Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Affective and cognitive responses to threatening people and places Leighton, Dana Charles 2004

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

Item Metadata

Download

Media
831-ubc_2004-0532.pdf [ 1.97MB ]
Metadata
JSON: 831-1.0091668.json
JSON-LD: 831-1.0091668-ld.json
RDF/XML (Pretty): 831-1.0091668-rdf.xml
RDF/JSON: 831-1.0091668-rdf.json
Turtle: 831-1.0091668-turtle.txt
N-Triples: 831-1.0091668-rdf-ntriples.txt
Original Record: 831-1.0091668-source.json
Full Text
831-1.0091668-fulltext.txt
Citation
831-1.0091668.ris

Full Text

AFFECTIVE AND COGNITIVE RESPONSES TO THREATENING PEOPLE AND PLACES by DANA CHARLES LEIGHTON B.A. Whi tman College, 2001 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS In THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming tp the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 2004 © Dana Charles Leighton, 2004 Library Authorization In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis'for financial gain shall not be al lowed without -my written permission. Dana Charles Leighton 06/25/2004 Name of Author (please print) . Date (dd/mm/yyyy) Title of Thesis: Affective and Cognitive Responses to Threatening People and Places Degree: Master of Arts Year: 2004 Department of Psychology The University of British Columbia Vancouver, BC Canada A B S T R A C T Humans have developed adaptive behaviors to cope w i t h threats to personal and col lect ive safety. One such behavior is increased vigi lance in situations that may connote danger. Ind iv idua l dif ferences, such as chronic dangerous w o r l d bel iefs, moderate personal responses to threatening situations. The present research examines the af fect ive and cogni t ive responses to t w o potent ial threats: immigrant outgroup members, and ambient darkness. Nei ther darkness nor immigran t outgroup threat alone affected integrat ive complex i ty ( IC) . The immig ran t outgroup cond i t ion resulted in more negative affect, especial ly i n combina t ion w i t h ambient darkness. B e l i e f i n a Dangerous W o r l d and Personal Need fo r Structure are shown to be moderators i n I C changes w h e n part icipants are under threat. i i TABLE OF CONTENTS Abstract i i Table o f Contents -- i i i L i s t o f Tables v Acknowledgement v i In t roduc t ion 2 Threatening Effects o f Immigran t Outgroups 2 B e l i e f i n a Dangerous W o r l d 4 Personal Need For Structure 5. A m b i e n t Darkness 6 Cogn i t i ve Complex i t y 7 Methods 9 Part icipants and Design -.9 Mater ia ls and Procedure 10 Results : 15 V o t i n g Behav ior 16 A f f e c t 16 Integrat ive Complex i t y 16 B e l i e f i n a Dangerous W o r l d 17 Personal Need for Structure 18 I m m i g r a t i o n Generat ion Effects 18 i i i Demograph ic Di f ferences 19 Discussion 20 Conc lus ion 27 References 30 Append ix : I m m i g r a t i o n V o t i n g Preference Measure 36 i v LIST OF TABLES Table 1 Means and Standard Deviat ions for A l l Dependent Var iables 38 Table 2 Correlat ions Between B e l i e f i n a Dangerous W o r l d ( B D W ) and A l l Dependent Var iables 39 Table 3 Correlat ions Between Personal Need for Structure (PNS) and A l l Dependent Variables 40 v A C K N O W L E D G E M E N T I thank Karen Devr ies for help designing this experiment, developing the vo t i ng preference measure, and, a long w i t h Rosemarie Ong, col lect ing and analyz ing the data. Ra j i v Jhangiani and A n d i K r a w c z y k were invaluable as m y re l iab i l i ty coders. M y labmates, Jason Faulkner and Justin Park, p rov ided advice and encouragement. I also thank Dr . M a r k Schaller and Dr . Peter Suedfeld for their generous f inancia l and laboratory faci l i t ies support, their patient reading o f numerous drafts, and Dr . D a n Per lman for serving on m y Master 's thesis commit tee. v i Threatening People & Places 1 A f fec t i ve and Cogni t ive Responses to Threatening People and Places Dana C. Le igh ton Univers i ty o f B r i t i sh Co lumb ia Threatening People & Places 2 A f fec t i ve and Cogni t ive Responses to Threatening People and Places The fo lks that are coming into this country are not jus t look ing for a j o b cut t ing your l a w n or replacing your roof. Some o f them, many o f them, are coming to replace y o u , you r very existence. ~ US Congressman T o m Tancredo (Un i ted States. Congress. House o f Representatives., 2003) I m m i g r a t i o n levels i n various countries around the w o r l d have waxed and waned, usual ly (at least i n democrat ic countries) as the result o f publ ic po l icy set by po l icy-makers in f luenced by the count ry 's populace. Some theorists attribute these po l icy changes to macroeconomic inf luences such as conf l i c t ing capital and labor interests, and the pol ic ies o f other potent ia l immigrant -accept ing countries ( for a rev iew see T i m m e r & W i l l i a m s o n , 1998). I n addi t ion to macroeconomic forces, there are also interpersonal factors at p lay in people 's thoughts and feelings about immigrants w h i c h affect their calls fo r po l i cy re fo rm. I n the Nether lands, once considered among the most immigrant- to lerant countr ies, people 's fears o f the degradat ion o f their o w n culture are d r iv ing immigra t ion po l icy to become less accept ing. A D u t c h c i t izen puts these fears succinct ly: " N o w there is anxiety and strange feel ings about foreigners coming here w h o do not want to l ive i n a Western w a y , " and a 27 year o ld M o r o c c a n immig ran t explains: " W h e n I was a k i d , the o ld Du tch ladies used to muss up m y hair ... n o w w h e n they see me, they wrap their handbag tw ice around their a r m " (Fle ishman, 2004) . Threatening Effects of Immigrant Outgroups Clear ly , the percept ion that immigrants pose a threat to the cul tural or personal safety o f the host count ry 's cit izens is an important factor i n prejudice toward immigran t outgroups. I n a German study speci f ical ly assessing perceived threat f r o m Turk ish immigrants , percept ion o f intergroup threat was posi t ive ly correlated w i t h desire for segregation, to expel the immig ran t Threatening People & Places 3 group, and favor ing cul tural assimi lat ion, and negat ively correlated w i t h acceptance o f immig ran t cul tural maintenance (Florack, P ion tkowsk i , Bohman, Balzer, & Perzig, 2003) . Perceptions o f immigrants as threatening are pervasive, but by no means universal . A survey o f Dan ish attitudes toward immigra t ion , for example, found on ly 2 5 - 3 0 % o f respondents indicated some degree o f prejudice, and this group bases its pre judic ia l bel iefs o n h o w immigrants look and act d i f ferent ly f r o m the dominant culture, their "soc ia l v i s i b i l i t y " (Enoch, 1994). Such v is ib le differences include the immigrants ' br ight c lo th ing, dark sk in , large fam i l y size, unusual ly pungent food (at least to Danes), and lack o f good Danish language ski l ls. Prejudice is l o w against Jewish immigrants to Denmark , but their social v is ib i l i t y is l o w compared to other ethnic immigran t groups that main ta in the cul tural and re l ig ious values f r o m their home societies. The use o f social v is ib i l i t y as a cue for immigran t threat may not be complete ly i r rat ional . One study on the l i nk between immig ra t ion and cr ime i n cit ies i n the Un i ted States showed that wh i l e there is no correlat ion between increasing rates o f immig ra t ion and increased cr ime, there is a posi t ive correlat ion between the propor t ion o f immigrants i n a c i ty and the c i t y ' s c r ime rate (Butcher & Piehl , 1998). Other research has shown that immigrants i n general are less l i ke ly to be invo lved i n cr ime than nonimmigrants , w i t h the except ion o f younger, less-educated males, and that the immig ra t ing group 's cul tural practices may legi t imize behavior that is i l legal i n the host cul ture (Mears, 2001). I n addi t ion, immigrants w h o feel marginal ized and reject the host cul ture 's norms and values may be more l ike ly to adopt c r imina l behavior (Junger-Tas, 2001) . The percept ion o f these social ly v is ib le immigrants as threatening to one's safety may then be seen as a reasonable inference intended to protect oneself f r o m being a v i c t i m o f cr ime. The percept ion o f threat extends beyond appearance, and economic threats are a strong inf luence i n ant i - immigrant sentiment. I n the Un i ted States and Canada, perceived threat posed Threatening People & Places 4 by immigrants to the wages o f domestic unsk i l led workers has affected immig ra t i on po l icy . A s income fo r l o w wage employees drops, creating dispari ty between h igh - and low-wage residents, po l i cy becomes more restr ict ive against immigrants ( T i m m e r & W i l l i a m s o n , 1998). I f percept ion o f immigrants as a threat is not universal , what m igh t account fo r ind iv idua l di f ferences i n immigran t outgroup threat perception? Researchers have considered a number o f ind iv idua l di f ference factors to help account for var iab i l i ty i n outgroup threat percept ion. A m o n g them are dissociat ion, social dominance or ientat ion, r i gh t -w ing author i tar ianism, re l ig ious fundamenta l ism, and j us t -wo r ld beliefs (Fiske, 1998, 2000). Pett igrew and Meertens (1995) posit a f o r m o f prejudice, "subt le pre judice," w h i c h encompasses defense o f t radi t ional values, exaggerat ion o f cul tural differences. Subtle prejudice also impl ies a denial o f posi t ive at t r ibut ions toward the outgroup, so that they are not perceived more negat ively than the ingroup, but rather the ingroup is perceived more posi t ively. I n addi t ion, researchers have evaluated regulatory focus (Shah, Brazy, & Higg ins , 2004) , trait anxiety (Ohman, F lyk t , & Lundqv is t , 2000) , and chronic beliefs i n a dangerous w o r l d (Schaller, Park, & Muel le r , 2003) . Belief in a Dangerous World Chronic dangerous w o r l d beliefs were one process studied by A l temeyer (1988) i n his exp lorat