs. ~Those whose eyes moved to the right 70 to 100 percent of the time were called right eye -movers, RMS; and those whose eyes moved to the left 70 to 100 percent of the time were called left eye -movers, LMS. This criterion has been used by Bakan in 10 11 his studies of CLEM. A11_S_s were right handed. The 148 _SS are divided into four groups by sex and eye movement; there are 37 male RMS, 33 female RMS, 41 male LMS and 37 female LMS. Subjective Report of Reactions to Anxiety The 148 _S_s were later given a "Body Feeling Inventory" questionnaire in a group setting; see Appendix B for a copy of this questionnaire. The list of 51 items was compiled by Paul Bakan. The questionnaire asked for the frequency of bodily reactions experienced by the S during stressful or anxiety provoking situations, e.g. , waiting in a doctor' 3 office, taking an important exam, before talking in front of a group, etc. The 51 reactions or symptoms were listed to be rated on a five -point frequency of occurrence scale: 1 --always, 2 - - often, 3 - - sometimes, 4 --rarely, and 5 --never. The determination of CLEM was done independently of acquiring and analyzing the questionnaire results. RESULTS General Response Tendencies Table la shows the distribution of average response scores over all 51 variables for all 148 ES. The entries in Table 1a are the number of_S_s with average response scores between the points on the frequency scale, e. g. , two male LMS had scores in the interval 2. 5 to 3.0 and 15 male LMS had scores in the interval between 3. O and the average response score over all Es of 3. 52.. The column to the extreme right gives the average response score for each group listed in the first column. See Appendix C for the average score by the four main groups for each of the 51 variables. In Table 1b are shown the number of SS in each group both below and above the overall average of 3. 52. The results of Chi- square computations are given; only the grouping of all males vs. all females shows a significant result. The correlation between sex and anxiety is . 14 so females report more symptoms or reactions to stress than do males. 12 13 TABLE 1a. -- Distribution of average response scores over all 51 variables for all 148 SS and groupaverage scores Scale NO' Grou Group over“ of P Some- all 83 AVE- Often times Avg. Rarely Never — 2.0 2.5 3.0 3.52 4.0 4.5 5.0 1* T T l l I I M-LM o 2 15 20 3 1 41 3. 564 M-RM 1 2 11 13 10 o 37 3.594 F-LM 1 3 17 10 5 1 37 3. 462 F-RM o 4 13 15 1 o 33 3. 449 Total 2 11 56 58 19 2 148 3. 520 Males 1 4 26 33 13 1 3.578 Females 1 7 30 25 6 1 3. 456 LMS 1 5 32 30 8 2 3. 516 RMS 1 6 24 28 11 0 3. 526 TABLE 1b. -- Number of_S_s~by group below and aboverthe overall average and Chi ~square tests for significance [Overall Average = 3. 52 Group . Chi-square d. f. P, Below " Above M-LM 17 24 .105- 1 n.s. M-RM 14 23 F-LM 21 16 .187 1 ‘ n.s. F-RM 17 16 Males 31 47 3.17 . 1 . 10 Females 38 32 LMS 38 40 . 279 1 n. s. RMS 31 3.9 14 Results of Analysis Over All 51 Variables All point -biserial correlations reported were obtained using the computer program PACKAGE [12] . For all 148 _S_s, Appendix D is a table of correlations between each of the 51 anxiety variables, the 8 group factors from the factor analysis solution, and the 3 vari— ables sex, eye movement and the product interaction between sex and eye movement. The correlations used in Tables 3a, 3b, 8a and 8b for males and females considered separately are not included in the Appendices. The absolute value of the correlations over all 148 _Ss to be significant at the .05 level must be . 17 or greater; at the . 10 level, . 14 or greater; at the .20 level, . 11 or greater. Only a few corre- lations are . 30 or above; most reported are .20 or below. In Tables 3a and 8a for the 78 male _S_s considered separately, the absolute value of the correlations necessary for given significance levels are as follows: at the .05 level, r_>_ .23; at the . 10 level, r: . 19; and at the .20 level, r_>_ . 15. Similarly, in Tables 3b and 8b for the 70 female _S_s considered separately, the correlations neces- sary are: at the .05 level, r: . 24; at the .10 level, r_>_ . 20; and at the . 20 level, r: . 16. Again the correlations reported are on the order of . 15 to . 20. The tentative conclusion here is that the vari- ables of sex and eye movement are relatively weak predictors of performance on this set of anxiety items. 15 LM - RM Differences Table 2 is a list of eye movement effects over all 148 _S_s in. order/of significance up to p < .20. All of the items significant for p < . 10 are RM items. Muscle tics or twitches (27) and inability to speak coherently (37) have been shown to be RM items previously by Bakan; item 27 is seemingly related to the increased muscle tension of RMS as reported by Day, and item 37 appears to support the hypothesis that any emotional disturbance that affected speech and hence the left hemispherewould affect RMS more. Cold hands (3) and feet cold (44) seem related to peripheral vasoconstric- tion, an adrenergic (sympathetic) reaction. This result appears TABLE 2. -— Eye movement effects in order of significance up to p < .20 for all 148 §SS point -biserial correlations are given r Igim p < LM Items RM Items . 19 27 .05 Muscle tics or twitches . 17 3 .05 Cold hands . 15 44 . 10 Feet cold .. 14 37 . 10 Inability to speak coherently . 14 4 . 10 Underarm perspiration -. 12 48 .20 Smell fingers, scratch scalp or other activities bringing hand to face -. 12 7 . 