Empathy in a Citizen Deliberation Experiment

Empa hy in a Ci izen Delibe a ion Expe imen Kimmo G € onlund*, Kaisa He ne † , and Maija Se € al€ a ‡ Despi e inc eased schola ly a en ion, he e is s ill limi ed knowledge on how empa hy wo ks in democ a ic delibe a ion. This a icle examines he ole o empa hy in ci izen delibe a ion wi h he help o a delibe a i e expe imen on immig a ion. Fi s , a andom sample o ci izens was su eyed ega ding hei opinions on immig a ion. Based on hei opinions, hey we e hen di ided in o a pe missi e o a non-pe missi e encla e, and andomly assigned in o like- minded o mixed-opinion g oups o delibe a ion. A e delibe a ion, hey we e su eyed anew. The s udy analyzes: (a) empa hy di e ences be ween pe missi e and non-pe missi e pa icipan s; (b) changes in ou g oup empa hy owa d immig an s as a esul o delibe a ion; and (c) di e ences in p osocial beha io (i.e., dona ing o cha i y). The esul s show ha he pe missi e esponden s had mo e empa hy, especially owa d immig an s, han he non- pe missi e esponden s. Among pa icipan s, ou g oup empa hy inc eased du ing delibe a ion. Rega ding p osocial beha io , he pe missi e pa icipan s dona ed mo e o en o cha i y a he end o he expe imen . In oduc ion Democ a ic delibe a ion appea s o be he key o in o med and ai collec- i e decision making. In democ a ic delibe a ion, pa icipan s’ awa eness o ac s and causali ies ela ed o poli ical issues a e expec ed o inc ease oge he wi h hei unde s anding o o he people’s pe spec i es. This s udy conce ns especially he la e aspec – ha is, he abili y o delibe a- o s o empa hize wi h o he people. We a e speci ically in e es ed in he ques ion o whe he he capaci y o conside he pe spec i es o o he s is a p econdi ion o delibe a ion, and o wha ex en delibe a ion imp o es his capaci y. * Kimmo G € onlund, P o esso o Poli ical Science, Social Science Resea ch Ins i u e, ASA- building, A ˚bo Akademi Uni e si y, 20500 A ˚bo, Finland. E-mail: [email p o ec ed] † School o Managemen , Uni e si y o Tampe e, Finland. ‡ Depa men o Philosophy, Con empo a y His o y and Poli ical Science, Uni e si y o Tu ku, Finland. This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 457 ISSN 0080-6757 Doi: 10.1111/1467-9477.12103 V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion While democ a ic delibe a ion is p ima ily a collec i e p ocess o ea- soning, i also en ails indi idual p ocesses o e lec ion o , as Robe Goodin (2000) desc ibes i , ‘delibe a ion wi hin’. In e nal delibe a ion equi es ha people a e e lec i e o e hei own iewpoin s as well as empa he ic owa d o he s’ posi ions. J€ u gen Habe mas (1996, 162), who mos ly emphasizes in e subjec i e aspec s, ega ds ‘ideal ole- aking’ as a pa o mo al discou se. The e m ‘ideal ole- aking’, o iginally in oduced by Mead (1934), e e s o a p ocess o placing onesel in ano he ’s posi ion and ying o unde s and hei pe spec i e. In Habe mas’s discou se e hics, ole- aking is needed o ack whe he o he s ha e a gene alizable a i ude abou he alidi y o a no m (Mo ell 2010, 79). Recen ly, he e has been inc easing a en ion paid o he ole o a ec in democ a ic delibe a ion (e.g., K ause 2008). In his book Empa hy and Democ acy, Michael Mo ell (2010) c i icizes hose unde s andings o delib- e a i e democ acy ha lean oo hea ily on a ionali y and he easonableness o a gumen a ion. Mo eo e , he a gues ha democ a ic delibe a ion needs empa hy, unde s ood as a p ocess ha in ol es bo h a ec i e and cogni i e componen s. Mo ell (2010, 114) w i es: [W]i hou he p ocess o empa hy, i is highly unlikely ha ci izens will demons a e he ole a ion, mu ual espec , ecip oci y, and openness owa d o he s i al o delibe a i e democ acy o ul ill i s p omise o equal conside a ion ha is cen al o gi ing collec i e decisions hei legi imacy. In his a icle, we examine he ole o empa hy in a ci izen delibe a ion expe imen . Ou expe imen al design p o ides a way o es he e ec s o delibe a ion on pa icipan s’ le els o empa hy. We ha e also included a speci ic design ea u e allowing us o es al uis ic beha io among pa ici- pan s. Ou expe imen is designed o compa e so-called ‘encla e delibe a- ion’ (i.e., delibe a ion among like-minded people) wi h delibe a ion in c oss-cu ing g oups. The opic o he discussion – immig a ion poli ics – is ideal o es ing he causal link be ween delibe a ion and empa hy since delibe a ion on immig a ion seems o equi e conce n o o he s’ posi- ions, especially ega ding ou g oups. In ou empi ical analysis, we i s pay a en ion o he po en ial di e en- ces be ween hose who we e willing o pa icipa e in democ a ic delibe a- ion and hose who we e no . We e hose who pa icipa ed mo e empa he ic han hose who d opped ou ? We also look a he po en ial o delibe a ion o inc ease empa hy, especially owa ds membe s o an ou - g oup, which in he p esen con ex is immig an s. We in es iga e whe he ou expe imen al ea men in luences le els o empa hy by examining di - e ences be ween like-minded and mixed g oups. Finally, we examine p oneness o al uis ic beha io in di e en g oups. Al uis ic beha io is ope a ionalized h ough an ac o dona ing money o cha i y. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 458 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 The a icle is o ganized as ollows. Fi s , we p o ide a de ini ion o empa- hy, and conside i s ole in democ a ic delibe a ion. We hen p esen ou hypo heses and he expe imen al p ocedu es, which a e ollowed by he esul s and concluding ema ks. Empa hy as an Elemen o Democ a ic Delibe a ion Empa hy, unde s ood as a capaci y and mo i a ion o pu onesel in he posi ion o o he s, seems o be a necessa y elemen in he p ocess o delibe a ion. Al hough many heo is s ha e hough abou he ole o empa hy, he e a e no many empi ical s udies on empa hy, o ‘ideal ole- aking’, in he con ex o democ a ic delibe a ion. Apa om some ai ly ecen con ibu ions in his a ea (e.g., Mo ell 2010), empa hy is la gely neglec ed in s udies o delibe a ion. The e m ‘empa hy’ has been de ined in a numbe o ways, and no gene - ally sha ed de ini ion o he e m exis s (e.g., Ho man 2000; P es on & De Waal 2002; Wal e 2012). Fo his eason, he discussion on empa hy is o en elusi e (Mo ell 2010, 39–66). Fu he mo e, he e a e concep s such as ‘sym- pa hy’, ‘compassion’ and ‘empa he ic conce n’ ha a e closely ela ed o ‘empa hy’, bo h concep ually and empi ically. Some w i e s conside empa- hy mainly as an a ec i e p ocess whe e o he s’ emo ions a e ep oduced. O he s, in u n, make a dis inc ion be ween a ec i e and cogni i e elemen s o empa hy whe e he la e e e s o an unde s anding o o he s’ posi ions (Wal e 2012). Da is (1980) a gues ha empa hy is a mul idimensional con- cep and ha he eisaneed omeasu ebo hcogni i eanda ec i easpec s o empa hy (i.e., a pe spec i e- aking capaci y and emo ional eac i i y). Democ a ic delibe a ion, especially i s idea o equal conside a ion o all iewpoin s based on hei me i s, clea ly equi es empa hy, in e p e ed as a cogni i e p ocess o unde s anding o he s’ pe spec i es. No ably, Rawls’s (1971) concep o ‘public eason’ equi es ha people a e open o o he s’ pe spec i es and eady o jus i y hei own claims in e ms accep able o o he s. Habe mas’s (1996) idea ha he alidi y o mo al no ms is de ined h ough he uni e saliza ion es , which examines wha is accep able o all, is a simila concep . I seems ha bo h Rawls’s public eason and he ul illmen o Habe mas’s uni e saliza ion es equi e an unde s anding o o he s’ social posi ions and in e es s. Such an unde s anding is necessa y o de e mine wha kind o e ms o coope a ion can be accep able o o he s. Delibe a i e democ acy calls o an unde s anding o hose aspec s o o he people’s social posi ions ha a e ‘e hically ele an ’ – ha is, hose pe aining o gene alizable mo al p inciples. These e hically ele an aspec s could be in e p e ed as being ela ed o wha Rawls (1971) calls V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 459 ‘p ima y goods’. In he p ocess o delibe a ion on income dis ibu ion, o ins ance, i seems ele an o unde s and ha li ing wi h less han a mini- mum income can lead o poo nu i ion. Au ho s such as K ause (2008) and Mo ell (2010) ha e a gued ha emo- ional componen s o empa hy a e also ele an o democ a ic delibe a ion. While cogni i e aspec s o empa hy pe ain o he p ocess o gaining knowl- edge abou o he s’ posi ions as well as p e e ences and alues ela ed o hem, he emo ional aspec o empa hy e e s o a ec ions ela ed o hose posi ions. Emo ional empa hy equi es he ep oduc ion o he a ge ’s emo- ions based on one’s own an icipa ed emo ions in he posi ion o he a ge o al e na i ely on he obse ed o imagined emo ions o he a ge . This does no mean ha he subjec has o expe ience exac ly he same emo ions as he a ge ; a he , he subjec should eel some hing ha is in line wi h he a ge ’s emo ions. In his espec , emo ional empa hy is simila o empa he ic conce n o sympa hy. Ba son (2011), o example, a gues ha empa he ic conce n in ol es a a ie y o eelings, including sympa hy and compassion. Unde s ood in his way, emo ional empa hy seems ele an o democ a ic delibe a ion because i is likely o be linked o p o-social a i udes. We would no ega d a pe son as empa he ic i she o he uses hei unde s anding o wha causes pain o ha m o he s; no can we see how his kind o an a i ude would help democ a ic delibe a ion. In addi ion o objec i e social posi ions, unde s anding o ce ain kinds o emo ions ela ed o hose posi ions can be ega ded as e hically ele- an in democ a ic delibe a ion. Emo ional empa hy migh be needed, o example, o unde s and ha a pe son would eel dis espec ed as a ci izen i denied he igh o o e, o ang y o us a ed i ea ed un ai ly by public o icials. Ye , he e a e also good easons o being c i ical o he iew ha democ a ic delibe a ion would equi e ep oducing o he s’ emo- ions. Fi s , people can misunde s and o he s’ emo ions. Second, people may mis ep esen hei own emo ions in o de o enhance be e ou comes o hemsel es, al hough his would iola e Habe mas’s since i y condi ion. Thi d, he e is he p oblem o mo ally ‘w ong’ emo ions – o example, ‘ he pain he acis eels in he ace o an i-disc imina ion laws’ (K ause 2008, 166). Such an emo ion is ha dly jus i iable om a mo e gene alizable pe spec i e, which should be he ou come o a delibe a i e p ocess. Al hough he ole o a ec i e aspec s o empa hy is no s aigh o wa d, Mo ell (2010) claims ha open-mindedness and sensi i i y o o he s’ emo ions a e likely o be empi ically ela ed o he sense o ecip oci y and commi men o delibe a ion. In o he wo ds, i is he a ec i e ele- men o empa hy ha mo i a es conside a ions o o he people’s posi ions. I his is he case, a ec i e componen s o empa hy acili a e o can e en be a p econdi ion o he cogni i e ype o empa hy. Following Da is’s mul idimensional model o empa hy, Mo ell (2010, 60–3) p omo es a V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 460 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 ‘p ocess model o empa hy’ ha emphasizes in e ac ion be ween a ec i e and cogni i e elemen s. F om he pe spec i e o democ a ic delibe a ion, i is, howe e , impo an ha emo ional eac ions, whe he one’s own o o he s’, a e assessed in e ms o he e hical ele ance and gene alizabili y o he claims ela ed o hem. The unde s anding o he a ec i e componen s o empa hy helps o ecog- nize po en ial sou ces o biases in empa hizing. Such biases can be p oblem- a ic o he idea o equal conside a ion o he pe spec i es o all a ec ed, which is a cen al poin in democ a ic delibe a ion. Social psychological li e - a u e p o ides a numbe o examples o empa hy biases. People a e mo e likely o empa hize wi h hose who a e simila o close o hem han wi h membe s o socially mo e dis an ou g oups ( o e e ences, see Cika a e al. 2014). In o he wo ds, iends, ela i es and o he in-g oup membe s a e mo e likely o gene a e empa hy, which may lead o biased judgmen s. A sin- gle pa en unde s ands he posi ion o ano he single pa en ; a disabled pe - son is amilia wi h he speci ic p oblems o disabled people; and so on. Empa hy biases ha e also been obse ed in empi ical s udies. Fo example, judges who ha e daugh e s a e mo e likely o ake a eminis s andpoin in hei decisions compa ed o judges who ha e only sons, sugges ing ha he o me unde s and women’s pe spec i es be e han he la e (Glynn & Sen 2015). While people end o unde s and ing oup membe s’ pe spec i es, ou g oup membe s, in u n, a e no as likely o gi e ise o empa hy. In a con- lic si ua ion, compassion is lowe o an indi idual who ep esen s he o he side in a con lic (B uneau e al. 2012). Democ a ic delibe a ion, unde s ood as an in e subjec i e p ocess whe e di e en iewpoin s a e pu o wa d o balanced conside a ion, can be ega ded as a possible emedy o empa hy biases. Fo example, Mu z (2002) has shown ha exposu e o c oss-cu ing iews, which is ypical in p ocesses o democ a ic delibe a ion, helps o unde s and hings om ano he ’s pe - spec i e. I can be assumed ha inclusi e p ocesses o delibe a ion, whe e people a e encou aged o conside o he s’ posi ions, could enhance empa hy owa d ou g oups, especially i hey appea o be in a disad an aged posi ion. In his s udy, we sepa a e gene al empa hy om empa hy owa d a speci ic ou g oup – in his case, immig an s. ‘Gene al empa hy’ e e s o cogni i e and a ec i e empa hy el owa ds people in gene al, whe eas ‘ou g oup empa hy’ e e s o empa hy el owa ds immig an s. In addi ion o he ole o empa hy in he delibe a i e p ocess, we also s udy he connec ion be ween empa hy and p osocial beha io in he con- ex o democ a ic delibe a ion. I appea s ha an empa he ic unde s and- ing o someone else’s posi ion is no au oma ically linked o p osocial beha io . Binmo e’s (1998) example is a gun igh e who uses ‘his empa- he ic powe s o p edic an opponen ’s nex mo e wi hou losing he u ge o kill him’. Howe e , empi ical s udies in social psychology show ha V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 461 empa hy o sympa hy usually ends o p omo e p osocial ac ion (e.g., Mon ada & Schneide 1989; Pagano & Huo 2007). Fu he mo e, pe spec i e- aking abili ies lea ned in one con ex can ex end o o he con ex s (Ba son 2011, 175–6). S udies gi e suppo o he empa hy-al uism hypo hesis, which s a es ha empa he ic conce n e okes an al uis ic mo i a ion o help o he s – ha is, gi es ise o p osocial ac ion (e.g., Ba son 1997). Se e al s udies show ha inducing empa hy inc eases he willingness o help o ac ual helping o o he s, e en when he ac o help- ing is cos ly (e.g., Ba son 1997; Ba son e al. 1997, 2002). Expe imen al e i- dence also sugges s ha al uis ic dona ions migh be ela ed o he dono s’ belie s abou he esponden s’ wo hiness o help (Fong 2007). In sum, he e is a heo e ical link be ween democ a ic delibe a ion and empa hy, including empa hy owa ds ou g oups. Fu he mo e, empa hy is empi ically linked o p osocial a i udes o beha io . While empa hy is heo e ically connec ed o delibe a i e discussion, we a e no awa e o p e ious empi ical s udies on he connec ion be ween delibe a i e discus- sion and empa hy. Mo eo e , he e a e no p e ious s udies on he in lu- ence o g oup composi ion on empa hy in he con ex o delibe a ion. The e a e a numbe o s udies showing ha pa icipa ion in delibe a i e discussion leads o opinion changes (e.g., Luskin e al. 2002; Ba abas 2004; Fishkin & Luskin 2005; Se € al€ ae al.2010;G € onlund e al. 2015), as well as o o he expec ed consequences o delibe a ion, such as inc eases in poli ical knowledge, e icacy, us and p epa edness o poli ical pa icipa- ion (e.g., G € onlund e al. 2010; Himmel oos e al. 2017). None o hese empi ical s udies looks a he connec ion be ween delibe a ion and empa- hy. G € onlund e al. (2015) is based on he same expe imen as he one in his a icle, bu i s udies opinion changes and obse es ha people wi h an i-immig an a i udes become mo e ole an e en when hey discuss in like-minded g oups wi h delibe a i e ules and acili a ion. This change in opinions canno only be explained by gaining mo e in o ma ion on he immig a ion issue du ing delibe a ion. Finally, al hough he empa hy- al uism hypo hesis has been es ed in a numbe o social psychological s udies, we a e no awa e o any o hem connec ing he hypo hesis o democ a ic delibe a ion. Howe e , ea lie s udies sugges ha pa icipa- ion in o ganized delibe a ion enhances gene alized us and a willingness o con ibu e o collec i e ac