ion o f the personali ty correlates o f r i gh t -w ing authori tar ian attitudes ( R W A ) . Characterist ics o f persons h igh i n R W A include adherence to convent ional norms, submissiveness to author i ty f igures, and aggression toward people w h o m they perceive as a threat to social order. The B e l i e f i n a Dangerous W o r l d scale ( B D W ) measured the extent to w h i c h people bel ieved that dangerous people and situations were prevalent i n the w o r l d . Dangerous w o r l d be l ie f was found, along w i t h self-righteousness, to be a factor i n the hosti le attitudes and aggressive behaviors o f r i gh t -w ing authoritarians. People h igh i n B D W w o u l d be especial ly v ig i lan t to physical , cul tural , and economic surv ival threats posed by immigrants . Threatening People & Places 5 H i g h B D W indiv iduals migh t be predisposed to respond to unconvent ional immig ran t cul tural practices and perceived immigrant outgroup threats w i t h prejudice and host i l i ty . Threatening persons are strong indicators o f danger, and we l i ke ly have evo lved behaviors to avo id or confront the dangers posed by threatening outgroups. Physio log ica l indicators prov ide evidence o f autonomic affect ive and cogni t ive responses that may have evo lved to cope w i t h such threats. W h e n v i e w i n g facial expressions and asked to consider h o w the person seems to be feel ing, part icipants v i e w i n g " fear " or " a n g r y " faces used longer saccadic eye movements and increased foveal attention to the eyes, nose, and m o u t h (feature areas) than d id part ic ipants v i e w i n g neutral, happy, or sad faces (Green, W i l l i a m s , & Dav idson, 2003) . The authors concluded that people use dist inct v isual scanning processes fo r threat-related s t imu l i , and that the saccadic scanning migh t indicate increased autonomic responses, wh i l e the increased at tent ion to feature areas m igh t indicate more active cogni t ive appraisal. Ind iv idua l differences in trait anxiety, conceptual ly related to dangerous w o r l d bel ief, have also been shown to moderate people 's responses to threat — speci f ical ly , threatening faces. Low-anx ious indiv iduals have an attentional bias toward nonthreatening s t imu l i , w h i l e h igh -anxious ind iv iduals have a selective bias toward processing threat-relevant s t imul i (Ca lvo , 2000; Eysenck, 1991), such as a threatening face in a c rowd o f happy faces (Byrne & Eysenck, 1995). Us ing this " f l ex ib le adaptive system" (Ca lvo , p. 764), h igh-anxious ind iv iduals cogn i t ive ly elaborate and explore the meaning o f the threat, i n order to f i n d ways to overcome the threat. Personal Need For Structure Another personali ty variable related to trait anxiety is Personal Need fo r Structure (PNS) , w h i c h is the tendency for an ind iv idua l to prefer more s imple, rule-based, t igh t ly organized ways o f interact ing w i t h the social envi ronment (Neuberg & N e w s o m , 1993). The PNS scale was shown by N e w b e r g & N e w s o m to be moderately correlated w i t h trai t anxiety, especial ly social Threatening People & Places 6 anxiety (rs = .15 to .31). One study o f the moderat ing effect o f PNS on erroneous stereotype fo rmat ion found that h igh-PNS indiv iduals made more erroneous group stereotype at tr ibut ions, except w h e n to ld they w o u l d need to pub l ic ly jus t i f y their impressions. I f h igher-PNS ind iv iduals also have higher levels o f trait anxiety, they may be more sensitive to the pressure o f pub l i c l y j u s t i f y i n g their group attr ibutions. PNS has also been shown to affect the l i ke l ihood o f engaging i n stereotyping behaviors. Neuberg & N e w s o m (1993) found that h igh PNS levels increased the l i ke l ihood that ind iv iduals w o u l d engage i n gender stereotyping. I t has been proposed that h igh-PNS persons w i l l endeavor to reduce the complex i ty o f the in fo rmat ion in the envi ronment by using s imp l i f i ed categorical reasoning (Neuberg & N e w s o m , 1993; Schaller, B o y d , Yohannes, & O'Br ien, 1995). A s h igh-PNS ind iv idua ls , w h o also may be more l ike ly to have higher trai t anxiety, experience the anxiety o f intergroup threat, they may be especially mot ivated to s imp l i f y the complex array o f si tuat ional in fo rmat ion , and more l ike ly to engage in more s imple, categorical processing o f in tergroup in fo rmat ion . Ambient Darkness L i k e threatening outgroups, ambient darkness also acts as a strong cue o f danger i n the envi ronment , and occasions autonomic behaviors to help us cope w i t h the dangers. I n a cata loging o f fears among chi ldren and adolescents conducted a century ago by G. Stanley H a l l , "strange persons" and darkness t ied for th i rd place among prevalent fears, after thunderstorms and repti les. A b o u t darkness and the u n k n o w n terrors i n the darkness, he wrote : " T h e o l d n ight o f ignorance, mother o f al l fears, st i l l rules our nerves and pulses in the dark despite our better k n o w l e d g e " (1897, p. 189). Bo th b iopsychologica l and sociological research has indicated that there are fundamental changes i n people 's responses to s t imul i w h e n exposed to darkness. The acoustic startle ref lex was shown to be more pronounced for part icipants i n a dark r o o m Threatening People & Places 7 cond i t ion than a l ighted r o o m (Gr i l l on , Pe l lowsk i , Mer ikangas, & Dav is , 1997). I n another study, part ic ipants reading vignettes o f situations i nvo l v ing being in dark publ ic places reported more fear o f v i c t im iza t ion than those reading vignettes o f being i n l ighted publ ic places (Warr , 1990). The power fu l effects o f the si tuat ion on people 's response to darkness are unmistakable. However , there are also strong intrapersonal inf luences in responses to threatening situations and people. I n their research on startle reflexes in the dark, G r i l l o n , et al. (1997) found that on ly part ic ipants w h o reported being afraid o f the dark as ch i ldren exhib i ted a negative af fect ive response to darkness. Another study (Schaller et al. , 2003) had Caucasian and A s i a n part ic ipants wa tch a slide show o f black men 's faces either i n total darkness or w i t h some ambient l ight , and measured danger-relevant stereotyping behavior. Darkness alone was not a strong predictor o f whether subjects w o u l d use danger-relevant stereotyping, but rather darkness i n combina t ion w i t h h igher levels o f chronic beliefs the w o r l d is a dangerous place. Darkness, especial ly for those w i t h more disposi t ional be l ie f i n a dangerous w o r l d , promotes thoughts o f " e v i l , death, and danger" (Schaller et al. , p. 647), w h i c h may also activate emot ional responses congruent w i t h such thoughts. Cognitive Complexity One mental construct studied by researchers to understand cogni t ive responses toward threat is cogni t ive complex i ty . A l t h o u g h l i t t le research exists speci f ical ly addressing cogni t ive complex i ty and intergroup threat, some cogni t ive complex i ty research has evaluated the complex i ty o f people 's perceptions o f threatening and nonthreatening others. Three such studies ( B a l d w i n , 1972; Hogan, 1977; Kuna , 1976) found that part icipants used more complex i t y w h e n evaluat ing others w h o m they d is l iked, others w h o exhib i ted more negative behaviors, or threatening others w h o el ic i ted negative affect. The researchers concluded that increased complex i ty resulted f r o m the need to understand and predict the behavior o f threatening others. Threatening People & Places 8 Ano ther study used expected or unexpected shock s t imul i to evoke threat, and found that l o w -complex i t y part icipants were more v ig i lant to threat, were more anxious in the face o f an unpredictable threat, and desired a more structured situation (Harr is , 1981). Cogn i t i ve complex i ty is clearly a potent factor i n interpersonal and si tuat ional threat. M o s t research on complex i ty focuses on integrative complex i ty ( IC ) , w h i c h is the ab i l i ty to di f ferent iate mu l t ip le facets o f a si tuat ion, and to conceptualize the f l u i d interrelat ionships between those dimensions. Integrat ive complex i ty is a process that is in f luenced by contextual factors, par t icu lar ly situational or interpersonal stressors. Research in bo th laboratory settings and rea l -wor ld contexts has found a curvi l inear relat ionship between stress and complex i ty , whereby very l o w and very h igh stress levels are associated w i t h lower complex i ty , but moderate stress levels are required for h igh complex i ty (e.g. Porter & Suedfeld, 1981 ; Schroder, Dr iver , & Streufert, 1967; Suedfeld & Granatstein, 1995; Suedfeld & Tet lock, 2001) . Integrat ive complex i t y affects decision mak ing processes, and i t also has interpersonal and intergroup consequences i n communica t ion and negot iat ion (e.g. Sil lars & Parry, 1982; Streufert & N o g a m i , 1989; Suedfeld, Le igh ton , & Conway, i n press). A s the foregoing expl icat ion has i l lustrated, stereotyping and prejudice toward immigrants is a complex mix tu re o f power fu l si tuational, intergroup, interpersonal, and intrapersonal factors. O n the one hand, economic impacts, and perceived or real threat f r o m those w i t h v is ib ly di f ferent cul tural differences, affect people 's acceptance o f immigrants , wh i l e on the other, the magni tude o f response l ike ly depends on intrapersonal factors i n the percept ion o f threats posed by immigran t outgroups. S imi lar ly , people d i f fer i n their responses to ambient darkness as a threatening environmental st imulus. B D W may serve as a moderator fo r either af fect ive or cogni t ive responses to threat. Threatening People & Places 9 The present research aims to study h o w B D W and PNS both moderate af fect ive, cogni t ive, and behavioral consequences o f t w o potent ia l ly threatening situations: exposure to