20 Dryness in mouth 16 opposite to Day' 3 [ 1 ] report of RMS reporting cholinergic reactions. Underarm perspiration (4) confirms Day' 3 [3 ] specific hypothesis regarding this as a RM symptom. The less significant LM item, smell fingers, . . . hand to face (48), is less clear as to its pos- sible meanings; possibly such activity could be hypothesized to relate to the visual inattention and preoccupation with thoughts and sensa - tions attributed to LMS by Day. Also, dryness in mouth (7) is a possible adrenergic (sympathetic) reaction attributed to LMS by Day. Table 3 is a list of eye movement effects for males and for females considered separately in order of significance up to p < . 20. Table 3a for the 78 male _Ss reveals an interesting pattern of items: all of the LM items refer to some function of the alimentary canal, from oral behavior to gastrointestinal activity. The items upset stomach (19), need to urinate (38) and need to defecate (39) are all directly cholinergic (parasympathetic) stimulation. This result appears counter to Day' 3 attribution of adrenergic reactions to LMS. Diarrhea (20) may be grouped with items 19, 38, and 39 above, although it could also have some connotations of sickness for LMS. The LM item dryness in mouth (7) is again a possible adrenergic (sympathetic) reaction and hence does not fit in this dimension with the other items. Take a drink (alcohol) (50) is possibly related to Bakan' s finding that LMS are more likely to be alcoholic. 17 The RM items, in contrast, all deal either'with the body surface or with a motor function. Items 27, 4, 3 and 37 have already been discussed as RM items from Table 2. Breathing deeper (14) is a cholinergic (parasympathetic) reaction and is specifically noted by Day as a RM item. Many bodily reactions (1) seems too vague an item to be interpreted along any dimension mentioned. Table 3b for the 70 female _S_s does not show the same interior vs. exterior dimension so evident for males. The LM item difficulty in talking(22) runs counter to the expectation of the hypothesis that left-hemisphere disturbance impairs RMS speech more. Perhaps this is an indication that females have greater integration of speech function than do males. Item 48 has already been discussed from Table 2 as a LM item. Items 3 and 44 have already been discussed as RM items from Table 2. Use a tranquilizer (51) appears to be a RM item for females since 8 of 33 female RMS report some use and only 4 of 37 female LMS report some use. Aware of change in breathing (13) is unclear as to the direction of the changeand hence not readily interpretable. Sex Differences Table 4is a list of sex effects over all 51 variables for all 148 _S_s in order of significance up to p < .20. The primary effect 18 TABLE 3a. -- Eye movement effects in order -of significance up to p < . 20 for the 78 male _S_s; point-biserial correlations are given r 1122111 p < LM Items (Male) RM Items (Male) .25 27 .05 Muscle tics or twitches -. 25 20 . 05 Diarrhea .21 4 . 10 Underarm perspiration -. 20 7 . 10 Dryness in mouth -. 18 19 . 20 Upset stomach . 18 3 .20 Hands cold -. 18 50 .20 Take a drink (alcohol) . 17 37 . 20 Inability to speak coherently . 16 14 . 20 Breathing deeper . 16 1 . 20 Many bodily reactions -. 15 38 .20 Need to urinate -. 15 39 . 20 Need to defecate TABLE 3b. -- Eye movement effects in order of significance up to p < . 20 for the 70 female Es; point-biserial correla- tions are given Item r No p < LM Items (Female) RM Items (Female) -. 21 22 . 10 Difficulty in talking .20 51 . 10 Use a tranquilizer -. 19 48 .20 Small fingers, etc. , bringing hand to face . 18 3 . 20 Hands cold . 18 44 . 20 Cold feet . 16 13 . 20 Aware of change in breathing 19 evident here is the large number of items significant for females as Opposed to the small number for males. This appears to support the idea from Table 1 that there may be a sex -linked response set in operation with this data. It is perhaps more socially acceptable for females to report more symptoms more often and for males to report themselves as staying "cool. " Additional support for this hypothesis comes from the fact that relaxation (34) is the only item reported significantly more often for males with p < . 10. Results of Cluster and Factor Analyses One cluster analysis and one factor analysis solution are included here, followed by analysis of sex and LM-RM differences over the groups provided primarily by the factor analysis solution. An Oblique Multiple Groups Solution The computer program PACKAGE by Hunter and Cohen [ 12] was used to cluster analyze the data for all 148 _S_s. Basically the program computes a correlation matrix and offers several options, e. g. , blind ordering procedures, computations of communalities, multiple grouping of variables, etc. , enabling one to cluster analyze the data. Initial runs showed a large subset of the variables forming a positive manifold of high correlation coefficients indicative of a strong general factor in the data. The centroid of this manifold was 20 TABLE 4. -- Sex effects over all 51 variables for all 148 _S_s, in order of significance up to p < .20; point-biserial correlations are given r Igim p < Male Items Female Items . 33 3 .05 Hands cold . 32 19 . 05 Upset stomach . 27 2 . 05 Face hot or flushed . 26 10 . 05 Headache .23 29 . 05 Shivering .23 45 . 05 General restlessness .20 44 .05 Cold feet .20 46 .05 Nervous stomach . 19 33 . 05 Fainting . 19 21 .05 Sinking feeling in stomach . 17 35 . 05 Dizziness . 15 30 . 10 Nausea . 15 36 . 10 Hand shaking or tremor . 15 28 . 10 Face gets pale . 15 24 . 10 Pick on skin around fingernails -. 14 34 . 10 Relaxation . 14 37 . 10 Inability to speak coherently . 13 20 .2 0 Diarrhea . 12 42 .20 Neck and shoulders tight or rigid . 11 18 . 20 Choked up feeling -. 11 47 .20 Pick on skin or pimples . 11 41 . 20 Move around more -. 