ion (Se € al€ a e al. 2010; G € onlund e al. 2010). Delibe a ion on Immig a ion: Hypo heses on Empa hy To sum up he heo e ical discussion, empa hy – unde s ood as a willing- ness o conside o he s’ pe spec i es and emo ions – can be ega ded as a V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 462 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 p econdi ion o democ a ic delibe a ion, and i can also be expec ed o de elop in he cou se o delibe a ion. The ole o empa hy may a y acco ding o he ype o issue; i can be expec ed o be pa icula ly impo - an when delibe a ing abou an issue such as immig a ion, which calls o conside a ion o o he people’s pe spec i es. Those capable o empa hiz- ing a e likely o be mo i a ed o delibe a e on immig a ion because hey a e willing and able o unde s and o he s’ pe spec i es and emo ions. Mo eo e , success ul delibe a ion is expec ed o enhance unde s anding o o he s’ pe spec i es, especially hose in a disad an aged posi ion. In his s udy, we compa e he ole o empa hy in encla e delibe a ion – ha is, delibe a ion among a like-minded g oup – o c oss-cu ing delibe - a ion whe e g oup membe s hold di e en iews. The inclusion o di e - en iewpoin s in he p ocess o exchanging a gumen s is one o he key ea u es in democ a ic delibe a ion. Indeed, he p esence o con lic ing iewpoin s is o en ega ded as a necessa y p econdi ion o delibe a ion (Thompson 2008, 502). Di e si y o opinions can be expec ed o enhance conside a ion and e lec ion on di e en iewpoin s (e.g., Me cie & Landemo e 2012). Cass Suns ein (2002, 2007, 2009) discusses he p oblems and isks ela ed o encla e delibe a ion, including, mos impo an ly, g oup pola iza ion. O he schola s (e.g., Ka powi z e al. 2009; see also Suns ein 2007, 76–7) poin ou ha delibe a ion in like-minded con ex s migh acili a e a icula ion o speci ic in e es s and alues as well as enhance mobiliza ion, especially o hose in disad an aged posi ions. Mansb idge (1994, 63), who was p obably he i s o use he e m ‘encla e delibe a ion’, calls o ‘encla es o p o ec ed discou se and ac ion’ as an elemen o a jus socie y. In ou expe imen , he esponden s we e di ided in o wo encla es based on hei opinions on immig a ion exp essed in a baseline su ey (T1). Responden s wi h nega i e a i udes o immig a ion we e classi ied in o a con encla e, and esponden s wi h a posi i e iew on immig a ion in o a p o encla e. Fu he mo e, pa icipan s we e andomly assigned in o: like-minded g oups consis ing o people om he same encla e; mixed g oups, including an equal numbe o pa icipan s om bo h encla es; and a con ol g oup. Subjec s in he i s wo g oups ook pa in he delibe a ion e en , whe eas he con ol g oup only illed in h ee mail- in su eys. The pu pose o ou expe imen was o compa e p ocesses and consequences o delibe a ion in like-minded and mixed g oups. 1 Al hough g oup composi ion was subjec o manipula ion, he o he s anda d p oce- du es o he delibe a i e ‘package’ (Mu z 2006, 61) we e applied. These p ocedu al ea u es include he p o ision o balanced in o ma ion o all pa icipan s as well as he use o mode a o s and discussion ules o encou age a p ocess whe e pa icipan s lis en o each o he ’s a gumen s and e lec on hei own iews. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 463 Empa hy was measu ed h ough su eys conduc ed among he expe imen- al subjec s be o e and a e delibe a ion (a T2 and T4). To measu e gene al empa hy, we used h ee i ems om Da is’s (1980) la ge empa hy ques ion- nai e. We ocused on i ems o pe spec i e- aking and empa he ic conce n, which we conside ele an om he poin o iew o democ a ic delibe a- ion. 2 In addi ion o gene al empa hy, we also analyzed ques ions designed o he speci ic pu poses o his s udy. These ques ions measu e pa icipan s’ willingness o conside he iews o a speci ic ou g oup (i.e., immig an s). As poin ed ou in he heo e ical discussion, we conside empa hy as a ac o acili a ing success ul delibe a ion. Based on his, we expec ha hose who we e ac ually willing o pa icipa e in he delibe a i e expe i- men a emo eempa he ic han hosewhod oppedou . Since he pa icipan s we e di ided in o p o and con encla es, we a e able o compa e empa hy in hese wo g oups. Based on ea lie s udies (e.g., Newman e al. 2015), we expec people in he p o encla e o be mo e empa he ic when i comes o ou g oup empa hy, ope a ionalized as a willingness o conside he pe spec i es o immig an s. Fu he mo e, ou expe imen al subjec s delibe a ed ei he in like- minded o in mixed g oups, which allows us o analyze he e ec s o he exposu e o con lic ing iewpoin s in he cou se o delibe a ion. The di e - si y o iewpoin s, pu o wa d in a delibe a i e p ocess, is expec ed o help pa icipan s o see hings om o he people’s pe spec i es, including ou g oups such as immig an s. This kind o de elopmen is expec ed o ake place especially among hose pa icipan s who ha e nega i e a i udes owa d immig a ion and who delibe a e in mixed g oups. This is because hese pa icipan s a e likely o hea a gumen s and iewpoin s in suppo o mo e pe missi e iews owa ds immig a ion. Ano he issue is ha acco ding o Suns ein’s (2002) ‘law o g oup pola iza ion’, people delibe - a ing in like-minded g oups can become mo e ex eme in hei iews. Especially, pa icipan s wi h an i-immig a ion a i udes can be expec ed o ein o ce each o he ’s nega i e a i udes owa d immig an s, which may dec ease hei willingness o conside immig an s’ pe spec i es. In addi ion o he su ey da a measu ing empa hy, ou s udy yields e i- dence on he pa icipan s’ p oneness o beha e al uis ically. A he end o he expe imen , he pa icipan s had he possibili y o dona e e5o hei e90 emune a ion o he Red C oss. Mo eo e , hey could choose be ween domes ic o in e na ional cha i y. Based on ea lie indings on empa hy and al uism, we assume ha membe s o he p o encla e a e mo e likely o con ibu e o cha i y compa ed o he membe s o he con encla e. As people in he p o encla e can be expec ed o be mo e empa he ic owa d ou g oups, hey can also be assumed o be mo e willing o con ibu e o in e na ional aid han hose belonging o he con encla e. On he basis o he p e ious discussion, we o mula e he ollowing hypo heses. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 464 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 H1: Empa hy, especially empa hy owa d ou g oups, is highe among he pa icipan s in he p o encla e han in he con encla e. H2a: Empa hy owa d ou g oups inc eases du ing delibe a ion. H2b: The la ges inc ease in ou g oup empa hy occu s among hose pa icipan s in he con encla e who delibe a e in mixed g oups. H3: Pa icipan s in he p o encla e dona e mo e o en o he Red C oss han pa icipan s in he con encla e, especially o he In e na ional Ca as ophe Fund. Expe imen al P ocedu es Ou analysis is based on an expe imen whe e ci izens, d awn om a an- dom sample, delibe a ed on immig a ion. The expe imen was held in Fin- land in 2012. A sho su ey (T1) was i s mailed ou o a simple andom sample o 12,000 adul s in he egion o Tu ku (A ˚bo). O he add essed sample, 39 pe cen (n54,681) esponded o he su ey. T1 consis ed o 14 ques ions measu ing esponden s’ immig a ion a i udes. The ques ions a e lis ed in he Appendix. Figu e 1 shows he ini ial dispe sion o a i udes among hose esponden s (n 53,232) who allowed u he con ac om he esea ch g oup. 3 Figu e 1. The Ini ial Dispe sion o A i udes owa d Immig a ion. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 465 ea men s and encla es, we ind ha he sligh inc ease among he con pa icipan s in he like-minded g oups is no signi ican , whe eas he inc ease om 0.39 o 0.47 among he con pa icipan s in mixed g oups is signi ican ( 53.98, p 50.000). The inc ease om 0.67 o 0.71 in he p o like- minded g oups is signi ican ( 52.63, p <0.01), whe eas he mino dec ease among he p o pa icipan s in mixed g oups is no ( 521.51, p <0.14). The esul s seem o con i m ou expec a ion ha aking pa in delibe a ion on immig a ion inc eases conce n o immig an s’ pe spec i es (H2a). Mo e- o e , he inc ease seems o be especially la ge among pa icipan s in he con encla e delibe a ing in mixed g oups (H2b). 12 Ne e heless, when we look a he di e ences-in-di e ence, we ind ha he con pa icipan s’ la ge inc ease in ou g oup empa hy in he mixed ea men is no s a is ically highe han he inc ease in he like-minded ea men . This inding was also con i med h ough sepa a e eg ession analyses wi hin he wo encla es (no shown he e) whe e he de elopmen o a i udes owa d immig a ion was added o he dummy a iables o ea men , ime (pos -delibe a ion) and ea men * ime. Wi hin he con encla e, only he inc ease o ole an a i udes explains inc ease in ou g oup empa hy. Wi hin he p o encla e, ou g oup empa hy was ound o inc ease by he like-minded ea men bu no by change in a i udes owa d immig a ion. In he con ol g oup, ou g oup empa hy did no change. In conclusion, H2a gains suppo , bu he inc ease occu s among all pa icipan s, wi h he excep ion o he p o- immig a ion pa icipan s in he mixed ea men . Thus, we canno con i m H2b, s a ing ha he pa icipan s wi h non-pe missi e iews on immig a ion would ha e gained mo e ou g oup empa hy in mixed g oups. These esul s a e in line wi h he indings by G € onlund e al. (2015), which show an o e all de elopmen in o mo e pe missi e a i udes owa d immig a- ion among he pa icipan s. A i udes we e measu ed wi h 14 i ems such as ‘Finland should ake mo e immig an s. Do you hink his is a bad o a good sugges ion?’. De elopmen in o mo e pe missi e a i udes happened in bo h ea men s. Also a he indi idual le el, he change in ou g oup empa hy is posi i ely co ela ed wi h he change in immig a ion a i udes owa d a mo e ole an iew (Kendall’s au 0.183, p 50.000). Mo eo e , i is no able ha , on a e age, he con pa icipan s delibe a ing in like-minded g oups became mo e ole an o immig a ion, which is con a y o ea lie indings on discus- sion in like-minded g oups, labeled by Suns ein (2002) as g oup pola iza ion. Ou esul s suppo he iew ha delibe a ion is di e en om o he o ms o alk, and ha discussion p ocedu es can ha e a s ong impac on ou - comes. The ‘delibe a i e package’, including in o ma ion and acili a ed small g oup discussions, seems o ha e an impac on how g oups discuss and how a i udes de elop (G € onlund e al. 2015; S andbe g e al. 2017). So a , he analysis has been based on a i udinal i ems. We will now u n o al uis ic beha io among he pa icipan s (H3). A he end o he V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 472 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 delibe a ion, he pa icipan s had an oppo uni y o dona e e5o hei e90 emune a ion o he Finnish Red C oss. This oppo uni y o dona e was nei he amed no p imed in any way; and he pa icipan s we e no old abou his in ad ance. The dona ion i sel was included as a ques ion a he end o Su ey T4, and he pa icipan s could make hei dona ion by icking a box in he ques ionnai e. In his espec , he dona ion was en i ely a he pa icipan s’ disc e ion and he possible in e en ion o social desi abili y o g oup p essu e was minimized. Figu e 3 shows he dis ibu ion o dona ions wi hin he wo encla es and ea men s. 13 As Figu e 3 shows, he al uis ic ac o dona ing was clea ly mo e com- mon in he p o encla e. Hal o he pa icipan s in he p o encla e made he dona ion, whe eas he p opo ion o hose who dona ed in he con encla e was 28 pe cen . The di e ence is s a is ically signi ican ( 53.13, p5001). 