darkness, and immigran t outgroup in format ion. Exposure to the doubly- threatening si tuat ion o f immig ran t in fo rmat ion i n the dark should result i n more negative affect and also reduced integrat ive complex i ty , especially for higher B D W indiv iduals as their ab i l i ty to cope w i t h dangerous w o r l d threats is challenged. H igher PNS persons should s imi lar ly respond to immig ran t outgroup threat by using more simpl ist ic cogni t ion, result ing in lower integrat ive complex i ty . H igher B D W persons should also have more negative affect and reduced complex i ty i n the face o f threats f r o m darkness or immigran t outgroups. I w o u l d also expect higher B D W ind iv iduals to be more l ike ly to endorse ant i - immigrant po l i t i ca l posi t ions, especial ly i n the darkness and immigran t outgroup threat condit ions. Methods Participants and Design The part icipants were 125 undergraduate students at a large Western Canadian univers i ty . They part ic ipated in an experiment t i t led " T h i n k i n g A b o u t People," fo r approx imate ly 50 minutes, i n part ia l fu l f i l lmen t o f a psychology department subject poo l requirement. There were 102 w o m e n and 23 men i n the sample. W h e n se l f - ident i fy ing ethnic i ty , 51 indicated Chinese, 35 as Canadian, Caucasian, or European, 31 as other As ian , and 8 d id not specify ethnic i ty , or indicated mul t ip le ethnicit ies. There were 70 ls t -generat ion immigrants w h o had l i ved in Canada f r o m 3 months to v i r tua l ly their entire l ives ( M = .5 l i fe t ime, SD =. .26). The remain ing 55 were Canadian-born part icipants, o f w h o m 21 were second-generation immigrants . Threatening People & Places 10 The part icipants were tested in group settings w i t h f r o m 2 to 9 people per group. Each group was randomly assigned to one cel l i n the 2 ( immigra t ion vs. budget slide show) x 2 (ambient dark vs. l igh t i l l umina t ion level) between-subjects design. Materials and Procedure The experimenter we lcomed the part icipants and showed them into a r o o m w i t h four tables, each hav ing t w o or three chairs fac ing a pro ject ion screen. The part ic ipants i n the ambient L dark cond i t ion entered the r o o m w i t h the f luorescent l ights turned on to ha l f br ightness, wh i l e the ambient l igh t cond i t ion part icipants had f u l l l igh t ing. A f te r being seated, the part ic ipants were in fo rmed that the study w o u l d invo lve f i l l i n g out several questionnaires and watch ing a b r i e f slide show. Participants then gave in fo rmed consent in fo rmat ion and the study began. Individual Difference Measures. Each part ic ipant completed a questionnaire packet contain ing a demographic questionnaire and several scales. The f i rst section o f the packet contained ind iv idua l di f ference measures. The B e l i e f i n a Dangerous W o r l d ( B D W ) scale (A l temeyer , 1988), assesses the extent to w h i c h ind iv iduals perceive the w o r l d around them as being chronical ly dangerous. I t contains 12 i tems such as " A n y day now, chaos and anarchy could erupt around us. A l l the signs are po in t ing to i t , " and "Despi te what one hears about ' c r ime in the street,' there probably isn ' t any more n o w than there has ever been." The part ic ipant rates agreement w i t h the i tems along a scale f r o m 1 to 7. A composite " B D W index" score was obtained by reverse-scoring several i tems, and comput ing the mean o f al l i tems. The Personal Need for Structure (PNS) scale (Neuberg & N e w s o m , 1993) was used to measure people 's tendency to prefer structured, rule-based situations. The scale has 12 i tems, rated o n agreement f r o m 1 to 7, worded such as " I enjoy hav ing a clear and structured mode o f l i f e , " and " I enjoy being spontaneous." A composite " P N S index" was computed by reverse Threatening People & Places 11 scor ing several i tems, and calculat ing the mean o f the i tem ratings. F o l l o w i n g the ind iv idua l di f ference measures was a page instruct ing the part icipants to stop and wa i t for instruct ions f r o m the experimenter. Illumination manipulation. W h e n al l the part icipants had f in ished the f i rst section o f their questionnaire packets, they were t o l d that there w o u l d n o w be a b r ie f slide show, and that i n order to let their eyes adjust to the darkness, the l ights w o u l d be turned o f f for 30 seconds pr io r to the slides. I n the ambient l igh t cond i t ion , a door to an adjacent br ight ly l i t experiment r o o m was lef t open three inches, a l l ow ing a soft ambient l igh t to f i l ter in . I n the ambient dark condi t ion, the door was closed, resul t ing in total darkness. A f t e r the l ights were turned o f f for 30 seconds, the slide show began. Each slide was projected onto the screen fo r a per iod o f 10 seconds, f o l l owed by 6 seconds o f b lack screen. The f ina l slide was fo l l owed by 30 seconds o f b lack screen. Slide show manipulation. The slide show manipu la t ion consisted o f t w o condit ions: "budge t " and " i m m i g r a t i o n . " I n the i m m i g r a t i o n condi t ion , part icipants watched a set o f 6 slides, each conta in ing a statement either i n favor of , or against, immigra t ion . The statements were taken f r o m various p ro - and ant i -i m m i g r a t i o n web sites, and also transcripts f r o m U.S. congressional speeches. O n l y the statement i tself, not the source o f the statement, was g iven on the slide. Three slides were i n favor o f immig ra t ion , and contained photos o f relat ively f r iend ly look ing persons, w h o appeared clean and professional. T w o o f these slides had photos o f men , one apparently East A s i a n and one South As ian . One slide had a photo o f an apparently East As ian w o m a n . The p ro - immig ra t i on slides featured text such as " Immigran ts are essential to the foster ing o f cosmopol i tan ism and Threatening People & Places 12 product ive d ivers i ty , and u l t imate ly to the economic prosperi ty o f the nat ion i n the twenty- f i rs t century . " The three other slides were opposed to immigra t ion , and featured photos o f re lat ive ly un f r iend ly , unkempt look ing persons. T w o o f these slides featured photos o f w o m e n , one apparently A f r i c a n and one apparently La t i n Amer ican . One slide contained a photo o f a man, apparently A f r i c a n . The ant i - immigra t ion slides had quotes such as " Immig ran ts are coming to replace y o u , your very existence. They are coming across porous borders, and the on ly w a y that i t can ever be dealt w i t h is, I reiterate, to prov ide a major mi l i ta ry or l a w enforcement presence on that border . " Th is research was interested i n h o w ind iv idua l differences m igh t sensitize h i g h B D W part ic ipants to immigran t threat, so that both p ro - immigra t ion and an t i - immigra t ion slides were used to avo id unintent ional ly inducing negative affect and an t i - immigra t ion sentiment among al l the part ic ipants i n the immigra t ion condi t ion by having on ly negative i m m i g r a t i o n in fo rmat ion presented. Groups o f subjects i n the " i m m i g r a t i o n " cond i t ion were randomly assigned to one o f t w o counterbalancing condit ions, where the slide show either began w i t h a p ro - immig ra t i on slide and ended w i t h an an t i - immigra t ion slide, or vice-versa. There was no effect o f counterbalancing on the dependent variables. I n the "budge t " condi t ion, the part icipants watched a slide show consist ing o f six slides, each o f w h i c h contained a posi t ion statement f r o m a f ic t i t ious po l i t i c ian , on the ways in w h i c h a budget surplus should be spent. Statements contained text such as " W e should be target ing surplus funds to social needs such as education, health-insurance coverage, ch i ld care and j o b t ra in ing . " Each slide contained a photo relevant to the pos i t ion (e.g., a school, a je t f ighter , a luxury yacht) . Threatening People & Places 13 F o l l o w i n g the slide show the l ights were turned back on, to the levels used p r io r to the slide show (ha l f brightness for dark — fu l l brightness for l ight ) . The part ic ipants then completed the second part o f the questionnaire packet. Dependent variable measures. The second part o f the packet contained measures to assess the dependent variables i n the study: affect, vo t ing choice, and integrat ive complex i ty . The f o l l o w i n g three scales were presented i n the order l isted. The Posit ive and Negat ive A f fec t Scale ( P A N A S ) (Watson, Clark , & Tel legen, 1988) was used to assess general mood. This questionnaire contains 10 posi t ive ly valenced affect ive adjectives (interested, exci ted, strong, enthusiastic, p roud, alert, inspired, determined, attentive, act ive) and 10 negat ively valenced adjectives (distressed, upset, gu i l ty , scared, host i le, i r r i table, ashamed, nervous, j i t te ry , afraid). The part ic ipant rates each adjective f r o m 1 to 5 fo r the extent to w h i c h he or she feels that way. The P A N A S instrument is d iv ided into posi t ive affect ( P A ) and negative affect ( N A ) factors. The scale was scored by comput ing the mean o f responses fo r each o f the t w o factors to derive a P A and a N A index. Internal re l iab i l i ty for each o f the factors was h igh (Cronbach 's a > .92), consistent w i t h previous re l iab i l i ty studies (Watson et al.) . The P A and N A factors were moderately correlated (Spearman's p = . 