11 5 .20 Much sweating 21 partialed out of the matrix to make the factor structure clearer. This method is equivalent to remOving the effect of the general factor, which might be called ”general anxiety. " Table 5 shows the oblique multiple groups solution resulting after this step. There are eleven group factors that group very closely items of similar content; this shows a high degree of reliability in the data due to item content. The content of most of the eleven groups is self -evident except for group 8, which combines oral and eliminatory symptoms with the general malaise connotation of such items as vomiting (31), dizziness (35), headache (10) and tired (43). A Varimax Rotation to Five’Factors Solution The data were further analyzed with a computer routine which computes a principal components solution and then rotates using both Quartimax and Varimax methods with several options of criteria for stopping the factoring. These computer runs were made with the original 51 variables plus one for sex and one for eye move- ment. Several of these combinations of options produced sets of factors in many ways similar to the oblique multiple groups solution in Table 5. In particular one standard and widely accepted combination of factor analysis is the Varimax rotation of the principal components 22 TABLE 5. -- Final oblique multiple groups solution after positive manifold has been partialed out ‘r Group Item No. No. Item Name 1 21 Sinking feeling in stomach 46 Nervous stomach 19 Upset stomach 30 Nausea 2 13 Aware of change in breathing 16 Breathing more rapid 15 Breathing shallower 3 22 Difficulty in talking 37 Inability to speak coherently 4 12 Heart beats faster 11 Aware of heartbeat 2 Face hot or flushed 17 Blood rushes to head 5 47 Pick on SKlll or pimples 48 Smell fingers, scratch scalp or other activities bring - ing hand to face 24 Pick on skin around fingernails 23 Bite fingernails 6 25 Tension in jaw 26 Bite down on teeth 7 42 Neck and shoulders feel tight or rigid Muscle tension 23 TABLE 5. —-Continued G123?" 113:? Item Name 8 50 Take a drink (alcohol) 39 Need to defecate 38 Need to urinate 31 Vomiting 33 Fainting 49 Smoke cigarette 51 Use a tranquilizer 35 Dizziness 10 Headache 43 Tired 9 5 Much sweating 4 Under arm perspiration 6 Sweat on palms 10 8 Lump in throat 7 Dryness in mouth 11 3 Hands cold 44 Cold feet 24 solution with a Kiel -Wrigley criterion for stopping the factoring. This results in the set of factors I through V shown in Table 6. This set is particularly interesting as it effectively groups the variables by content into separate areas and functions of the body. Factors VI through VIII are specific factors from the oblique multiple groups solution and are listed because of their independence of con- tent to the first five factors. The eight groups in Table 6 were used to test for LM-RM differences and for sex differences using the original data. The first five groups were used since they more nearly represent an orthogonal factor solution than the previous oblique solution. The results of the analysis over these groups should give approximately the same pattern of results as the analysis over the 51 variables. A list of the eight factors in the factor analysis solution is given in Appendix E along with a table of intercorrelations of each of these eight factors with the other factors and with the three variables sex, eye‘movement and the product interaction of sex and eye move- ment. Included at the bottom are the standard score coefficient alphas or the reliabilities of the eight factor groups. This Appendix has been included for the interest of the more quantitatively oriented reader. 25 TABLE 6. -- Factor analysis, Varimax rotation to five factors, Kiel —Wrigley criterion, factors VI-VIII from previous multiple groups solution; factor loadings are given Factor Factor Item Item Name Factor Name No. Loadmgs No. I . 74 46 Nervous stomach Visceral (stomach) . 64 21 Sinking feeling in reactivity stomach . 64 19 Upset stomach . 58 30 Nausea . 56 34 Relaxation (reflected) II -. 70 50 Take a drink Lower‘visceral (alcohol) and oral -. 67 39 Need to defecate reactivity -. 46 38 Need to urinate (alimentary -. 55 20 Diarrhea canal) and -. 63 31 Vomiting malaise -. 50 35 Dizziness -. 40 33 Fainting -. 38 51 Use a tranquilizer -. 49 49 Smoke cigarette III . 70 11 Aware of heartbeat Upper visceral . 70 12 Heartbeat faster (heart) and . 68 16 Breathing more rapid respiratory . 67 13 Aware of change in reactivity breathing . 64 15 Breathing shallower . 51 14 Breathing deeper IV -. 62 40 Inability to concen- Muscle tension trate -. 63 26 Bite down on teeth -. 58 25 Tension in jaw -. 52 41 Move around more -. 46 42 Neck and shoulders feel tight or rigid -. 35 9 Muscle tension TABLE 6. —-Continued 26 w Fiztor LESIIOIES 113:? Item Name Factor Name V . 52 23 Bite fingernails Motor/hand to . 50 47 Pick on skin or face pimples . 50 24 Pick on skin around fingernails . 42 48 Smell fingers, scratch scalp or other activities bringing hand to face VI 5 Much sweating Sweat 4 Under arm per- spiration 6 Sweat on palms VII 22 Difficulty in talking Speech disruption 37 Inability to speak coherently VIII 3 Cold hands Sympathetic 44 Cold feet (adrenergic) reaction 27 LM - RM Differences Table 7 is a list of eye movement effects in order of significance up to p < . 50 over the eight factors listed in Table 6 over all 148 _S_s. TABLE 7. -- Eye movement effects in order of significance up to p < . 50 for all 148 _S_s; point-biserial correlations are given r ngtor p < LM Items RM Items .23 VIII .05 Sympathetic (adrener- gic) reaction -. 13 V . 20 Motor/hand to face .09 III . 