14 A he indi idual le el, he e is a weak posi i e co ela ion be ween gene al baseline empa hy and dona ing (Kendall’s au 0.13, p<0.05). The bi a ia e co ela ion be ween ou g oup empa hy (a e he expe imen ) and dona ing is sligh ly s onge and s a is ically signi ican (Kendall’s au 0.19, p <0.01). Among hose who dona ed, pa icipan s in he p o encla e we e mo e likely o gi e money o he In e na ional Ca as ophe Fund han hose in he con encla e, as expec ed. In he p o encla e, 47 indi iduals (77.0 pe cen o he dona o s wi hin he encla e) dona ed o he ca as ophe und, compa ed o 14 pe sons in he con encla e (58.3 pe cen ). The numbe o dona o s o he domes ic und was 14 in he p o and 10 in he con encla e. The di e ences in whe e people Figu e 3. The Dis ibu ion o Dona ions o he Red C oss a he End o he Expe imen . V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 473 dona ed o is no , howe e , s a is ically signi ican be ween he wo encla es ( 51.61, p 50.116). Rega ding he di e ences in e ms o dona ing be ween he wo ea men s, he di e ence is signi ican wi hin he p o encla e bu no wi hin he con encla e. When p o-immig a ion people discussed wi h each o he , his seems o ha e a ec ed hei willingness o dona e o cha i y in a signi ican way: 43 pe sons dona ed, 33 did no , compa ed o 17 dona ing in mixed g oups and 27 no dona ing. ( 51.97, p 50.051). 15 Somewha su p isingly, con pa icipan s delibe a ing in like-minded g oups dona ed (14 dona ed, 28 did no ) mo e o en han hei pee s discussing in mixed g oups ( en dona ed, 34 did no ), al hough his di e ence is no s a is ically signi ican ( 51.09, p 50.28). Wi hin he con encla e, Figu e 3 would sugges ha he e is a ia ion in he ex en o which dona ions a e a ge ed o domes ic o in e na ional aid. How- e e , hese di e ences a e no s a is ically signi ican ( he di e ence is close o being signi ican in he like-minded ea men , p <0.12). 16 In o de o e i y he pa e ns ound in bi a ia e analyses be ween empa hy and p osocial beha io , we also conduc ed a bina y logis ic eg ession analysis whe e empa hy, ou g oup empa hy, change in ou g oup empa hy and he p esence o a pa icipan wi h an immig an backg ound in he small g oup we e included as co a ia es. Table 6 shows he esul s o he eg ession. The only s a is ically signi ican a iable is baseline ou - g oup empa hy, which has an odds a io o o e 13. O e all, he model does no p edic dona ions e y well (62 pe cen co ec ). In o he wo ds, we did no ind a link be ween gene al empa hy and al uism, which has been es ablished in some o he s udies (Ba son 1997; Ba son e al. 1997, 2002), bu he willingness o conside immig an s’ iews inc eased he like- lihood o al uis ic dona ing, as assumed by H3. Conclusions I has been con incingly a gued ha bo h cogni i e and a ec i e aspec s o empa hy a e essen ial elemen s o democ a ic delibe a ion (e.g., Table 6. Bina y Logis ic Reg ession B Odds a io S.E. p Gene al baseline empa hy 0.83 2.30 1.10 0.450 Ou g oup baseline empa hy 2.61 13.59 0.98 0.007 Ou g oup empa hy change 0.92 2.50 1.27 0.472 Immig an in small g oup 0.17 1.19 0.30 0.557 Cons an 22.54 0.08 0.90 0.005 Nagelke ke pseudo R 2 0.075 % co ec ly p edic ed 62.3 No es: Dependen a iable: Dona ed o he Red C oss. S.E. 5S anda d e o . V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 474 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 K ause 2008; Mo ell 2010). In addi ion, i seems plausible o assume ha he ole o empa hy a ies acco ding o he ype o issue delibe a ed – he issue o immig a ion is ce ainly an issue ha equi es an unde s anding o o he s people’s posi ions (c . Newman e al. 2015). Mo eo e , delibe a ion abou immig a ion can be expec ed o aise conce ns ega ding he pe - spec i es o he ele an ou g oup (i.e., immig an s). Ou analyses show ha esponden s wi h p o-immig a ion a i udes had a somewha highe baseline le el o gene al empa hy han espond- en s wi h es ic i e a i udes owa d immig a ion. Gene al empa hy was no , on he o he hand, connec ed wi h a i ion in he expe imen . Empa he ic people we e no mo e inclined o ake pa in delibe a ions on immig a ion, and we did no ind signi ican changes in he pa ici- pan s’ gene al empa hy le els in he cou se o he expe imen . Nei he delibe a ing in small-n g oups, no being subjec o he like-minded o mixed ea men had any e ec on he pa icipan s’ gene al empa hy a he agg ega e le el. The e we e, howe e , mo e p ominen di e ences in he subjec s’ will- ingness o ake in o accoun immig an s’ pe spec i es. This second index, which was c ea ed as a p oxy o ou g oup empa hy, consis s o h ee i ems on esponden s ecep i i y owa d immig an s’ iewpoin s. The p o- immig a ion encla e was a mo e willing o conside immig an s’ iews han he less pe missi e con encla e. I is also no able ha hose con encla e esponden s who we e less willing o conside immig an s’ pe - spec i es we e also less p one o pa icipa e in delibe a ion. Among hose who pa icipa ed in delibe a ion, he e