RGFWLPIM PW: NC? C3?“ '3'.‘§EE RQ2223C...ACH WKBLQTS ‘i‘itxsmi; for {+20 519951200 0? £243. A. MECHICEAN ST’ffi 3“ JNQVERSETY E22222... 325150 A? M'ém 0.2.3 r275 114E515 LIBRAR Y Michigan State University l f E5319; . . arm/“791* v bHLtU-{o 02". A numbc: for the Forge. studies sugg- erence betwec Jects. The ditions pref. tne blots an be related t \- 53‘1‘ In C O Jame bet A TEIationsh; .. level Of ("1? U\A‘ 9rCew1 of tag |Q 1... egg} DH: ““3311th w" I ABSTRACT ERCEPTUAL PREFERa-zcs OF THE RCiRSCZ-L’ACE-I INKBLO TS by Richard A. Brown A number of studies have investigated preference for the Rorschach inkblots. The results of these studies suggest that there may be differences in pref- erence between normal and unstable or neurotic sub- jects. The studies also suggest that under some con- ditions preference may be related to the complexity of the blots and under other conditions preference may be related to the sexual suggestiveness of the cards. This study was designed to investigate differences in perceptual preference between the cards, differences in preferences for cards between threatening and non-. threatening instructional sets, and differences in pref- erence between sex of the subjects. In addition, the relationships between perceptual preference and anxiety level of subjects, complexity of the cards, and sexual suggestiveness of the cards were also investigated. Perceptual preference was measured by the length of time each card was attended when the cards were presented in the standard sequence through a slot in . -‘.. “n “yd 2‘ .1 tech» “' a screel . 9 college 943.3 «4"» f- as 10.;5 as " ESLLICtiOIIF significmtl as indicate the sexual ' were signif and percept and differt sets. Perc be higher 5 threatening t:0 decreas; Richard A. Brown a screen. Under the non-threatening instructions college subjects were requested to look at each card as long as they were interested. Under the threatening instructions subjects were told, in addition, that their interest in the cards revealed their underlying per- sonality characteristics. Significant differences in perceptual preference were found between cards. These differences were not significantly related to sex of subject, a xiety level as indicated by the Taylor'Manifest Anxiety Scale, or the sexual suggestiveness of the cards. However, there were significant relationships between card complexity and perceptual preference and between card complexity and differences in preference between instructional sets. Perceptual preference of the cards was found to be higher for more complex cards, but under the more threatening instructional set this relationship tended to decrease. Approved: :39‘4? Qp~§%fi_z_‘ Date: Thesis Committee: Paul Bakan, Chairman Lucy C. Rau Terrence M. Allen 4 . p- 12F ‘ I ‘5st -~ ’JI :3 'f'] PERCEPTUAL PREFEREJCE OF HE RCRSClLACI-l INKBLOTS By Richard A. Brown A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER.OF ARTS Department of Psychology 1965 .Il" ‘rm-‘y‘ The a appreciati o ‘ .fJ'T‘. his guide. in the dev Dr. Lucy 511;; stior Stu dy. ACKNOZ'JLE G} EN T The author wishes to express his gratitude and appreciation to his adviser, Dr. PaulBakan, for his guidance in this research and for his assistance in the develOpment of the manuscript. He also wishes to express gratitude to Dr. Lucy Han and Dr. Terrence Allen for their helpful suggestions and comments during the course of this study. ii __..._—. .—_-—. ———-— ———v —~. *7: ~... mPQ-‘DIF‘; Afl'—..i-I- ' '1' than“; «1-d- LIST CF LIST OF Chapter II. III. Iv, TABLE OF LIST OF TABLE . . . . . . LIST OF APPENDICES. . . . Chapter I. INTRODUCTION . . II o I'fET}DD o o o o 0 Subjects Procedure III. RESULTS. . . . . CONTENTS Perceptual Preference of the Cards Effects of Anxiety Level Perceptual Preference and Other Measures IV. DISCUSSION . . . V. CONCLUSIONS. . . REFERENCES. . . . . . . . APPENDICES. . . . . . . . iii Table 1. LIST‘OF TABLES Means of Raw and Transformed Looking Time on the Rorschach Inkblots. . . . . Analysis of Variance for Perceptual Preference. . . . . . . . . . . . . . . Analysis of Variance for Anxiety Level. . Differences in Perceptual Preference Between Instructional Sets. . . . . . . Card Complexity: Means, Standard Deviations, and Standard Error of Means Correlates Between Perceptual Preference and.0ther Stimulus Qualities of the Rorschach Ink Blots . . . . . . . . . . Perceptual Preference of Cards: Newman-Keuls Test on All Ordered Pairs 0f Means. o o a o o o o o o o o o Stimulus Properties of the Rorschach Ink Blots: Other Measures. . . . . ... iv 22 23 24 44 "hitilbl i iii-ulna:- LIST OF APPEL‘JDICES Appendix . Page A. Perceptual Preference of Cards: Newman-Keuls Test on All Ordered Pairs of Means. . . . . 