50 Upper visceral (heart) and respiratory reactivity Factors VIII and V, sympathetic reaction for RMS and motor/hand to face for LMS, are the same as conclusions made following Table 2. The additional RM factor III is less significant and confusing since it contains both adrenergic (sympathetic) and cholinergic (para- sympathetic) symptoms. From the original data each of the six items in this factorwere reported more often by RMS of both sexes. This tends to show the weakness of the sympathetic -parasympathetic dichotomy for this factor over all SS. 28 Table 8a is a list of eye movement effects in order of significance up to p < .50 for the 78 male _S_s considered separately over the eight group factors. These effects have all been described before in part; the lower visceral and oral factor (II) as a LM factor and the sympathetic reaction factor (VIII) and speech disrup- tion factor (VII) as RM factors from Table 3a; and the motor/hand to face factor (V) as a LM factor from Table 2. Table 8b is a list of eye movement effects in order of significance up to p < . 50 for the 70 female is considered separately over the eight group factors. The effects of the sympathetic reaction factor (VIII) and the motor/hand to face factor (V) as RM and LM factors respectively have been discussed following Table 3b. There was also some indication from Table 3b that aware of change in breathing (13) was a female RM item; the result in Table 8b has the entire factor upper visceral (heart) and respiratory reactivity (III) as a female RM factor. From the original data all items in this factor are reported more often by RMS of both sexes. Females report items 11, 12 and 16 more often; these items all relate to an increase in heart and breathing rates. This does not support Day' s attribution of such reactivity to LMS. 29 TABLE 8a. --Eye movement effects in order of significance up to p < . 50 for the 78 male gs; point-biserial correlations are given r F13?” p < LM Items (Male) RM Items (Male) . 17 VIII . 20 Sympathetic (adrener- gic) reaction - 13 II . 50 Lower visceral and .oral reactivity (alimentary canal) and malaise . 11 VII . 50 Speech disruption -.09 V . 50 Motor/hand to face TABLE 8b. --Eye movement effects in order of significance up to p < . 50 for the 70 female _S_s; point-biserial correlations are given Factor r No p < LM Items (Female) RM Items (Female) .23 VIII . 10 Sympathetic (adrener- gic) reaction . 13 III . 50 Upper visceral (heart) and respiratory reactivity -. 13 V . 50 Motor/ hand to face 30 Sex Differences Table 9 is a list of sex effects in order of significance up to p < . 50 for all 148 _S_s over the eight group factors. That females tend to report more symptoms more often than males is in evidence here. Only on the sweat (VI) factor at p < . 50 do males report more often. And again it is likely that sweating is less socially desirable and admittable for females than for males. Hunter [16.] has suggested that males may have more sweat glands, especially under the arms where it is noticed more. TABLE 9. --Sex effects in order of significance up to p < . 50 for all 148 Es; point-biserial correlations are given r Fggtor p < Male Items Female Items . 38 VIII . 05 Sympathetic (adrener - gic) reaction . 30 I . 05 Visceral (stomach) reactivity . 13 VII . 20 Speech disruption . 13 IV .20 Muscle tension . 11 II . 50 Lower visceral and oral reactivity (alimentary canal) and malaise -. 06 VI . 50 Sweat DISCUSSION General Response Tendencies The significant Chi -square in Table 1b shows that women report symptoms of anxiety more often than men. One possible explanation is that women may be more anxious than men. On the other hand, women may simply be more honest in their reporting (or in their memory) of reactions that reflect "fear under‘stress. " The results in Tables 4 and 9 that males admit only to relaxation (34), pick on skin or pimples (47), much sweating (5) and, at p < . 50, to sweating (VI) more often than females add support to a socially acceptable response set hypothesis: women are supposed to be emotional creatures and can report emotional reactions but men are supposed to be unemotional and cannot admit them. Thus prevailing social mores may act to decrease male awareness but certainly the reporting of emotional reactions. 31 32 Discussion of Analysis Over All 51 Variables Throughout the discussion, it should be noted that the correlations quoted are on the order of . 15, . 20 and less often . 30. The discussion is to be weighed by the reader in this light. Although the variables of sex and eye movement are relatively weak pre- dictors of performance over this set of anxiety items, the resulting patterns are significantly interesting to add support to the general theses on which this work is based. LM-RM Differences for Males and Females Combined The results for RMS support previous hypotheses and observations by Day and Bakan except for the cold hands and feet cold items. These latter items contradict Day' s easy dichotomy of LM-sympathetic (adrenergic) symptoms and RM—parasympathetic (cholinergic) symptoms. What does emerge is a picture of RMS having more muscle tics and twitches perhaps related to greater muscle tension, increased underarm perspiration, cold hands and cold feet. Coherent speech is disrupted also, presumably because of some disturbance in the left hemisphere. These symptoms all relate to the body surface: to the-skin, the muscle sheath, selected sweat glands and to the motor function of speech. 