3 5 1 , p = .000). Th is posi t ive corre lat ion may be related to a h igh propor t ion o f As ian part icipants i n the sample, an issue that w i l l be explored i n the results section. Previous re l iab i l i ty studies (Watson et al.) have f o u n d these factors to have smal l to moderate negative correlations (rs f r o m -.15 to - .23) , but these studies were per formed w i t h M idwes te rn U.S. universi ty populat ions, w h i c h probably contained a smaller percentage o f As ian part icipants. A vo t ing preference measure (see Append ix ) was used to test h o w vo t ing preferences were affected by the manipulat ion. Th is measure asks the part ic ipant to consider 7 po l i t i ca l Threatening People & Places 14 candidates and vote for the one they w o u l d favor based on the issue o f i m m i g r a t i o n alone. I t assessed whether the part ic ipant preferred a pol i t ica l candidate w i t h a moderate or more extreme ant i - or p ro - immig ra t ion posi t ion. The part ic ipant 's choice resulted i n a " v o t i n g preference" score ( l = m o s t an t i - immigra t ion ; 7=most p ro- immigra t ion) . A vo t ing ext remi ty var iable was calculated by subtract ing the part ic ipant 's vo t ing score f r o m the median o f a l l part ic ipant scores. The vo t ing scale was pre-tested w i t h 26 part icipants f r o m the same subject popula t ion three months before data col lect ion for the present study. The pre-testing consisted o f hav ing part ic ipants sort the seven i tems based on h o w pro- or an t i - immigra t ion the i tems were. The i t e m sort ing achieved a re l iab i l i ty o f .94 (Cronbach's a ) across subjects. Integrat ive complex i ty was measured w i t h the Paragraph Comple t ion Test (PCT) , w i t h six sentence stems (three on each page). The PCT asks part icipants to complete each sentence stem, and to f o l l o w it w i t h a f e w more sentences to complete the paragraph. The sentence stems cover three cogni t ive domains: interpersonal ( " W h e n a f r iend acts d i f ferent ly . . . " , "Strangers. . . " ) , intrapersonal ( "Confus ion. . . " , " W h e n I feel anxious.. .") , and inst i tut ional ("Rules. . . " , " T h e heal th care system i n Canada..."). The part icipants were to ld that they had three minutes to complete each paragraph but were not t imed. Except for a f e w subjects, al l paragraphs were completed before the end o f the experiment. The P C T is usual ly used to study conceptual complex i ty , or the re la t ive ly stable d imension o f complex i ty associated w i t h a g lobal , domain nonspeci f ic, organizat ion o f cogni t ive structure (see Coren & Suedfeld, 1990), wh i le a domain-speci f ic w r i t i n g sample is of ten used to study integrat ive complex i ty as i t is affected by experimental manipulat ions. The t w o measures have shown a moderate correlat ion ( r = .33), indicat ing they both measure the structural elements o f t h ink ing — d i f ferent iat ion and integrat ion (de Vr ies & Walker , 1987). The P C T was used i n Threatening People & Places 15 this study to measure the impact o f threatening people and places on the more g lobal leve l , rather than the integrat ive complex i ty about any specif ic domain. The materials were scored fo r complex i ty using guidel ines publ ished i n a scor ing manual ( B a k e r - B r o w n et al . , 1992). Each paragraph is coded for the level o f conceptual comp lex i t y exhib i ted i n the paragraph, on a scale o f 1 to 7. A score o f 1 indicates un id imens ional , d ichotomous th ink ing ; 3 indicates the recogni t ion o f mul t ip le dimensions; 5 indicates the recogni t ion o f integrat ive relationships between the dimensions; 7 indicates di f ferent iated and h igh ly integrated conceptual processing. Scores o f 2, 4, or 6 are intermediate levels i n the scale. The mean o f a l l paragraphs is taken to represent an overal l integrat ive complex i ty score on a scale o f 1 to 7. Th is measure was coded by the experimenter, b l i nd to the exper imental cond i t ion for each paragraph. A subset o f paragraphs was scored independently by t w o other t ra ined complex i ty coders, w h o achieved interrater re l iab i l i ty o f .84 (Cronbach's a ) w i t h the experimenter. A f t e r al l part icipants had completed the questionnaires, the part icipants were debriefed about the intent o f the experiment, its hypothesis, and a l lowed to ask questions o f the experimenter. Results T o analyze m a i n effects and interact ion effects for the dependent variables o f interest, I used a 2 x 2 A N O V A . Table 1 lists al l dependent variables along w i t h their means and standard deviat ions for each cel l i n the 2 x 2 A N O V A , and signi f icant effects are indicated. Since the hypotheses invo lve the doubly threatening effects o f darkness combined w i t h immig ran t outgroups, p lanned contrast tests were conducted in addi t ion to the omnibus A N O V A interact ion. The dark / immigra t ion condi t ion was contrasted against the other three condi t ions to f i n d theoret ical ly relevant relationships. I n order to analyze the moderat ing effects o f B D W and Threatening People & Places 16 PNS, regression analyses (details explained i n the B D W section be low) were conducted. The results w i l l be presented for each o f the dependent variables, B D W and PNS moderat ion, and i m m i g r a t i o n effects w i l l be explained. Correlat ions between the dependent variables and B D W and PNS are l isted i n Tables 2 and 3, respectively. Retrospective power values for the 3-way interact ions are also l isted i n Tables 2 and 3. Unless stated otherwise, a levels o f .05 are used fo r signi f icance test ing, and exact signif icance values are always reported. Negat ive affect scores had an extremely posi t ive ly skewed dist r ibut ion, so analyses o f correlat ions w i t h negative affect were done w i t h Spearman's p, as i t is robust to v io lat ions o f normal i ty (Gu i l f o rd , 1965). Voting Behavior There were no signi f icant ma in effects or interact ion effects for the independent variables on vo t i ng behavior. The means for each group were i n the expected d i rect ion on ext remi ty , and opposite to the expected d i rect ion on vo t ing score. Affect A n analysis o f variance showed that immig ra t ion slide groups reported a higher level o f negative affect than the budget groups, F ( l , 1 2 1 ) = 4.03,p = .047. Negat ive affect was higher i n the dark - immigra t ion condi t ion, wh i le being s imi lar ly lower i n the other three condi t ions. The planned contrast test was signif icant, ^(43.85) = 2.06,p = .045. These results indicate that part ic ipants experience more negative affect when exposed to immig ra t i on slides than budget, but this effect may be accentuated by being i n the dark. Th is f i nd ing is consistent w i t h the hypotheses. The manipulat ions had no signi f icant effect on posi t ive affect. Integrative Complexity There were no signi f icant ma in effects or interact ion effects o f the man ipu la t ion condi t ions on the I C measure. The I C measure is d iv ided into three domains: interpersonal, intrapersonal, and inst i tut ional . T o further examine possible effects, I analyzed each doma in and Threatening People & Places 17 inc luded the results i n Tables 1, 2, and 3. However , interpretat ion o f domain-speci f ic results should be made w i t h caut ion — each domain only contains t w o sentence stems, but at least f ive stems should be scored for complex i ty results to be rel iable and va l id (P. Suedfeld, personal commun ica t ion , A p r i l 2001). Belief in a Dangerous World T o examine the ma in effects and moderat ing effects o f B D W , I conducted a series o f regression tests. Us ing a method explained i n Schaller, Park, & Meu l le r (2003) , I standardized the independent variables: i l l umina t ion ( - 1 , 1), slide ( - 1 , 1), and B D W (z-score t ransformat ion) , and calculated each o f the interact ion terms by mu l t i p l y i ng the standardized values together ( I l l um ina t ion x Sl ide, I l l umina t ion x ' B D W , Slide x B D W , and I l l um ina t ion x Slide x B D W ) . The resul t ing seven variables were entered as predictors i n a series o f regression analyses on each o f the dependent variables. Table 2 shows the correlations between B D W and a l l dependent variables. These analyses showed no signi f icant effect o f B D W on the vo t ing behavior measure either as a m a i n effect or as a moderat ing variable. There was a signi f icant m a i n effect o f B D W on negative affect (|3 = .187, p = .045), and on overal l integrat ive complex i ty (|3 = - .24, p = .011). The di rect ion o f the effects supports the hypotheses that h igher B D W is associated w i t h higher negative affect and lower complex i ty . I n addi t ion to the ma in effects o f B D W , this study also is interested i n the degree to w h i c h B D W migh t act as a moderator o f people's affect ive and cogni t ive responses to darkness and immig ra t ion . Several interesting results emerged f r o m our analysis o f B D W ' s moderat ing effects. The B D W x I l l um ina t ion interact ion predicted posi t ive affect ( P A ) , a result that approached but d id not attain signif icance ((3 = -.\76,p- .064). T o explore the nature o f this interact ion, a correlat ion was done, w h i c h showed that B D W and P A were negat ively , a l though Threatening People & Places 18 not s ign i f icant ly , correlated in the dark (r = -.\9,p = .122), wh i le the corre lat ion i n the l igh t was weaker ( r - .095, p = .474). There was no signi f icant moderat ing effect o f B D W on negative affect i n any o f the interactions. The analysis found a three-way B D W x I l l umina t ion x Slide interact ion fo r integrat ive complex i ty (|3 = -.185, p = .048). Th is interact ion was dr iven by a strong negative corre lat ion between B D W and integrat ive complex i ty i n the dark / immigra t ion cond i t ion ( r = - .429, p - .01), whereas correlat ions in the other three condi t ions were weaker and nonsigni f icant (rs < -.289 and ps > .16). This supports the hypothesis that the moderat ing effect o f B D W on complex i t y change is stronger i n the dark / immigra t ion double threat condi t ion than i n the other condi t ions. Personal Need for Structure A s imi lar regression analysis was per formed to investigate any m a i n effects or moderat ing effects o f Personal need for structure (PNS) on the dependent variables. There were no m a i n effects o f PNS on any o f the dependent variables. Table 3 shows the correlat ions between PNS and al l dependent variables. The PNS x Slide interact ion showed a signi f icant effect for vo t ing ext remi ty (|3 = .198,/? = .041), and integrat ive complex i ty (|3 = - . 