43 B. Stimulus Properties of the Rorschach Ink Blots: Other Measures. . . . . . . . . 45 Heme Interprets a rather 1' 1942). II over 2,00C "PSyCholog Cf studies number of ther defin themselves comes QUit cePt in te and subjec Altho. ambiguous, SChach pla city and 8 he Placed CHAPTER I INTPODUCTION Herman Rorschach (1884-1922) regarded his Form Interpretation Test as a psychological exPeriment with a rather incomplete theoretical foundation (Rorschach, 1942). In 1958 Baughman estimated that there had been over 2,000 reported studies concerning Rorschach's "psychological experiment" and since 1958 the quantity of studies has rapidly continued to increase. A large number of these studies have been concerned with fur- ther definition of the stimulus qualities of the blots themselves. Knowledge of these stimulus qualities be- comes quite important when evaluating a subject's per- cept in terms of the interaction between the stimulus and subject, himself. Although an important feature of the test is the ambiguous, unstructured nature of the inkblots, Ror- schach placed general requirements of symmetry, simpli- city and good composition on the blots. Furthermore, he placed specific restrictions on the stimulus proper- ties of each of the ten blots: Plate 1. Black. Failure is almost never encountered. Stimulates form and movement responses equally. Easy to interpret as a Whole and in De- tails. There are a number of small Details that frequently give rise to answers. Plate II. Black and Red. Kinaes- thesias more easily seen than in I. Contains a prominent intermediate. figure. Color is introduced and oc- casionally induces a suggestion of "color shock". The red runs into the black. Plate III. Black and Red. Kinaes- thesias.easiest to see. The.red separated from the black.~ Plate IV. Black. Form and movement answers both comparatively difficult. More difficult to interpret as a whole than in details. The figure is gen- erally considered "beautiful" but the interpretation is difficult. Plate V. Black. The easiest form to interpret.. Almost always interpreted as a "bat", or a "moth". SchiZOphrenics frequently fail to answer on this fig- ure, or they see, for example, moving peeple in it. Plate VI. Black. Generally called the most difficult of all figures. Plate VII. Black. The essential part is the white intermediate figure, a rather obvious oil lamp, rather than the black figures. This Plate presents the converse of Plate V in that normals rarely see the lamp while schiZOphrenics frequently do. Plate VIII. Multicolored. Harmonious color and form. Color shock.apparent in neurotics. Easily interpreted, at least in detail. Plate IX. JMulticolored. Discordant color and form. Kinaesthesias easily aroused. In addition, a definite intermediate figure. Plate x. Multicolored. Disparate blots. Whole answers almost impos- sible. (1942, p. 52) Studies on the stimulus values of the blots have further differentiated the qualities of the ten Ror- schach cards. Baughman (1959) and others (Cantor, 1958; Exner, 1959, 1962; Kuntz, 1964; Marks, 1961; and wallen, 1948) have used modified versions of the blots to tease out the effects of such stimulus dimensions as color, texture, form, and.mass on percepts. Symbolic meanings of the cards have been investigated by the use of Os- good's (1957) semantic differential (Davis, 1961; Kamano, 1960; Little, 1959; Loiselle and Kleinschmidt, 1963; Rabin, 1959; Rosen, 1960; Zax and Benham, 1961; and Zax and Loiselle, 1960a, 1960b) and by other methods (Charney, 1959; Hafner, 1961; Hirschstein and Rabin, 1955; Levy, 1958; Meer and Singer, 1950; Pascal et a1, 1950; Rosen, 1951; Scheifer and Hine, 1960; Shaw, 1948; Taniguchi et a1, 1958; and Zelin and Sechrest, 1963). Preference for the cards has also been investigated by a number of investigators (George, 1955; Hershenson, 1949; Mitchell, 1952; Smith, 1962; and Wallen, 1948). Although there have been a great many studies, thus far there has been no research reported on the relative "attention pulling power" or interest value of the Rorschach inkblots. It seems reasonable to assume that the studies on preference are most closely related to the relative interest of the cards as preference of stimuli seems to be closely related to their interest and "attention pulling" value (Berlyne, 1963; Gilgen and Bakan, 1965). For some preliminary speculation about perceptual interest of the cards, then, it is apprOpriate to review the literature on preference. Wallen (1948) used men at a military training