was an inc ease in ou g oup empa- hy wi hin he con encla e. E en hough he inc ease seemed o be mo e e iden in he mixed ea men , he di e ence-in-di e ence analysis showed no s a is ically signi ican di e ence be ween he ea men s. The esul s sugges ha people whose ini ial a i udes owa d immig a ion we e es ic i e we e especially a ec ed by aking pa in delibe a ions in g oups wi h mixed opinions; hei willingness o unde s and immig an s’ pe spec i es inc eased no ably. Also in he like-minded ea men , delibe - a ion led o somewha inc eased ou g oup empa hy, especially wi hin he p o encla e In o he wo ds, delibe a ion on he immig a ion issue seems o ha e enhanced a speci ic ype o pe spec i e- aking. Because ou measu e o ou g oup empa hy only aps in o cogni i e aspec s o empa hy, i emains possible ha he inc eased ecep i i y o immig an s’ pe spec i es was ins umen al o e en sel -in e es ed in cha ac e . Fo example, pa icipan s could ha e s a ed o hink ha immig an s’ iewpoin s should be consid- e ed mo e ca e ully in o de o educe he isks o e o ism. Howe e , he ac ha g ea e willingness o hea immig an s’ iewpoin s was associ- a ed wi h an inc eased pe missi eness (i.e., ole an a i udes) owa ds V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 475 immig a ion, especially wi hin he con encla e, sugges s a mo e p o ound change in pa icipan s’ a i udes. When i comes o he beha io al dependen a iable – ha is, whe he o dona e o cha i y o no – we ound in acco dance wi h H3 ha he an i-immig an con encla e was less willing o dona e a po ion o hei emune a ion o he Red C oss. Pa icipan s in he con like-minded g oups dona ed much less ( wo hi ds did no dona e) han pa icipan s in he p o like-minded g oups (43 pe cen did no dona e). We did no ind a link be ween gene al empa hy and likelihood o dona e, whe eas ou measu e o ou g oup empa hy was posi i ely linked o he likelihood o dona ion. The e was also a signi ican di e ence wi hin he p o encla e be ween hose delibe a ing in like-minded g oups (43 pe cen did no dona e) and hose delibe a ing in mixed g oups (61 pe cen did no dona e). This ind- ing seems o be in line wi h s udies in poli ical beha io (e.g., Mu z 2002, 2006; also Ka powi z e al. 2009) showing ha discussions wi h like- minded people os e willingness owa d collec i e ac ion, whe eas c oss- cu ing exposu e dec eases people’s willingness o ac . ACKNOWLEDGEMENTS The expe imen was unded by he Academy o Finland (Resea ch Council o Cul u e and Socie y, p ojec 251222) and A ˚bo Akademi Uni e si y Founda ion (Cen e o Excellence Funding). The au ho s a e g a e ul o he h ee anonymous e e ees as well as he Edi o o Scandina ian Poli ical S udies, Ca l Hen ik Knu sen, o aluable commen s. Pa icipan s a se e al con e ences, mos no ably in he Panel ‘Me hods and Measu es o Assessing Demo- c a ic Inno a ions’ a he ECPR Gene al Con e ence in Glasgow in 2014, p o ided commen s on ea lie e sions o his a icle. NOTES 1. Fo esul s ega ding opinion changes in he expe imen , see G € onlund e al. (2015). 2. We a e awa e o he c i ique ela ed o measu ing empa hy h ough sel - epo scales. In pa icula , such scales a e likely o inc ease he likelihood o socially desi able answe s. In he p esen s udy, Da is’s measu e was used because i i ed well wi h he gene al su ey wi h many i ems. Due o he la ge numbe o su ey i ems, we could no use all he ques ions om Da is’s measu e, which includes ou scales – namely a ‘ an asy scale’, a ‘pe spec i e- aking scale’, an ‘empha ic conce n scale’ and a ‘pe sonal dis ess scale’. Ins ead, we used hose ha in ou iew bes ep esen he kind o empa hy and pe spec i e- aking ele an in democ a ic delibe a ion. I is also no ewo hy ha because o andom alloca ion, social desi abili y should no a y be ween he expe imen al ea men s. 3. The ques ions we e i s pilo es ed wi h s uden s a wo uni e si ies in o de o mea- su e he app op ia eness o he ques ions o he pu pose o he expe imen . All su ey i ems wo ked well bo h in he pilo s and in he ac ual su ey conduc ed among he an- dom sample (T1). In he su eyed sample, all 14 i ems loaded on a single ac o and C onbach’s Alpha o he sum a iable eached 0.94. The e o e, we we e able o con- s uc a sum a iable o he 14 i ems. Each i em was i s ecoded in o a scale om 0 o 1, wi h 1 indica ing he mos immig a ion- iendly a i ude. 4. A e he expe imen , each pa icipan in delibe a ions ecei ed a gi ce i ica e wo h e90, whe eas e e yone in he con ol g oup ecei ed e15. 5. Fo a mo e de ailed accoun o ec ui men and a i ion, see Rapeli and Ka jalainen (2015). 6. T3 was only comple ed by hose who ook pa in he delibe a ion e en . V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 476 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 7. Acco ding o he webpages o he Finnish Red C oss (h ps://www. edc oss. i/), domes- ic wo k is abou counselling in heal h- ela ed issues, p o ision o i s aid in la ge public e en s and p o ision o olun a y ‘ iends’ o lonely people, such as o he elde ly. The mos isible pa o he Ca as ophe und is o aid people in ca as ophes all o e he wo ld, bu ac ually he und also inances domes ic wo k, like helping ic ims o i es. We belie e, howe e , ha he gene al public unde s anding in Finland would be ha he Ca as ophe und is o o eign wo k and he domes ic und o he wo k in Finland. 8. The da a analyzed in he in he p esen a icle, including ques ionnai es, a e deposi ed a he Finnish Social Science Da a A chi e (G € onlund 2014). They can be o de ed ee o cha ge o esea ch and eaching pu poses. 