2 0 1 , p = .034). Participants i n the budget slide show cond i t ion exhib i ted a moderate negative correlat ion between PNS and vo t ing ext remi ty ( r = -.288,/? = .033). Participants i n the immigra t ion slide show cond i t ion showed a moderately negative correlat ion between PNS and integrative complex i ty , ( r = -.283,/? = .019). Relat ionships i n the other condi t ions were weak and nonsigni f icant (rs < .09, ps > .51). Immigration Generation Effects Since the slide show condi t ion and the vo t ing measure are both immigrat ion-re levant , I analyzed any moderat ing effects that being a 1 st or 2nd generation immigran t m i g h t have o n the vo t i ng measure and slide show condit ions. I entered a standardized var iable representing the Threatening People & Places 19 number o f generations since immig ra t ion to Canada (-1 for 1st generation, 0 fo r 2nd generat ion, 1 for th i rd or greater generation) into a regression equation as a m a i n effect, a long w i t h the 2 i m m i g r a t i o n and slide ma in effects and 4 interact ion variables. There were no ma in or moderat ing effects o f immig ra t ion generation on vo t ing behavior, and no Generat ion x Slide interact ion effects on any o f the dependent variables. There was a s igni f icant m a i n effect o f immig ra t ion generation on integrat ive complex i ty ((3 = .189,/? = .041), w i t h a posi t ive correlat ion that approached signif icance (r = .\6S,p = .061). A n A N O V A compar ison was signif icant, F(2, 122) = 6.409, p = .002, and a Tukey ' s H S D post-hoc compar ison showed that 2nd generation immigrants were s igni f icant ly higher i n comp lex i t y than 1st generat ion immigrants ( M = 1.81 vs. 2.20), but there were no s igni f icant di f ferences between 3rd or greater generation immigrants and the other groups. A n A N O V A and Tukey ' s H S D test also revealed immig ra t ion generat ion dif ferences fo r B D W and PNS. The 2nd generation immigrants had lower PNS scores than 1st generat ion immigrants (M= 3.31 vs. 3.80, F(2 ,122) = 4.317,p = .015). The 2nd generation immigrants also had lower B D W scores than 1st and 3rd or greater generation immigrants (M- 3.11 vs. 3.77 and 3 .71 , F (2 ,122) = 5.225,p = .007). Demographic Differences One demographic variable that may affect the results is the presence o f a large propor t ion o f Chinese part icipants (40%) compared to other ethnicit ies (other As ian = 2 5 % ; Caucasian = 2 8 % ) . A n A N O V A revealed a signi f icant di f ference between the ethnic groups, F (3 ,121) = 6.60, p < . 001 , and a Tukey ' s H S D showed Chinese part icipants to be s igni f icant ly higher than Caucasians i n B D W (Ms = 3.93 vs. 3.15), and showed less var iab i l i ty i n the measure (SDs = .70 vs. .83). Threatening People & Places 20 The h igh propor t ion o f part icipants w h o indicated they were o f As ian ethnic i ty (65%) may help exp la in the puzz l ing moderate posi t ive correlat ion between P A and N A observed i n the data. Chinese and other As ian part icipants showed a signi f icant moderate corre lat ion between P A and N A (Spearman's p = .438,/? < .001), wh i le non-As ian part icipants had a weaker nonsigni f icant correlat ion (Spearman's p = .248,/? = .109). Some hypothesized reasons for this di f ference w i l l be explored in the discussion section. Discussion Contrary to the hypotheses, higher B D W was not a moderator i n people 's responses to either darkness or threatening outgroups, but rather higher B D W people seem to general ly experience s l ight ly more negative affect across al l the condit ions. H igher B D W ind iv iduals showed a general ly lower level o f integrative complex i ty , w h i c h is congruent w i t h research on author i tar ianism and complex i ty . Consistent w i t h the hypotheses, the doub ly threatening experience o f immigran t outgroup threat i n the darkness takes an especial ly severe to l l on integrat ive complex i ty for higher B D W people. Th is f i nd ing concurs w i t h the d isrupt ive stress hypothesis (Schroder et al . , 1967; Streufert & Schroder, 1965), w h i c h posits that after surpassing an " o p t i m u m " cogni t ive load imposed by in fo rmat ion load and stressors, people 's cogni t ive complex i ty is reduced. I n the case o f the higher B D W part icipants i n this study, the nox i t y o f the darkness combined w i t h the threatening outgroup may have chal lenged their ab i l i ty to cope w i t h these stressors, resul t ing i n reduced complex i ty . L o w e r B D W part icipants were less affected by these stressors. The present research supports the general not ion that w h e n faced w i t h i n fo rmat ion about immigran t outgroups, people experience more negative affect. Th is supports in tergroup v ig i lance theory (Schaller, 2003) , and identi f ies one d imension o f experience that may func t ion i n intergroup v ig i lance. Exposure to a potent ia l ly threatening outgroup arouses a response i n people Threatening People & Places 21 that includes greater negative feelings. This effect is especially strong in condi t ions o f darkness, w h i c h may be a potent t r igger for feelings o f danger and thoughts o f vu lnerab i l i t y , a result i n accord w i t h the hypotheses. Interpretat ion o f this result needs to be qual i f ied because the budget slide show condi t ion may have s imply reduced negative affect, as the part ic ipants considered on ly the posi t ive ways a budget surplus could be al located to help society. However , i t w o u l d be expected in this case that posi t ive affect w o u l d be higher for part icipants i n the budget cond i t ion , w h i c h was not the case. There were no signi f icant effects o f the manipulat ions on the vo t ing preference measure, w h i c h is contrary to the hypotheses. The result ing means indicated that part ic ipants expressed vo t ing preferences more favorable toward immigra t ion after watch ing the i m m i g r a t i o n slide show than after watch ing the budget slide show. Since the results were nonsigni f icant , no inferences can be made about them. Regardless, the results are interesting. I t may be that the pro-immig ra t i on slides were more persuasive than the ant i - immigra t ion slides, or that many o f the part ic ipants were reminded o f the so-called po l i t ica l incorrectness o f an t i - immigra t ion v iewpo in ts , occasioning the endorsement o f moderate or p ro - immig ra t ion posit ions. I f these hypotheses were correct, I w o u l d expect that this general movement i n vo t ing toward the pro-i m m i g r a t i o n pos i t ion w o u l d compress the var iab i l i ty o f the vo t ing measure against a ce i l ing effect. Instead, the immigra t ion slide show occasioned greater var iab i l i ty i n i m m i g r a t i o n vo t ing than d id the budget slide show. This greater var iab i l i ty seems to indicate the presence o f a th i rd var iable that is in f luenc ing some, but not a l l people to endorse more p ro - immig ra t ion posi t ions i n the immig ran t condi t ions. Th is result may be due to impression management ( I M ) concerns aroused by watch ing the i m m i g r a t i o n slide show. The immigra t ion slide show, f o l l owed by the immigra t ion-spec i f i c vo t i ng measure, m igh t act to arouse the so-called Gamma, or moral is t ic , constel lat ion o f I M Threatening People & Places 22 concerns (see Paulhus & John, 1998). Subjects higher i n I M concerns w h o may have reacted to the slide show w i t h ant i - immigrant sentiment m igh t subsequently endorse more moderate or pro-i m m i g r a t i o n candidates to al lay their I M concerns. I n the budget condi t ion , such concerns w o u l d not have been aroused fo r higher I M concerned part icipants, and the vo t ing preference score w o u l d not be moved toward a more pro- immigrant posi t ion. Table 1 showed the mean i m m i g r a t i o n vo t ing score was more p ro - immigra t ion in the immig ra t ion slide cond i t ion , w h i c h w o u l d support this hypothesis. Since I M was not measured, its effects cannot be cont ro l led for to test this hypothesis. Ano ther explanat ion exists for the vo t ing preference di f ference --. each immig ran t generat ion responded di f ferent ly to the immigra t ion condi t ion. Participants w h o were 1st and 2nd generat ion immigrants indicated more p ro - immigra t ion vo t ing preferences i n the i m m i g r a t i o n condi t ion , wh i le 3rd and greater generation immigrants chose vo t ing preferences less p ro - immig ra t i on i n the immig ra t ion slide condi t ion than the budget slide condi t ion . Since the sample had a h igh percentage o f 1st and 2nd generation immigrants , their in f luence may have caused an increase in the overal l mean. The differences were nonsigni f icant , and i t is impossible to rule out sampl ing error as the cause for these differences. There were no signi f icant differences between condi t ions for integrat ive complex i ty or vo t i ng ext remi ty . I t is l i ke ly that for most part icipants (higher B D W part ic ipants be ing an except ion) , the st imulus materials were not power fu l enough to el ic i t such responses f r o m indiv iduals . However , the d i rect ion o f the means was as predicted. A s shown i n Table 1, integrat ive complex i ty was lower, and vo t ing ext remi ty was higher, for part ic ipants i n the i m m i g r a t i o n condi t ion. The 2nd generation immigrants were, on who le , s igni f icant ly higher i n complex i ty than their 1st or 3rd or greater generation counterparts. A l t h o u g h there are no publ ished studies o f complex i ty differences between immigran t generations, I can postulate some Threatening People & Places 23 possible reasons for this dif ference. One possible factor is the use o f t w o languages i n the home. I t w o u l d be expected that parents o f 2nd