33 Of the two LM items, dryness in mouth (7) supports Day' 3 ascription of sympathetic (adrenergic) autonomic reactions to LMS. However, the implications of smell fingers, . . . hand to face (48) are not clear. It is a motor function, it affects the skin surface and may be linked to Day' 3 observation that LMS are unable to maintain external visual attention because they tend to be obsessional about emerging impulses and become preoccupied with thoughts and sensa- tions. LM-RM Differences for Males Considered Separately The RM items for males are nearly the same as for RMS over all _S_s; they all are reactions at the body surface or are motor functions. Even the additional breathing deeper (14) item is closely associated with the body' s muscle sheath. The LM items for males in striking contrast all relate directly to the functioning of the alimentary canal, represent pri- marily parasympathetic (cholinergic) reactivity and refer to functions interior to the body. This result does not support Day' s observation that LMS report adrenergic (sympathetic) sensations. The LM items in contrast to the RM items strongly suggest an inner vs. outer awareness dimension for LM and RM males respectively. 34 Fisher and Cleveland [ 13] have found that males with definite body image boundaries experience greater physiological reactivity at the body surface and lesser internal (heart response) activity. Males with Rorschach scores indicating indefinite body image boundaries showed an opposite pattern. Thus these results suggest that male LMS might be found to have less definite body image boundaries than male RMS. Fisher [13] replicated this study using female adolescent _S_s and found the results to be in exactly the same directions. These resulting patterns of reactivity for males and females in both studies were very significant under conditions of high emotional arousal and only borderline during periods of rest. Fisher [14] discusses the need for alternatives to describ- ing physiological reactivity in terms of a sympathetic -parasympathetic dimension. His preference is for an inner-vs. outer dimension based upon ratings from Rorschach protocols of "barrier" vs. "penetration of boundary" scores. Persons who visualize their boundaries as thick and armored manifest a higher level of aspira- tion, more drive toward self -expression and more motivation for competitive advancement than do persons with indefinite body -image boundaries. Thus from the results in Table 3a male RMS should show these high drive characteristics and male LMS would not. 35 LM-RM Differences for Females Considered Separately The same inner vs. outer reaction dichotomy is not evident in the results for females in Table 3b, or at least not along the CLEM dimension. The cold hands and feet, and the aware of change in breathing items are all associated with the outer body layers, the skin and muscle sheath. The smell fingers, . . . hand to face item has been previously discussed as a LM item. Use of a tranquilizer was reported by 8 of 33 female RMS and by 4 of 37 female LMS. Bakan' s suggestion [7 ] that RMS are more tense might be linked to a greater use of tranquilizers by RM females. Difficulty in talking (22) is a LM item for females only. The parallel content item inability to speak coherently (37 ) is a RM item for males and for all _Ss considered together. From the fact that the left hemisphere is generally dominant for verbal activity, it seems to follow that RMS would be affected more by any disturbance in that hemisphere. That females and especially LM females report disturbance in speech may be evidence of greater integration of and/ or less definitely located speech centers in females. Sex Differences If there are any differences in emotional reactivity to anxiety or stress between males and females other than those 36 previously discussed in the subsection General Response Tendencies, they have been obscured in this data. Discussion of Cluster and Factor Analyses The Oblique Multiple Groups Solution The cluster analysis in Table 5 contains 38 of the 51 vari- ables in the study and shows that the data has a high degree of reliability with respect to the content of the items. The eleven groups represent the following bodily reactions: group 1-—all un- comfortable stomach sensations; group 2 --all involve an increase in respiratory activity; group 3 --both involve a disturbance of speech; group 4--a11 involve an increase in heart rate and blood rising to the face and head; group 5 ~—all involve bringing the hand to the face or mouth; group 6--both are tension in the jaw; group 7 -- both are muscle tension items; group 8--are items that involve both oral and eliminatory system symptoms plus items with a general connotation of malaise, i. e. , vomiting (31), dizziness (35), head- ache (10) and tired (43); group 9--all involve increased sweating; group 10--both are indicative of a general sympathetic (adrenergic) reaction in the mouth and throat; and group 11 --both involve the sympathetic (adrenergic) reaction of peripheral vasoconstriction. 37 The Varimax Rotation to Five Factors Solution The factor analysis in Table 6 reorders the variables into five groups which reflect fairly well—defined body locations and/or related body functions. Again, this shows high content reliability of the items. The three factors VI, VII and VIII are specific factors from the cluster analysis in Table 5 and were included for analysis because their content is fairly independent of the first five factors. Factor I is essentially the same upset stomach group as group 1 from Table 5. Factor 11 is essentially the same as group 8, . lower visceral and oral reactivity with malaise. Factor III com- bines group 2, breathing rate increase items, with part of group 4, increase in heart rate items, plus an additional breathing rate item, breathing deeper (14), a decrease in breathing rate item. Perhaps awareness of this area or function of the body is more salient to reporting symptoms there than a sympathetic -parasympathetic awareness dimension. Factor IV combines groups 6 and 7, tension in jaw, shoulders and in general, and two items indicating restless - ness, inability to concentrate (40) and move around more (41). Factor V is identical to group 5, a motor/ hand to face group. From the original hypotheses regarding LM-RM differences, one might expect factors 111, V and VIII to be reported by LMS, and factors I, II, IV, VI and VII to be reported by RMS. 38 Lorr Still. [15] factor analyzed personality ratings of outpatients in psychotherapy and found three clusters of body com- plaints. The first was a factor of "endodermal dysfunction" including the upper and lower portions of the endodermal tube (the alimentary canal as previously referred to in this study) and gastrointestinal symptoms. The second cluster'was of "mesodermal origin" includ- ing anergic, respiratory and cardiovascular-symptoms. The third cluster‘was of "cerebrotonic order" including primarily skin com- plaints. These symptoms occur in a layer—like fashion. Lorr' s first group is very similar to a combination of factors I and II from Table 6. His second group seems to parallel factor 111, and his third group might include parts of factors V, VI and VIII which are all functions occurring in or associated with the skin. Thus, the results presented in Table 6 appear to add sup- port for a hypothesis of differential awareness of physiological reactivity in different layers of the body. RM - LM Differences The picture of LMS from Table 7 is sparse, only showing the factor motor/hand to face (V). The sympathetic reaction factor VIII and the upper visceral factor III attributed to LMS by Day are reported more often by RMS. Factors 1, II, IV, VI and VII do not 39 even tend to be significant considered over all 148 _S_s. The most significant conclusion appears to be that RMS experience the sympathetic reaction of vasoconstriction at the skin' 3 surface and increased heart rate and breathing rate. The results in Table 8a for male SS parallel the results already discussed in Table 2. Male LMS report lower visceral and oral reactivity with malaise (generally cholinergic reactivity) and motor/hand to face factors 11 and V. Male RMS report the sympa- thetic reaction (VIII) and the speech disruption (VII) factors. Nothing new is added by this analysis. The results in Table 8b for females show only the motor/ hand to face (V) factor for female LMS, and the sympathetic reaction (VIII) and upper visceral (heart) and respiratory reactivity (III) factors for female RMS. The only addition to what has been dis- cussed before is the inclusion of the entire upper visceral factor for female RMS. This result is counter to Day' 3 observation that LMS report sympathetic (adrenergic) reactivity. Sex Differences Table 9 shows that females report nearly all symptoms more often than males. Only the upper visceral (III) and motor/ hand to face (V) factors do not tend to be answered more often by females. From 40 the original data the items in factor 111 are split between males and females; and for factor V females answer the "fingernail" items (23 and 24) more often and males the "pick on skin" and "hand to face" items (47 and 48) more often. Problems and Suggestions for Further Study [The problems with self -report data are worth noting here. The _Ss must report their own awareness of physical reactions. Thus the degree of awareness is a variable that could have the effect of obscuring the results. Also _S_ response set is not explicitly con- trolled for. Since the results do show a possible sex linked response set, it may be necessary to obtain instrumental measures of physio- logical reactions to circumvent the problems of awareness and response set. Another problem is that the basal levels of reactivity for each_S_are not measured. The effects of this lack may be mitigated if we make the following assumptions. If a S has alow basal level of any symptom, his report of that symptom should indicate its presence in his awareness and be relatively good data. If a _S_ has a high basal level of any symptom, then any report of same under stress‘would tend to underrate and obscure the effect of that symptom in the data. If we can make the assumption that the great majority 41 of these 51 symptoms or reactions to stress are not experienced chronically by most of the _S_ population of students (recall the over- all average was 3. 52, between sometimes and rarely), and are experiencedin recognizable states of anxiety or stress, then their data will be relatively good. Most of the effect of high basal level of anxiety Es will be to obscure the differences in the data. This also suggests one way to account for basal reactivity level by getting an independent measure of anxiety proneness. A second method-might be to ask _S_s to rate both their basal or rest level of eachof these symptoms and the level in stress situations on the same scale and analyze the differences. A direct way around most of the pitfalls of self -report data would be to make instrumental measurements of selected physio- logical reactions both before and during experimentally derived stress situations. Summary There do appear to be differences in reactions to anxiety or stress when the population is divided on the basis of right vs. left conjugatelateral eye movement. Some of Day' 8 observations have been supported and some not supported. A LM-adrenergic (sympathetic) and RM-cholinergic (parasympathetic) symptom report 42 dichotomy does not hold up in the data presented; in fact a case might be made in part for the opposite association of RMS with adrenergic reactions and LMS with cholinergic reactions. An inner vs. outer dichotomy of reactivity is apparent for males. Male LMS report symptoms pertaining to the alimentary canal and .to general parasympathetic (cholinergic) reactivity. Male RMS report reactions in the body' s outer layers, the muscle sheath and the skin surface. The