station as subjects. To investigate differences in preference between normal and unstable men two groups were used: Normal recruits undergoing psychiatric screening and unstable men about to leave the training program because of psychiatric discharge. The method was to present the cards, one at a time, to each sub- ject and ask whether the design was liked or disliked. To investigate the effect of color, achromatic versions of the cards were presented to some subjects and stan- dard versions to others. It was found that cards 11, VI, and IX were disliked more in unstable men. Although unstable men preferred achromatic versions of the cards the only card of the three where color was important in differentiating the groups was card II. Nallen's explanation was that in card II the red was important because it was associated with blood and aggression and, therefore, was not preferred by unstable subjects. Hershenson (1949) asked male and female high school juniors to participate in a research project and made it clear to her subjects that the purpose was not to provide information about them personally. After each subject had been administered the Rorschach they were asked to rank the cards in order of preference. She found that although there were no highly significant differences in preference between sexes there was a ten- dency for card II to be more disliked by female subjects. This difference might be attributed to a female defensive reaction to the red on the card as "blood". Using groups of psychotics and'psychoneurotics from the clinic pOpulation of a Veteran's Administration HOS- pital and a group of non-patients, Mitchell (1952) admin- istered the Rorschach and then asked each subject which of the cards was liked best and which of the cards was liked least. The results confirm Wallen's (1948) find- ings about card IX which was liked more by the normal than the psychoneurotic subjects. In addition it was found that card VII was liked less by normals than the psychoneurotic group. Also, card V had a tendency to be preferred more by subjects with less intelligence and less education. In 1953 George pointed out an interesting rela- tionship between the preference of Wallen's (1948) subjects and the sexual suggestiveness of the cards as determined by the number of sex associations pro- duced for each card by male college students (Shaw, 1948). A negative correlation between card preference and sex associations suggests the Operation of some kind of defense mechanism affecting card preference. George (1955) later ran his own study in card preference. Utilizing college students as subjects he asked them to choose their favorite card after the usual Rorschach administration. He then ranked their preferences and compared them to ranked data from a number of other sources including Hershenson's (1949) card preference, Wallen's (1948) card preference, Shaw's (1948) sexual suggestiveness, a similar study on sexual suggestiveness by Pascal et a1. (1950), and Beck's (1948) Z values for the difficulty of organizing whole reSponses to each of the cards. The Z values were used as an index of card complexity, equating degree of complexity with the difficulty of producing whole responses to the cards. The George and the Hershenson ranked preferences correlated highly with each other and with complexity as indicated by the Z values. However, none of these was found to be very highly related to the preference of wallen's subjects or to the sexual suggestiveness of the cards. George eXplained the difference in preference of Wallen's group by the lower intelligence of the military groups. Citing Mitchell's finding concerning card V, he hypo- thesized that subjects of lower intelligence and edu- cational level prefer cards that are less complex. Although it seems to be true that intelligence affects the preference of complex stimuli (Hoats et al., 1963) another explanation of the difference is also possible. That is the difference in set between the groups of subjects. The college groups were simply participating in "research projects" while the men undergoing psychia- tric examination were in a somewhat more threatening situation. A study by Henry and Rotter (1956) has shown that if subjects are told the Rorschach is a device for personality evaluation they exhibit more and stronger defense mechanisms when taking the test. Also, Luft (1953) has shown that a warm, non-threatening atmosphere provokes more preference for home-made inkblots than a cool, threatening atmosphere. A recent study by Smith (1962) separated college students into depressive and non-depressive