generation immigrants use bo th their nat ive language and the host cul ture 's language, wh i le at the same t ime their ch i ldren are learning and using the host cul ture language in school. I f this were the case, there m igh t be an increased cogni t ive f l ex ib i l i t y or complex i ty as a result o f navigat ing both languages simultaneously. Ano the r possible explanat ion is cul tural disparit ies i n interpretat ion. I f the ch i ldren o f 1st generat ion immigrants are exposed to mul t ip le , contradictory, cul tural interpretations o f the same event or s i tuat ion, they may learn to use more dif ferent iated and integrated th ink ing . These interpretations w o u l d be very valuable hypotheses to test. The signi f icant strong posit ive correlat ion between negative affect and bo th B D W and PNS that emerged i n the l ight/budget condi t ion was unexpected. U n l i k e the other three condi t ions, the l ight /budget condi t ion also had a moderate posi t ive correlat ion between B D W and posi t ive affect, a l though i t was nonsigni f icant. W h y higher B D W and PNS levels m igh t be associated w i t h more negative affect i n a condi t ion o f no threat is inconsistent w i t h the hypotheses and remains to be explained. Taken together, what do the results te l l us about the affect ive and cogni t ive consequences o f threatening people and places? Darkness itself, at least o f short durat ion, does not have a s igni f icant effect on posi t ive or negative affect, nor is there an effect on integrat ive complex i ty . Exposure to in fo rmat ion about a threatening outgroup, i n this case, immigrants , led people to " f e e l " more negat ively, and i n the darkness this effect was accentuated. Integrat ive complex i ty was affected by the dark / immigra t ion threat, but was moderated by B D W , such that h igher B D W part ic ipants experienced more reductions i n complex i ty . I t appears that B D W moderates people 's responses by in f luenc ing cogn i t ion rather than affect. The results that Schaller, et al. (2003) obtained showing more danger-relevant Threatening People & Places 24 stereotyping i n higher B D W indiv iduals may be due to changes in cogni t ion , perhaps cogni t ive s impl i f i ca t ion . The prevalent concept ion o f stereotyping identi f ies i t as a cogni t ive process rather than an affect ive one (cf. Fiske, 1998). The present research supports this concept. Another personal i ty variable that moderated cogni t ive processing i n this study was personal need fo r structure (PNS). The results showed that higher PNS ind iv iduals exhib i ted s imp l i f i ed cogn i t ion w h e n presented w i t h the immigrant- re levant in fo rmat ion . Th is result supports the hypotheses, and is i n l ine w i t h previous research on stereotype fo rmat ion (Schal ler et al . , 1995) w h i c h found that h igh PNS indiv iduals use more s impl is t ic reasoning strategies i n f o r m i n g erroneous stereotypes. The present results lend support to the no t ion that these ind iv iduals use what Schaller, et al. (1995) cal led a "s impl is t ic two-d imens iona l cataloging sys tem" (p. 553) when encoding group-relevant in format ion. However , the present results showed this effect to be present only i n the immigrant slide show condi t ion , so the effects o f PNS on cogn i t ion are activated i n intergroup threat condi t ions, but not env i ronmenta l threat (darkness) condi t ions. B D W and PNS showed a l o w correlat ion ( r = .147,/? = .10), w h i c h indicates that they are measur ing largely di f ferent moderat ing processes i n this experiment. The data reveal that their moderat ing effects on cogni t ion operate in di f ferent condit ions. W h i l e B D W moderates cogn i t ion i n the combined immigran t outgroup and darkness condi t ion, PNS operates in the cond i t ion o f immig ran t outgroup threat. W h i l e higher PNS indiv iduals may be more wor r ied about the threat to social order presented by immigran t outgroups, higher B D W indiv iduals are more affected w h e n the threats are combined. These results bolster previous research by Schaller (i.e. Schaller et al . , 1995; Schaller et al. , 2003) , and i l lustrate the importance o f these t w o ind iv idua l di f ference variables in understanding these ind iv idua l differences in social cogn i t ion w h e n we are exposed to threatening persons or places. Threatening People & Places 25 Some o f the characteristics o f the part icipants i n this exper iment require discussion. There were 91 (72%) 1st and 2nd generation immigrants i n the sample. First generat ion immigrants indicated a more p ro - immigra t ion vo t ing preference than 2nd and 3rd or greater generat ion immigrants (Ms = 4.65, 4.43, & 4.38, respect ively), but the dif ferences were not s igni f icant , F (2 ,121) = .577. The lack o f a s igni f icant di f ference between the generations is puzz l ing. I t w o u l d be expected that part icipants w h o were recent immigrants or ch i ldren o f immigrants w o u l d be more l ike ly to endorse pro- immigrant candidates than part ic ipants w h o have a longer generational history i n the country. T w o surveys lend l im i ted support to this expectat ion. A nat ional telephone po l l i n the Un i ted States o f A m e r i c a ( U S A ) (Puente, 1993) and a survey conducted speci f ical ly w i t h M e x i c a n Amer i can immigrants i n Texas (Binder , Pol inard, & W r i n k l e , 1997) showed that support for restr ict ive immig ra t ion pol ic ies increases w i t h the number o f generations immigrants are i n the U S A . These surveys d i f fer f r o m the present research i n that they were conducted i n the U S A and i n the general populat ion, wh i le this research draws f r o m a Canadian univers i ty student popula t ion. L i t t l e empir ica l research exists on Canadian immigran t attitudes toward the subject, but anecdotal evidence f r o m Br i ta in indicates that 2nd generation teenage immigrants experience considerable tension between support ing immigra t ion and concern fo r the protect ion o f their host culture (Vasagar, 2001). Support for the inf luence o f univers i ty educat ion on attitudes toward i m m i g r a t i o n is prov ided by an analysis o f a national survey in the U S A . H a v i n g baccalaureate or graduate degrees is associated w i t h attitudes more favorable toward immig ra t i on (Haubert & Fussel l , 2004) . Another generational di f ference was that 2nd generation immigrants had lower B D W and PNS levels than 1st generation immigrants, and lower B D W levels than 3rd or greater generat ion immigrants . W h i l e no publ ished studies have examined immigran t generational di f ferences i n Threatening People & Places 26 B D W or PNS, i t may be that the chi ldren o f 1st generation immigrants to Canada ( w h o w o u l d have higher B D W and PNS levels) may develop a sense o f security and f l ex ib i l i t y comparat ive ly lower than their parents, as a result o f comparat ively more stable po l i t i ca l or economic condi t ions in Canada. A s their parents relate stories embedded w i t h higher dangerous w o r l d bel iefs or needs fo r structure, the ch i ldren may feel a comparat ive sense o f security and f lex ib i l i t y . The effect o f parents' B D W and N F S levels on the development o f their ch i ldren, i n bo th immig ran t and non- immigran t populat ions w o u l d be an interesting avenue o f research to pursue. Another demographic variable that affects the results is the presence o f the large p ropor t ion o f Chinese part icipants. Chinese part icipants were s igni f icant ly higher than Caucasians in B D W , and showed less var iab i l i ty i n the measure. Th is characteristic o f the sample makes the results less generalizable outside the populat ion sampled. One possible explanat ion for this di f ference in B D W scores and variance w o u l d be that socio-cul tural background differences affect h o w Caucasian and Chinese part icipants interpret the i tems on the scale. The B D W scale was developed w i t h Canadian univers i ty students, and there are no publ ished reports o f va l idat ion cross-cultural ly. I f ound a moderate posi t ive correlat ion between P A and N A , w h i c h may be related to the h igh p ropor t ion o f As ian part icipants (65%) in the sample. A s imi lar pattern o f correlat ions has been observed in previous research using the P A N A S w i t h As ian part icipants (S. J. Heine, personal communica t ion , M a y 12, 2004) , and may be explainable by a theory that people f r o m Chinese and other East As ian cultures are more adept at dialect ical th ink ing that a l lows the maintenance o f t w o contradictory posit ions on the same issue, wh i l e so-cal led Western th ink ing tends toward analyt ic th ink ing processes that polarize contradictory v iewpoin ts (cf. Nisbet t , Threatening People & Places 27 Peng, Cho i , & Norenzayan, 2 0 0 1 ; Peng & Nisbett , 1999). The degree to w h i c h this theory holds fo r affect w o u l d be a valuable avenue o f research to pursue. The present research demonstrated the moderat ing effects o f t w o ind iv idua l di f ference factors, B D W and PNS, on affect ive and cogni t ive responses to a threatening outgroup and si tuat ion. A l t h o u g h the results are qual i f ied by a sample that inc luded a disproport ionate number o f immigrants and ethnic As ian part icipants, the results ho ld , at least i n the popu la t ion studied. A n y addi t ional research should address the l imi tat ions o f this research. First, the doma in o f the vo t i ng task and the threatening outgroup slides should not be confounded. Second, the non-outgroup slide cond i t ion should include posi t ive ly and negat ively valenced slides to avo id un in tent ional ly induc ing posi t ive affect or reducing negative affect i n the non-threatening condi t ion . T h i r d , the va l id i ty and re l iab i l i ty o f the P A N A S measure should be studied among As ian immig ran t populat ions so comparisons w i t h non-As ian groups can be made more conf ident ly . F ina l