literature suggests that male LMS may have an indefinite body image as opposed to RMS and that male RMS have greater competitive and achievement drives. Many specific observations of both Day and Bakan are sup- ported, primarily by the data for males, specifically male RMS have'more muscle tics, underarm perspiration, difficulty speaking anda tendency to breathe deeper (which as a cholinergic reaction supports Day), and male LMS are more likely to be alcoholic. The hypothesis of increased heart and breathing rates for LMS is not supported, since female RMS report it. The hypothesis that RMS have a slowing and strengthening of heart and breathing rates gets some support since male RMS report the item breathing deeper. The hypothesis that LMS are preoccupied with obsessive intrusions is supported by the fact that LMS over all _S_s and female LMS report the motor/hand to face item(s). RM concern with body outer layers 43 supports the hypothesis of high muscle tonus in RMS. The speech disruption hypothesis is supported by the data over all _S_s and for male RMS; the exception is the difficulty in talking item reported by female LMS. This result, however, also tends to support earlier findings that suggest greater hemispheric integration for females. The cluster and factor analyses produced groups of symp- toms that apply to specific locations on and/ or functions of the body and tend to confirm other reports in the literaturethat an inner vs. outer or a body layer dimension for the investigation of physiological reactivity may be meaningful. Perhaps the most outstanding result from this study is that given the low level of correlation between the variables sex and eye movement and the set of anxiety symptoms, and the relatively low predictive power this implies, the pattern of responses is as striking as it is. Problems with self-report data were discussed and sug- gestions made for further study. LIST OF REFERENCES 10. LIST OF REFERENCES Day, M. E. An eye movement phenomenon relating to attention, thought and anxiety. Perceptual and Motor Skills, 1964, 19, 443 -446. Libby, W. L. , Jr. Eye contact and direction of looking as stable individual differences. Journal of Experimental Research in Personality, 1970, 4, 303 -312. Day, M. E. An eye -movement indicator of type and level of anxiety: some clinical observations. Journal of Clinical Psychology, 1967, 23, 438-441. Duke, J. D. Lateral eye movement behavior. Journal of General Psychology, 1968, 78, 189-195. Day, M. E. An eye -movement indicator of individual differences in the physiological organization of attentional processes and anxiety. Journal of Psychology, 1967, 66, 51—62. Bakan, P. Hypnotizability, laterality of eye -movements and functional brain asymmetry. Perceptual and Motor Skills, 1969, 28, No. 3, 927-932. Bakan, P. The eyes have it. Psychology Today, April 1971, 4, No. 11, 64. Bakan, P. , and Svorad, D. Resting EEG alpha and asymmetry of reflective lateral eye movements. Nature, 1969, 223, 975-976. Strayer, F. Cited by Bakan in Reference [7 ] . An unpublished study done at Simon Fraser University, Vancouver. Bakan, P. Personal communication, April, 1971. 44 11. 12. 13. 14. 15. 16. 45 Sperry, R. W. Hemisphere deconnection and unity in conscious awareness. American Psychologist, 1968, 23, 723 -7 33. Hunter, J. , and Cohen, S. PACKAGE, A computer library tape program available at The Michigan State University Computer Center. Fisher, S. , and Cleveland, S. E. Body Image and Personality. 2d Rev. Ed. New York: Dover Publications, 1968, 448 p. Fisher, S. , and Cleveland, S. E. An approach to physiological reactivity in terms of a body -image schema. Psychological Review, 1957, 64, 26-37. Lorr, M. , et al. A factor analysis of personality ratings of outpatients in psychotherapy. Journal of Abnormal and Social Psychology, 1953, 48, 507 -514. Hunter, J. Personal communication, May, 1971. APPENDICES APPENDIX A PROV ERBS USED TO DETERMINE CLEM 1. The hardest work is to go idle. 2. A rolling stone gathers no moss. 3. A watched pot never boils. 4. Better a good enemy, than a bad friend. 5. If you can' t bite, then don' t show your teeth. 6. A poorworker blames his tools. 7. He that lies on the ground cannot fall. 8. Better a bad peace than a good war. 9. They who are mute-want to talk most. 10. What saddens a wise man, gladdens a fool. 46 IAITPTHNIHI(]3 Body Feeling Inventory Below you will find a list of reactions to stress. This is n.study of how often people have these reactions in stressful situations. Examples of stressful situations are: a) before an important interview b) before talking in front of a group c) waiting in doctor's or dentist's office d) taking an important exam e) being afraid or anxious etc. Each reaction is to be rated on sis-point scale on the basis of how often :33 experience the reaction in a stress situation. The five spaces on the IBM answer sheet correspond to the 5 points of the scale shown below: 1 2 3 l; 5 Always Often Sometimes Rarely Never Thus if you experience a reaction often you would fill in the number 2 space on the answer sheet. If you never experience the reaction you would fill in the number 5 space on the answer sheet. You may use any number from I to 5 in answering. Consider each of the reactions independently of the others in responding. 1. Aware of many bodily reactions. 2; Face hot or flushed. 3. Hands cold. h. Under arm perspiration. 5. Much sweating. 6. Sweat on palms. 7. Dryness in mouth. 8. Lump in throat. 9. Muscle tension. 10. Headache. (please go to next page; the scale is repeated at top of page) 47 Scale 48 '12_3 h 5 Always. Oftenh Sometimes Rarely Never 31..Aware of heartbeat. 12$ Heart beats faster. l3. Aware of change in breathing. 1h. Breathing deeper. 