groups on the basis of MMPI scores and investigated the differ- ence in preference between achromatic cards I and VI and chromatic cards 11 and VIII. It was found that. depressives preferred the achromatic cards and the non-depressives preferred the chromatic cards. How- ever, inferences about the specific effect of color cannot be made because the two groups of cards differ in other aSpects. For example, cards II and VIII seem to be more complex in design than cards I and VI. Also, it is difficult to make inferences about the Rorschach from this study as only four cards were used. Since simply changing the presentation sequence of the cards has been shown to have the effect of changing their symbolic meanings (Zax and Loiselle, 1960b), it may be that the four cards in Smith's study had vauired dif- ferent stimulus values after extraction from the gestalt of the standard ten card sequence. There have been a number of studies investigating preference for the Rorschach inkblots, but no reported study has yet used looking time as a measure of prefer- ence. The use of looking time as a measure of perceptual preference has the advantages that the cards may be gag)». , In ”I...“ 75"! $.qu presented in the standard ten card sequence main- taining a gestalt similar to that of the traditional Rorschach administration and that the length of vol- untary attendance to a card seems to be a natural, continuous variable indicating perceptual interest or preference (Berlyne, 1963; Gilgen and Bakan, 1965). Previous findings on card preference suggest a number of questions of primary interest in the present study: 1. Are there differences in perceptual pref- erence between cards? Are there differences in perceptual pref- erence between threatening and non-threat- ening instructional sets? Are there differences due to the sex of the subjects? Are there differences due to the anxiety level of the subjects? Is perceptual preference related to the complexity of the cards under a non- threatening instructional set? 10 6. Is perceptual preference related to the Sexual suggestiveness of the cards under a threatening instructional set? The only available measure of complexity was that of Beck's (1949) Z values. These values were developed to evaluate the relative difficulty of organizing a reSponse to each of the cards and not the stimulus complexity of the cards. Because of this, a separate study was run to obtain a better index of the complexity of each of the Rorschach inkblots. F . WEI-{I‘I'F' CHAPTER II IflETHDD Subjects The subjects were 50 male and 50 female volun- teersfrom undergraduate psychology courses at Michi- gan State University. No subject had seen the Ror- schach cards before the eXperiment. Each subject was assigned randomly to either of two experimental groups with the restriction that there be an equal number of each sex in each group. For the major analysis, then, there were four subgroups of 25 subjects each: Group I, Male; Group I, Female; Group II, Male; and Group II, Female. Fifteen more subjects from the Michigan State pOpulation were used in a separate study to obtain an index of complexity for each of the Rorschach inkblots. Procedure In this study looking time is used as a measure of perceptual preference. The dependent variable is defined as the amount of time a subject voluntarily attends to each of the Rorschach inkblots presented in their standard sequence. 11 12 Each subject was seated before a rectangular screen which had been placed on a desk. The experi- menter was hidden behind the screen during presentation of the Rorschach cards to avoid distraction of atten- tion from the cards. The cards were presented one at a time through a slot in the screen and the subject's looking time for each card was recorded by a stOpwatch. The instructions were given orally by the eXperi- menter and at the same time presented on a card that was passed through the slot in the screen. Group I (25 male and 25 female subjects) was presented with instructions designed to give the subjects a rather non-threatening set: The Rorschach cards are a series of inkblots that are frequently used by psychologists. This experiment is concerned with finding out how interesting these inkblots are to . college students. You are requested to look at a number of Rorschach cards. Look at each card as long as it interests you, then push it back through the screen. Another card will then be presented. Push this card back through the screen. The instructions for Group II (25 male and 25 female subjects) were designed to mobilize any defense Operations that may have an effect on