l y , the use o f an immigrat ion-speci f ic integrat ive complex i ty measure m igh t help i l lustrate better the complex i ty -changing effects o f the threatening outgroups. Conclusion The ways we respond to threatening people and situations are l i ke ly to be evo lved, adaptive behaviors that have persisted for m i l l i ons o f years. W e have means fo r detect ing intergroup threats based, i n part, on h o w outgroup members deviate f r o m our cul tural norms and values, w h i c h w o u l d be an effect ive strategy when evaluat ing newcomers to a group. A s we encounter immigrants f r o m very di f ferent cultures, our affect and our cogn i t ion are go ing to be affected by these perceived threats, and we w i l l respond to cope w i t h the threats. One such w a y to cope is to categorize and reduce the complex i ty o f the in fo rmat ion we are fac ing about our env i ronment , a l l ow ing us to then evaluate and act on the threats more ef f ic ient ly . The desire for Threatening People & Places 28 structure may be one way we have evolved to cope w i t h the complex i ty o f i n fo rmat ion about outgroup threats. S imi la r ly , we may have evolved ways o f qu ick ly detecting and act ing on threats i n the envi ronment . A s we are faced w i t h situations o f vary ing degrees o f dangerousness, we must decide whether we need to be more v ig i lant , tax ing our l im i ted cogni t ive resources, or whether we can be less v ig i lant , and may devote more resources to other tasks. Chronic dangerous w o r l d bel iefs may be one way we have evolved to deal w i t h such contingencies. I f we can more e f f ic ient ly deal w i t h si tuational dangers, we can be effect ive cogni t ive managers (Suedfe ld , 1992), a l locat ing cogni t ive resources to where they are needed. W h e n we move these processes into the contemporary context, they serve us fo r better or worse. I n the present study, the moderat ion o f B D W and PNS was shown to apply i n specif ic condi t ions o f threat. When we are faced w i t h in fo rmat ion about an immigran t outgroup, w e may respond w i t h negative affect, especially i f we perceive a si tuational threat such as darkness. W e also may experience more or less cogni t ive s impl i f ica t ion, depending on our ind iv idua l B D W or PNS levels. Those o f us w i t h a h igh levels o f PNS may respond to immigran t outgroup members w i t h more s impl i f ied cogni t ions, w h i c h may occasion categorizat ion and group stereotypes. Those w i t h a h igh B D W level may experience more s impl i f ied cogn i t ion i f we perceive the double threat o f an immigran t outgroup and threatening situations. The threatening si tuat ion we used in the present research, darkness, is a potent, basic fear that el ici ts responses in a w ide range o f people. I t w o u l d be interesting to k n o w whether there are threats posed by immigran t outgroups that also tap such basic fears and e l ic i t cogni t ive and behavioral consequences. For example, migh t threats to resource avai labi l i ty (e.g. immigrants us ing up f resh water supplies or deplet ing f o o d stocks) t r igger such fears o f danger? I f so, I w o u l d expect that such a threat response w o u l d be moderated by B D W . Threatening People & Places 29 Ano ther interesting question unanswered by the present research is whether there are specif ic varieties o f immigran t outgroup threats that h igh PNS indiv iduals are more v ig i lan t about. One such specif ic threat may be the threat to social order posed by immigran ts ' d i f fe r ing bel iefs and cul tural practices. Another such threat m igh t be the conf l i c t ing , complex array o f benefi ts and costs to the host culture that are associated w i t h immigrants . H i g h PNS ind iv iduals may be mot iva ted to attempt to reduce the complex i ty o f this in fo rmat ion , by ignor ing some in fo rmat ion , or th ink ing more s impl is t ical ly . This experiment lends some credence to this hypothesis, but such a hypothesis w i l l need to be more fu l l y explored. A s we face a more g lobal ly integrated future, w i t h complex sets o f costs and benefi ts o f g lobal integrat ion, the issues o f ind iv idua l responses to immigran t outgroup threat that I have raised here w i l l become more cr i t ical . I t is important to k n o w h o w our responses are moderated by our dif ferences i n danger bel iefs and needs for structure. I f i t is true that these responses are the product o f m i l l i ons o f years o f evolut ionary history, the l i ke l ihood that we w i l l change our automatic responses is l i t t le or n i l . However , the awareness o f our propensi ty to make automatic at t r ibut ions, and their costs and benefits, w i l l a l l ow us to make better decisions about impor tant issues l ike immig ra t ion po l icy . Threatening People & Places 30 References A l temeyer , B. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco, C A : Jossey-Bass/Pfeiffer. ; B a k e r - B r o w n , G., Ba l la rd , E. J., B luck , S., de Vr ies , B., Suedfeld, P., & Tet lock, P. E. (1992). The conceptual/ integrat ive complex i ty scoring manual. I n C. P. Smi th (Ed.) , Motivation and personality: Handbook of thematic content analysis (pp. 400-418) . Cambr idge: Cambr idge Un ivers i ty Press. B a l d w i n , B. A . (1972). Change i n interpersonal cogni t ive complex i ty as a func t ion o f a t ra in ing group experience. Psychological Reports, 30(3), 935-940. B inder , N . E., Pol inard, J. L., & W r i n k l e , R. D. (1997). M e x i c a n A m e r i c a n and A n g l o attitudes toward immig ra t ion re fo rm: A v i e w f r o m the border. Social Science Quarterly, 78(2), 324-337. Butcher, K. F., & Piehl , A . M . (1998). Cross-ci ty evidence on the relat ionship between i m m i g r a t i o n and cr ime. Journal of Policy Analysis and Management, 17(3), 457-493. Byrne , A . , & Eysenck, M . W . (1995). Tra i t anxiety, anxious m o o d , and threat detect ion. Cognition & Emotion, 9(6), 549-562. Ca lvo , M . G. (2000). Ear ly v ig i lance and late avoidance o f threat processing: Repressive coping versus l o w / h i g h anxiety. Cognition & Emotion, 14(6), 763-787. Coren, S., & Suedfeld, P. (1990). A power test o f conceptual complex i ty : Textua l correlates. Journal of Applied Social Psychology, 20(5), 357-367. de Vr ies , B., & Walker , L. J. (1987). Conceptual/ integrat ive complex i ty and attitudes toward capital punishment. Personality & Social Psychology Bulletin, 13(4), 448-457. Threatening People & Places 31 Enoch, Y . (1994). The intolerance o f a tolerant people: Ethnic relations i n Denmark . Ethnic and Racial Studies, 17(2), 282-300. Eysenck, M . W . (1991). Tra i t anxiety and cogni t ion. I n C. D. Spielberger, I. G. Sarason & et al. (Eds.) , Stress and emotion: Anxiety, anger, and curiosity, Vol 14 (pp. 77-84) . Wash ing ton , D C : Hemisphere Publ ish ing Corp. Fiske, S. T. (1998). Stereotyping, prejudice, and discr iminat ion. I n D. T. Gi lber t , S. T. Fiske & et al . (Eds.), Handbook of social psychology, Vol 2 (4 th ed., pp. 357-411). N e w Y o r k , N Y : M c G r a w - H i l l . Fiske, S. T. (2000). Stereotyping, prejudice, and d iscr iminat ion at the seam between the centuries: Evo lu t ion , culture, m i n d , and brain. European Journal of Social Psychology, 30(3), 299-322. Fle ishman, J. (2004, A p r i l 18, 2004). Tolerance and fear col l ide i n the Netherlands. Los Angeles Times, p. A 9 . Florack, A . , P ion tkowsk i , U . , Bohman, A . , Balzer, T., & Perzig, S. (2003). Perceived intergroup threat and attitudes o f host commun i ty members toward immigran t accul turat ion. Journal of Social Psychology, 143(5), 633-648. Green, M . , W i l l i a m s , L. M . , & Davidson, D. (2003). I n the face o f danger: Specif ic v i e w i n g strategies for facial expressions o f threat? Cognition & Emotion, 17(5), 779-786. G r i l l o n , C , Pe l lowsk i , M . , Mer ikangas, K. R., & Davis , M . (1997). Darkness faci l i tates the acoustic startle ref lex i n humans. Biological Psychiatry, 42(6), 453-460. G u i l f o r d , J. P. (1965). Fundamental statistics in psychology and education (4 th ed.). N e w Y o r k : M c G r a w - H i l l B o o k Company. H a l l , G. S. (1897). A study o f fears. American Journal of Psychology, 8(2), 147-249. Threatening People & Places 32 Harr is , R. M . (1981). Conceptual complex i ty and preferred coping strategies i n ant ic ipat ion o f tempora l ly predictable and unpredictable threat. Journal of Personality & Social Psychology, 41(2), 380-390. Haubert , J., & Fussell , E. (2004). Social bases ofperceptions of immigrants: New class theory ' and pro-immigrant sentiment. Unpubl ished manuscript. Hogan , H. W . (1977). Complex i ty , ext remity, affect, and threat as dimensions o f person percept ion. Journal of Psychology, 96(2), 321-325. Junger-Tas, J. (2001). Ethnic minor i t ies, social integrat ion, and cr ime. European Journal on Criminal Policy and Research, 9, 5-29. K u n a , D. P. (1976) . Threatening behavior and complex judgments : A test o f the v ig i lance and Pol lyanna hypotheses. Representative Research in Social Psychology, 7(1), 6-12. Mears, D. P. (2001). The Immigra t ion -Cr ime Nexus: T o w a r d an Ana ly t i c F ramework fo r Assessing and Gu id ing Theory, Research, and Pol icy. Sociological Perspectives, 44(1), 19p. Neuberg, S. L., & N e w s o m , J. T. (1993). Personal need for structure: Ind iv idua l di f ferences i n the desire for s impler structure. Journal of Personality & Social Psychology, 65(\), 113-131. Nisbet t , R. E., Peng, K., Cho i , I., & Norenzayan, A . (2001). Cul ture and systems o f thought : Ho l is t ic versus analyt ic cogni t ion. Psychological Review, 108(2), 291-310. Ohman, A . , F lyk t , A . , & Lundqv is t , D. (2000). Unconscious emot ion : Evo lu t ionary perspectives, psychophysio logica l data and neuropsychological mechanisms. I n R. D. Lane & L. Nade l (Eds.) , Cognitive neuroscience of emotion (pp. 296-327) : O x f o r d Un ivers i ty Press. Threatening People & Places 33 Paulhus, D. L., & John, O. P. (1998). Egoist ic and