15. Breathing shallower; 16. Breathing more rapid. 17. Blood rushes to head. 18. Choked up feeling. I9. Upset stomach. 20. Diarrhea. 21. Sinking feeling in stomach. 22. Difficulty in talking. 23. Bite fingernails. 2h. Pick on skin around fingernails. 25. Tension in jaw. 26. Bite down on teeth. 27. Muscle tics or twitches. 28. Face gets pale.; 29. Shivering. 30. Nausea 31. Vomiting. 32. Increased salivation (mouth waters) (Please go to next page; the scale is repeated at top of page) ?’ .. . 1! 9'3 Scale '1” ' 2 3 h 5 Always Often Sometimes Rarely' Never ## 33. Fainting 3h. Relaxation. 35. Dizziness. 36. Hand shaking or tremor. L0. L1. h2. h3. an. as. h6. h7. I48. h9. Inability to speak coherently. Need to empty bladder (urinate). Need to empty rectum (defecate). Inability to concentrate. Move around more. Neck and shoulders feel tight or rigid. Tired. Cold feet. General restlessness. Nervous stomach. Pick on skin or pimples. Smell fingers, scratch scalp or other activities bringing hand to face.-. Smoke cigarette. 50 . Take a drink (alcohol). 51. Use a tranquillizer. APPENDIX C AVERAGE RESPONSE SCORES OVER ALL 51 VARIABLES FOR THE FOUR MAIN SUBJECT GROUPS Subject Group Variable M-LM M-RM F-LM F-RM 1 3.17 2.95 2.84 2.97 2 3.54 3.57 3.08 3.00 3 3.90 3.54 3.19 2.82 4 2.71 2.38 2.54 2.42 5 3.17 3.32 3.51 3.36 6 2.80 2.97 2.95 3.06 7 3.24 3.65 3.51 3.55 8 3.83 3.78 3.62 3.64 9 3.44 3.49 3.32 3.30 10 3.93 3.92 3.38 3.42 11 3.17 3.08 3.16 3.00 12 2.93 2.89 2.95 2.67 13 3.39 3.46 3.73 3.42 14 3.68 3.41 3.68 3.67 15 3.71 3.78 3.78 3.67 16 3.49 3.43 3.49 3.27 17 3.95 3.97 3.92 3.88 18 3.83 3.84 3.54 3.70 19 3.63 3.97 3.05 3.12 50 51 Subject Group Variable M-LM M-RM F-LM F-RM 20 4.24 4.59 4.24 4.15 21 3.51 3.57 3.08 3.15 22 3.44 3.38 3.05 3.48 23 3.44 3.76 3.46 3.58 24 3.56 3.78 3.14 3.42 25 4.00 3.95 4.00 4.00 26 4.00 3.78 3.81 3.67 27 3.98 3.51 3.86 3.64 28 4.10 4.03 3.86 3.76 29 4.20 4.16 3.78 3.73 30 4.24 4.30 3.97 4.03 31 4.71 4.73 4.68 4.70 32 4.12 3.89 4.00 4.06 33 4.93 4.86 4.68 4.76 34 2.95 3.22 3.35 3.36 35 4.27 4.27 3.86 4.09 36 3.83 3.73 3.57 3.39 37 4.17 3.84 3.84 3.61 38 3.27 3.57 3.51 3.27 39 3.73 4.00 3.97 3.97 40 3.15 3.14 3.00 3.00 41 2.78 2.76 2.51 2.64 42 3.39 3.22 3.11 3.03 43 3.24 3.41 3.32 3.33 52 Subject Group Variable M-LM M-RM F-LM F-RM 44 4.05 3. 81 3.70 3.24 45 2.80 2.92 2.46 2.45 46 3.07 3.32 2.86 2.67 47 3.41 3.43 3.62 3.70 48 3.27 3.38 3.11 3.52 49 4. 56 4. 38 4. 51 4. 52 50 4. l7 4. 49 4. 43 4. 30 51 4. 80 4. 92 4. 86 4. 64 APPENDIX D TABLE OF CORRELATIONS OVER ALL 148 SS BETWEEN THE 51 ANXIETY VARIABLES, THE 8 GROI'J'P FACTORS FROM THE FACTOR ANALYSIS SOLUTION AND THE VARIABLES SEX, EYE MOVEMENT AND THE PRODUCT INTERACTION OF SEX AND EYE MOV EMENT* 3 ‘53 £5 '3 Factor E Id g 'g c x a) “’ I II III IV V VI VII VIH " g» g " g > :3 m 2 fl 2 46 85 20 33 43 28 36 42 43 20 - 2 - 9 21 7o 15 24 43 24 18 47 41 19 - 3 1 19 73 46 24 39 25 21 32 33 32 -1o - 5 30 59 39 24 39 29 31 31 4o 15 - 3 1 34 46 -16 21 3o 5 32 32 12...- 14 7 7 50 -1o 61 8 - o 17 - 8 -19 7 - 3 .- 6 -13 39 11 58 9 - 2 21 8 - 1 6 - 6 - 6 - 8 38 27 47 18 17 19 17 9 20 1 - 2 -13 20 4o 52 6 11 15 2 18 18 13 - 9 -13 31 17 55 9 22 16 - o 5 32 3 - 2 o 35 29 53 36 30 25 16 22 47 17 — 6 7 33 12 42 22 20 — o - 2 9 25 19 - 1 9 51 19 35 4 4 - 4 - 1 - 1 6 9 4 -14 49 - 5 43 2 -18 16 - 8 - 7 9 - 2 5 5 11 36 12 65 30 17 34 31 4o 2 6 - 1 12 38 13 66 31 15 35 41 37 5 7 - 5 13 27 23 78 34 17 24 35 38 - 8 6 -10 14 ‘11 25 57 26 9 13 20 31 - 7 8 8 15 20 12 56 39 33 4o 31 53 1 1 - 5 . 16 19 18 76 36 12 .24 32 45 4 7 — 4 4o 36 20 20 56 36 13 35 18 8 o 1 26 27 — 2 16 63 19 14 25 25 8 9 3 25 37 - 6 28 63 26 21 24 28 - 1 2 1 41 15 5 36 49 30 9 33 33 11 - 3 5 53 54 a) E" m E: 3 Factor 0 £3 0 as E E --4 o x o 3 >4 “9 ‘5 >4 8 > I II III IV V VI VII VIII 8‘3 £2 (32 42 45 20 40 62 20 32 32 39 12 7 4 9 45 27 32 56 20 32 33 45 7 - 1 - -1 23 16 11 8 12 54 4 10 3 3 - 8 - 4 47 17 29 24 30 62 32 24 25 -11 - 2 1 48 21 19 18 33 50 11 27 6 1 -12 6 24 27 9 13 28 75 11 . 17 16 15 - 9 2 5 29 6 29 35 20 83 28 -- 2 ~11 - '1 - 9 4 25 5 18 9 8 65 25 34 4 14 7 6 28 - 0 38 25 19 52 36 39 - 6 - 7 - 2 22 46 7 34 43 21 44 82 30 7 - 8 11 .37 46 6 45 44 32 31 82 39 14 14 4 3 41 18 37 34 7 38 36 70 33 17 2 44 29 35 49 41 22 11 22 70 20 . 15 - >3 1 16 - 16 30 20 4 16 24 35 9 3 10 2 35 7 53 31 13 39 36 40 27 1 - l 7 38 18 48 34 22 32 34 30 - 5 -12 -10 8 49 20 45 44 20 27 46 32 9 1 2 .10 34 47 36 48 38 10 30 40 26 - 1 3 17 36 25 45 46 11 29 42 35 4 0 - 2 18 60 16 47 60 41 31 60 50 11 - 4 4 27 22 25 41 38 29 20 30 34 - 0 19 7 28 41 39 42 53 38 28 51 58 . 15 5 0 29 60 37 29 52 23 20 47 52 23 2 1 32 25 36 36 34 23 31 27 25 - 1 5 8 36 38 21 46 35 24 38 47 28 -15 7 - l 43 19 53 23 47 33 12 15 39 - 0 - 5 - 4 45 49 14 36 51 29 25 31 39 23 - 3 - 2 *Table entries X 10 -2 = correlations APPENDIX E TABLE OF INTERCORRELATIONS BETWEEN THE 8 GROUP FACTORS FROM THE FACTOR ANALYSIS SOLUTION AND THE VARIABLES SEX, EYE MOVEMENT AND THE PRODUCT INTERACTION OF SEX AND EYE MOVEMENT; THE STANDARD SCORE COEFFICIENT ALPHAS OR THE RELIABILITIES OF THE 8 FACTOR GROUPS* Factor Variable** III III IV V VI VIIVIIIA B c Factor or Variable 100 31 38 58 33 42 55 5O 30 II 31 100 26 18 28 5 8 38 11 III 38 26 100 49 26 43 48 61 -'1 IV 58 . 18 49 100 43 35 52 54 13 V 33 28 26 43 100 24 32 21 3 VI 42 5 43 35 24 100 45 35 - 6 VII 55 8 48 52 32 45 100 42 13 VIII 5O 38 61 54 21 35 42 100 38 A 30 11 -l 13 3 -6 13 38 100 B -3 -5 9 4 -13 3 3 23 C -1 -9 -4 3 2 -2 9 -0 H I . H 01°C WCQCJOOOACDUOD I OU'IU'I OCDNNOOACOl-I N l 01 O H O H O Standard Score Coefficient Alphas 80 74 82 75 69 70 80 64 fir—V fi—f *Table entries X 10.2 = correlations **Variable A - - Sex Variable B -- Eye Movement Variable C -- Product Interaction of Sex and Eye Movement 55 ”7177717177 71': 11777! [1, IQFMI'IHILIIIIW“