perception of the inkblots: 13 The Rorschach cards are a series of inkblots that are frequently used by psychologists. This eXperiment is concerned with finding out how inte- rest in various cards reveals your inner conflicts and hidden person- ality characteristics. You are re- quested to look at a number of Ror- schach cards. Look at each card as long as it interests you, then push it back through the screen. Another card will then be presented. Push this card back through the screen. Following the presentation of all the cards the Taylor Manifest Anxiety Scale (Taylor, 1953) was admini- Stered to each subject to obtain an indication of anxiety level. A separate study was run to obtain an index of complexity for each of the Rorschach inkblots. In this study 15 subjects were each asked to look at all the inkblots and then arrange them on a table into 5 groups ranging from least complex to most complex. These groups were assigned values from 1 to 5 and the values of the cards were recorded for each subject. ivy 1.1.,Ild’ll'. u. CHAPTER III RESULTS Perceptual Preference gf the Cards Means of looking time in attention to the cards are reported in Table 1. Since the distribution of the dependent variable, looking time in seconds, was found to be skewed in the positive direction a log- arithmic transformation was used to reduce skewness and stabilize the variance (Winer, 1962, p. 221). To avoid values of zero a constant of 1 was added to all raw observations before the logarithmic transformation was performed. Analysis of variance (Winer, 1962, pp. 337-343) was used to test for differences in perceptual pref- erence between cards, between instructional sets, and between sexes. The results (Table 2) indicated dif- ferences between cards (p 4.01) and possibly between instructional sets on particular cards (p41.10). Differences between individual cards were inves- tigated by the Newman-Keuls method (Winer, 1962, pp. 80-85) and reported in Appendix A. The relative 14 15 menu mfiwooa oofiommsohu mo wage; mosooom ca. mans wfixooH mo mzmoza... H mum»; 9“.ngsz mufimommgu mm; 20 MZHH @2503 Omfimgmgdm. 07% Rad mo mZ 5 H S m 358 39111.11?!» .a egfi II 16 TABL E 2 ANALYSIS OF VARIANCE FOR PERCEPTUAL PREFERWCE Source of variation if MS F P1 Between subjects 22 . Instructions 1 .977 1.18 a Sex 1 1.453 1.76 IxS 1 .047 Subjects w. groups 96 .825 Within subjects 290 Cards 9 1.101 50.05** IxC 9 .039 1.77* SxC 9 .021 IxSxC 9 .008 stubj. w. groups 864 .022 *p A .10 *7’Cp 4 . 01 l7 perceptual preference of the cards may be summarized as follows: I V II IV III VI; VI IX VIII X The numbers of the cards are arranged in order of per- ceptual preference from least to most preferred. Cards underlined by a common line do not differ significantly in mean looking time. Cards not underlined by a common line differ significantly (p41.01) from each other. For example, cards I and V did not differ significantly from each other, but were preferred less than the rest of the cards. Card II was preferred less than VI, IX, VIII, and X but was not significantly different from , IV, III, and VII. Card VI was preferred less than VIII and X but was not significantly different from VI and VIII. Card VIII was preferred more than any of the non-multicolored cards but did not differ significantly from IX or X. Card X was preferred most but did not significantly differ from VIII. Tests at the .05 level of significance failed to further differentiate the cards in terms of perceptual preference. 18 Effects gf Anxiety £2131 In order to investigate the effects of anxiety level on perceptual preference of the cards, subjects were selected from the 100 subjects used in the main analysis so that there would be an equal number of high, moderate, and low anxiety subjects within each of the four sex by instructional set groups. The criterion for selection was each subject's score on the Taylor’Manifest Anxiety Scale. From each of the four sex by instructional set groups there were selected three subgroups of 7 subjects each. The high anxiety subjects' scores ranged from 22 to 37, moderate anxiety subjects ranged from.15 to 21, and low anxiety subjects from 1 to 14. Thus for the anxiety analysis there were twelve groups of 7 sub- jects each or a total of 84 subjects. An analysis of variance was performed to inves- tigate the effect of anxiety on perceptual preference between cards, instructional sets, and sex.of subjects. The results are presented in Table 4. There was no indication of any relation between anxiety level and perceptual preference. ANALYSIS OF VARIANCE FOR.ANXIETY LEVEL 19 TABLE 3 Source of variation ,Qf MS F Between subjects 8; Anxiety 2 .025 Instructions 1 .615 AxI 2 .369 Sex 1 1.693 1.85 AxS 2 .038 1x8 1 .098 AxIxS 2 .218 Subjects w. groups 72 .913 .ELEDEE subjects ‘126 Cards 9 .869 37.78** AxC 18 .015 1x0 9 .042 1.83* AxIxC 18 .025 1.09. SxC 9 .023 1.00 AxSxC 18 .023 1.00 IxSxC 9 .015 AxIxSxC 18 .022 stubj. w. groups 648 .023 *p41.10 **p¢:.01 20 Also, product moment correlations were obtained between looking times on individual cards and subjects' anxiety scores and between total looking times and an- xiety scores. These correlations were calculated for the total group of 100 subjects, each sex group, each instructional set group, and each sex by instructional set group. No correlations approached statistical significance. Perceptual Preference in}! 913113.; Measures To investigate the relationship of perceptual pre- ference to other measures product moment correlations were calculated between the following: 1. Perceptual preference of each card as measured by mean log looking time under non-threatening instructions. (Table 1) 2. Perceptual preference of each card as measured by mean log looking time under threatening instructions. (Table l) 3. Differences in perceptual preference for each card between non-threatening and threatening instructional sets as measured by differences in mean log looking time between instructional sets. (Table 4) 21 4. Complexity of the cards as measured by mean judgments calculated from 15 subjects who were asked to sort the cards into five groups from least complex (value = 1) to most complex (value = 5). (Table 5) 5. Complexity of the cards as measured by Beck's Z values (Beck, 1949) for the diffi- culty of organizing whole responses. (Appendix B) 6. Sexual suggestiveness of the cards as measured by total frequencies of sex associations to each card by Pascal et al.‘s (1950) 237 subjects. (Appendix B) 7. Relative preference of the cards as measured by the mean rank of each card calculated from Hershenson's (1949) 30 subjects who were asked to rank the cards from most liked to least liked. (Appendix B) The correlations are reported in Table 6. Because the Hershenson data was reported in terms of mean ranks the signs of these correlations have been changed so that the direction is consistent with the meaning of other'measures. 22 ooqa. wmoa. nooa. ammo. ooqo. nmqo. mono. memo. ammo. omNo. moocmuowmaa mcoauoshumca wsasoummunu Icon How oqNH.H mmma.a omqo.a omma.a mmqo.a mmam.a Hmma.a oHNN.A mmwm.a ommm.a oEHu woa :moz wcoauosuumaa wcacoumousu How Odom.a mmmm.a omqa.a qnma.a mmmo.a amom.a mNHN.H NmmN.H maam.a mmaq.a «Emu woa com: mmozoummwao mo HH> >H > HH H HHH> HHH H> XH x Hmpuo ma woumu 3 a w a e n a m N a $80 mHmm AdZOHHUDMHmZH zmmzemm mozmmmmmmm A HHH HH H >H He HH> > geese ea meaeo on a m a e m a m N a ameao mzamz mo.momma emaozaem ez< .moneme amaezaem .mzamz 24 mli- 1.. w Ho.uvmxe mo.uvmx em.- «ame. exam. Ne.- aaow. eamm. aeemammeae e.e0meeemamm .a I No. 00.: MN.I dm.u ON.: mfiOHumHOOmmm xom m..am um Hmommm .o u kaww. kom.n amp. rams. mosam> N m.xoom .m - aaoe.- aema. aemm. meaee mo seaeaaeeOo .a mumm amnoauosuumaa soozuon monouomowa Hosp : am.u xmo.u umooumm ma mononommam .m msoauosuumsa mcwsmumousu Moons 1 «ram. oucohomoum Hanummohmm .N msowuosuumza wnasmuwmucuuson Moon: 1 monouommum Hanumoohom .H o n a m N H meoam MzH enamommom may mo mmHHHaaso mpquHam mmmeo ez< mozmmmeumm naeemmoaam zaae mm mmeaameaoo m Mdm AnomfivmmJo oa.h mo.h aa.e mm.e ex.e oo.e oe.s Na.s so.m Aeoe.evea.a ea a m a o m e m N a mamas eoooosose x_ ma.o HHH> «xmq.oa mm.¢ MH kkmm.ma koo.oa mm.m H> shew.om keno.¢a tamm.oa Ho.q HH> kkmm.ou «kmw.¢a khan.oa on.d ma. HHH kkom.NN kkma.oa kkao.aa ma.© Nn.H nm.a >H kkNH.ON kkmm.ma khmo.ma kkmw.n mN.m ma.m om.a HH hamm.om ramm.¢m kaeq.om tamo.qa samo.oa samo.o xaom.m kaom.m > «kma.mm «amm.om axon.mm xxmo.oa xxam.ma axoa.ma axon.oa «amo.m mm.a H a HHmw1 xH H> Haw HHH >H HH > 11m oEHu MOH HmuOH a -#HH> ma. H> HH> _ HHH . >H , _ HH > .wm, asap man Hence we pruo ma mphmo an m w a o m e m N a noose $25.... mo madam QMMMQMO AA< ZO HmmH. mABMIH/ afifiz "mam/«.0 mo mun/BmmhmMm awafiofimm m mqmdd. APP ENDIX B '45 46 q.m N.m m.m m.n o.“ H.n m.o m.¢ «.0 m.© neon .e .aeaa .eoeeoe. susz .muothSm on Spam omumasoamo poxaa ummoa ou poxaa umoB_EOHm Econ com: aw mm on em mma an mm waa RNA mma Aaau .a .onen .Hw no Hwommmv .muOOHLSm nmm Eouw wnowumaoowwm wow «0 mocosomuw he omusmmme mm mmoso>aummwwam Hmnxom m.m m.m m.¢ m.m m.N o.a o.N m.m m.q o.a , Amam .a .aqma .xoomv .mmmsommou oaosz How mosaw> N m. oom an penummoa mm zuaxoa Eou N XH HHH> HH> H> > >H HHH HH moumo mmMDmdmZ Mammy. 0 "mm. Gamma/H I. 0430mm 3% mm“. m Mdm