moral ist ic biases i n sel f-percept ion: The interplay o f self-deceptive styles w i t h basic traits and mot ives. Journal of Personality, 66(6), 1025-1060. Peng, K., & Nisbet t , R. E. (1999). Cul ture, dialectics, and reasoning about contradict ion. American Psychologist, 54(9), 741-754. Pet t igrew, T. F., & Meertens, R. W . (1995). Subtle and blatant prejudice i n western Europe. European Journal of Social Psychology, 25(1) , 57-75. Porter, C. A . , & Suedfeld, P. (1981). Integrat ive complex i ty i n the corresponding o f l i terary f igures: Effects o f personal and societal stress. Journal of Personality & Social Psychology, 40(2), 321-330. Puente, M . (1993, July 14, 1993). U S A cool to huddled masses: Sentiment sours as rate o f arr ival rises. USA Today, p. 1A. Schaller, M . (2003). Ancestra l environments and mot ivated social percept ion: Goal - l i ke blasts f r o m the evolut ionary past. I n S. J. Spencer, S. Fe in & et al . (Eds.), Motivated social perception: The Ontario symposium, Vol 9 (pp. 215-231). M a h w a h , N J : Lawrence E r l b a u m Associates, Publishers. Schaller, M . , B o y d , C , Yohannes, J., & O'Br ien, M . (1995). The pre judiced personal i ty revis i ted: Personal need fo r structure and fo rmat ion o f erroneous group stereotypes. Journal of Personality & Social Psychology, 68(3), 544-555. Schaller, M . , Park, J. H. , & Muel le r , A r ( 2 0 0 3 ) . Fear o f the dark: Interact ive effects o f bel iefs about danger and ambient darkness on ethnic stereotypes. Personality & Social Psychology Bulletin, 29(5), 637-649. Threatening People & Places 34 Schroder, H. M . , Dr iver , M . J., & Streufert, S. (1967). Human information processing; individuals and groups functioning in complex social situations. N e w Y o r k : H o l t Rinehart and Wins ton . Shah, J. Y . , Brazy , P. C , & H igg ins , E. T. (2004). Promot ing us or prevent ing them: Regulatory focus and manifestat ions o f intergroup bias. Personality & Social Psychology Bulletin, 30(A), 433-446. Sil lars, A . , & Parry, D. (1982). Stress, cogni t ion, and communica t ion in interpersonal conf l ic ts . Communication Research, 9(2), 201-226. Streufert, S., & N o g a m i , G. Y . (1989). Cogni t ive style and complex i ty : Impl icat ions for I /O psychology. I n C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psychology 1989 (pp. 93-143). O x f o r d , England: John W i l e y & Sons. Streufert, S., & Schroder, H . M . (1965). Conceptual structure, env i ronmenta l complex i ty and task performance. Journal of Experimental Research in Personality, 7(2) , 132-137. Suedfeld, P. (1992). Cogni t ive managers and their cri t ics. Political Psychology, 13(3), 435-453. Suedfeld, P., & Granatstein, J. L. (1995). Leader complex i ty i n personal and professional crises: Concurrent and retrospective in fo rmat ion processing. Political Psychology, 16(3), 509-544. Suedfeld, P., Le igh ton , D. C , & Conway, L. G. ( in press). Integrat ive complex i ty and cogni t ive management i n international confrontat ions: Research and potent ial appl icat ions. I n M . F i t z d u f f & C. E. Stout (Eds.), Psychological approaches to dealing with conflict and war. Westport , CT : Praeger Publishers. Threatening People & Places 35 Suedfeld, P., & Tet lock, P. E. (2001). Ind iv idua l differences i n in fo rmat ion processing. I n A . Tesser & N . Schwarz (Eds.), Blackwell handbook of social psychology: Intraindividual processes (pp. 284-304) . Ma iden , M A : B lackwe l l Publishers, Inc. T i m m e r , A . S., & W i l l i a m s o n , J. G. (1998). Immig ra t i on po l icy pr io r to the 1930s: Labor markets, po l i cy interactions, and global izat ion backlash. Population and Development Review, 24(4), 739-771 . Un i ted States. Congress. House o f Representatives. (2003). Congressional Record. Retr ieved M a y 6, 2004, f r o m ht tp : / / thomas. loc .gOv /cg i -b iWquerv/C7r l08:7temp /M :108Ln7GDS Vasagar, J. ( 2 0 0 1 , M a y 23, 2001). Welcome to Br i ta in : Stuck in the middle. The Guardian, p. 6. Warr , M . (1990) . Dangerous situations: Social context and fear o f v ic t im iza t ion . Social Forces, 68(3), 891-907. Watson, D., Clark, L. A . , & Tel legen, A . (1988). Development and va l idat ion o f b r i e f measures o f posi t ive and negative affect: The P A N A S scales. Journal of Personality & Social Psychology, 54(6), 1063-1070. Threatening People & Places 36 Appendix Immigration Voting Preference Measure Voting Task Here are positions political candidates take on the issue of immigration. Please choose a single candidate you would vote for if you were voting on the basis of this one issue alone. Check the space before the position statement you would vote for. Priority should be given to Canadians when it comes to jobs and resources from this country. We have a small population, which cannot support immigrants living on welfare. Highly skilled workers with expertise that can't be found in Canada should be allowed to immigrate and better Canada's economy, but immigrants must not take jobs from Canadians. Immigration applications should be widely available, but the selection criteria for entering foreigners should be specifically defined. The government should allow people who are skilled workers or have significant business interests into Canada, but we should also allow for legitimate unskilled or unemployable persons to immigrate to Canada, such as refugees from political persecution (and in some cases, economic persecution). Refugees from war should be evaluated carefully but quickly. The Canadian economy sometimes needs highly skilled workers from other countries to fill jobs Canadians can't fill or don't want. Also, as global citizens, we have an obligation to help foreigners who are politically persecuted, or are refugees from countries at war. Only under these circumstances should our border be opened. Threatening People & Places 37 Canada should remain pure. Foreigners who come here end up stealing Canadian jobs or living on welfare. Canadian workers and taxpayers should not have to bear this burden. Under my administration, Canadians will not suffer from immigration. Refugees from regions of conflict, political instability, or political persecution should be granted immediate access to the country. Skilled workers, those who have family in Canada, and those who have business or investment interests should also be accepted. Also, we are a wealthy country, and can afford to harbor some immigrants who are unemployable, and those persecuted economically. Our borders should be open to all those who wish to enter or leave Canada. People should think of themselves as citizens of the world, and thus should be free to move from region to region without thinking about arbitrary lines. Workers who want to work in Canada and those who cannot work but contribute to our diversity should be welcomed. As Canadians, we have a responsibility to citizens of other nations of the world. We should share our resources by allowing all people fleeing from war, political persecution, political instability, and those who come from economically disadvantaged countries immediate access to Canada. All those who apply and are employable and will contribute something to Canada should also be allowed to enter. Threatening People & Places 38 Table 1 Means and Standard Deviations for All Dependent Variables Dependent Variables Budqet Slides Immiaration Slides S i a Effects Liaht (N=25) Dark (N=30) Liaht (N=34) Dark (N=35) Affect Negative Affect 1.37 (.36) 1.41 (.39) 1.47 (.44) 1.66 (.66) b ,d Positive Affect 2.00 (.86) 2.09 (.68) 2.02 (.72) 2.22 (.52) Voting Behavior Voting Preference 4.48 (1.16) 4.37 (1.27) 4.62 (1.39) 4.66 (1.37) Extremity .72(1.02) .83(1.02) .97 (1.17) 1.00 (1.14) Integrative Complexity Overall 1.95 (.50) 1.98 (.37) 1.87 (.51) 1.86 (.49) Interpersonal 2.06 (.57) 2.29 (.51) 1.88 (.63) 2.17 (.61) a Intrapersonal 1.86 (.70) 1.74 (.62) 1.81 (.64) 1.70 (.64) Institutional 1.90 (.76) 1.90 (.66) 1.93 (.65) 1.68 (.67) Note: Significant Effects Legend a = Illumination main effect b = Slide main effect c = Illumination x Slide interaction d = Contrast effect (Dark/Immigration vs. all other conditions) Threatening People & Places 39 Table 2 Correlations Between Belief in a Dangerous World (BDW) and All Dependent Variables /•for Budget Slides r f o r Immigration Slides Dependent Variable Overall r Light (N=25) Dark(N=30) Light (N=34) Dark(N=35) Power Sig. Effects Affect Voting Negative Affect .113 . 4 6 2 * .004 .017 .105 .250 a Positive Affect -.117 .290 -.172 -.05 -.236 .186 Behavior Voting Preference -.107 -.221 -.27 -.029 -.037 .055 Extremity .000 -.122 .099 .174 .152 .068 ative Complexity Overall -.202 * -.289 -.125 -.289 -.429 * .509 a , d Interpersonal -.01 .034 .056 .178 -.326 .327 Intrapersonal - . 1 8 8 * -.430 * -.011 -.093 -.306 .539 Institutional -.243 ** -.206 -.231 -.143 -.401 * .174 ** significance < .01 * significance < .05 Note. Significant Effects Legend (As Revealed By Regression) a = BDW main effect b = Illumination x BDW interaction c = Slide show x BDW interaction d = Illumination x Slide show x BDW interaction Threatening People & Places 40 Table 3 Correlations Between Personal Need for Structure (PNS) and All Dependent Variables Dependent Variable Overall r r f o rBudqe t Slides r for lmmiarat ion Slides Power Liaht (N=25i Dark (N=30) Liaht (N=34) Dark (N=35) Affect Negative Affect .100 . 4 1 6 * -.145 .031 -.008 .072 Positive Affect .028 .006 .006 -.290 .247 .249 Voting Behavior Voting Preference -.132 -.262 -.210 .006 -.103 .054 Extremity -.085 -.374 -.276 .104 .035 .056 Integrative Complexity Overall -.123 -.016 .215 -.346 * -.231 .053 Interpersonal -.069 .134 .068 -.266 -.176 .413 Intrapersonal -.093 -.123 .246 -.205 -.196 .172 Institutional -.108 -.035 .071 -.367 * -.151 .058 ** significance < .01 * significance < .05 Note. Significant Effects Legend (As Revealed By Regression) a = PNS main effect b = Illumination x PNS interaction c = Slide show x PNS interaction d = Illumination x Slide show x PNS interaction 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0091668/manifest

Comment

Related Items