9. C onbach’s Alpha o hese h ee i ems is 0.66 a T2 and 0.69 a T4. 10. C onbach’s Alpha o hese h ee imes is 0.67 a T2 and 0.64 a T4. 11. E en hough we do no ha e a hypo hesis on gene al empa hy, we also looked a he de elopmen o gene al empa hy be o e and a e aking pa in delibe a ion. The esul s a e shown in Appendix Table 1. Gene al empa hy emained s able among hose who ook pa in he delibe a i e expe imen . The e we e no changes in any compa i- sons among hose who delibe a ed o in he con ol g oup. 12. In a sepa a e es , we disco e ed ha he inc ease in ou g oup empa hy occu ed among male pa icipan s, especially in he con encla e. Al hough his inc ease among con male pa icipan s was mo e e iden in he mixed ea men (p 50.001), i also occu ed in he like-minded ea men (p 50.03). Among women, no s a is ically signi - ican changes in ou g oup empa hy can be aced in any o he g oups. 13. K uskall-Wallis H- es e eals a signi ican di e ence in dona ions be ween he ou g oups depic ed in Figu e 3 (p 50.010). 14. When conside ing hei dona ions o he Red C oss, pa icipan s we e engaged in a so- called ‘dic a o game’, which is a echnique equen ly used o es subjec s’ inclina ion o p osocial ac ion. In he s anda d o m o he game, he dic a o and he ecipien a e s uden subjec s, whe eas in ou expe imen he ecipien is a cha i y o ganiza ion. P e- ious esul s on such games show ha subjec s end o sha e a la ge pa o hei endowmen when he ecipien is conside ed as dese ing o needy, compa ed o he s anda d o m o he game (Engel 2011). A compa able s udy o ou own was con- duc ed by Eckel and G ossman (1996), who ound ha abou 73 pe cen o expe imen- al subjec s made an alloca ion o he Ame ican Red C oss. O e all, pa icipan s in ou expe imen dona ed conside ably less han he pa icipan s in Eckel and G ossman’s s udy. This migh be due o ou pa icipan s’ expe ience o ha ing ea ned he ewa d om aking pa in he expe imen , which in ol ed many s ages and ook place o qui e a long ime. This in e p e a ion is in line wi h he esul s o Engel’s (2011) me a- s udy o e dic a o games. 15. Acco ding o he Mann-Whi ney U- es , he di e ence be ween he wo ea men s wi hin he p o encla e is no s a is ically signi ican (p 50.137, exac one- ailed p50.072). 16. Acco ding o he Mann-Whi ney U- es , he di e ence be ween he wo ea men s wi hin he con encla e is no s a is ically signi ican (p 50.272, exac one- ailed p 50.155). Appendix The main i ems measu ing a i udes on immig a ion a e: 1. Finland should ake mo e immig an s. Do you hink his is a bad o a good sugges ion? 2. Mig a ion o o eigne s in o Finland should be es ic ed as long as he e is unemploymen in Finland. [ ] V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 477 3. Do you hink Finland will change in o a be e o a wo se place o li e when people om o he coun ies mo e o Finland? Ques ions 1–3 we e p esen ed on a scale om 0 o 10. 4. I is good o he Finnish economy ha people om o he coun ies mo e o Finland. 5. Immig an s ake away jobs om na i e Finns. [ ] 6. Immig an s should ha e he same igh o social secu i y as Finns e en i hey a e no Finnish ci izens. 7. The s a e and he municipali ies use oo much money o aid immi- g an s. [ ] 8. Immig a ion poses a se ious h ea o ou na ional o iginali y. 9. E e yone who wan s o come o Finland o li e and wo k should be allowed o do so. 10. Immig a ion policy should p ima ily a o Ch is ians ins ead o o he eligions. [ ] 11. Gene ally speaking, immig an s adap well in o he Finnish socie y. 12. I would be happy o ha e an immig an as a co-wo ke . 13. I would accep an immig an as a amily membe . 14. I would accep immig an s in my neighbo hood. Ques ions 3–14 we e p esen ed as a s anda d Like scale wi h ou alues. [ ] 5Re e sed coding in he sum a iable. Appendix Table 1. The De elopmen o Gene al Empa hy P e (T2) Pos (T4) T4–T2 p Con All 0.72 0.72 0.00 Like-minded 0.72 0.73 0.01 Mixed 0.73 0.71 20.02 P o All 0.75 0.75 0.00 Like-minded 0.74 0.75 0.01 Mixed 0.77 0.76 20.01 Con ol g oup Con 0.69 0.68 20.01 P o 0.75 0.74 20.01 No e:*p<0.05; **p <0.01; ***p <0.001. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 478 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 REFERENCES Ba son, D. C. 1997. ‘Sel -O he Me ging and he Empa hy-Al uism Hypo hesis: Reply o Neu- be g e al. 1997’, Jou nal o Pe sonali y and Social Psychology, 73(3), 517–22. Ba abas, J. 2004. ‘How Delibe a ion A ec s Policy Opinions’, Ame ican Poli ical Science Re iew, 98(4), 687–701. Ba son, D. C. 2011. Al uism in Humans. New Yo k: Ox o d Uni e si y P ess. Ba son, D. C. e al. 1997. ‘Is Empa hy-induced Helping Due o Sel -O he Me ging?’, Jou nal o Pe sonali y and Social Psychology, 73(3), 495–509. Ba son, D. C. e al. 2002. ‘Empa hy, A i udes and Ac ion: Can Feeling o a Membe o a S igma ized G oup Mo i a e One o Help he G oup?’, Pe sonali y and Social Psychology Bulle in, 28(12), 1656–66. Binmo e, K. 1998. Game Theo y and he Social Con ac , Vol. 2: Jus Playing. Camb idge, MA: MIT P ess. B uneau, E. G., Du ou , N. & Saxe, R. 2012. Social Cogni ion in Membe s o Con lic G oups: Beha iou al and Neu al Responses in A abs, Is aelis and Sou h Ame icans o Each O he ’s Mis o unes. Philosophical T ansac ions o he Royal Socie y B, 367, 717–30. Cika a, M. e al. 2014. ‘Thei Pain Gi es Us Pleasu e: How In e g oup Dynamics Shape Empa hic Failu es and Coun e -empa hic Responses’, Jou nal o Expe imen al Social Psy- chology, 55, 110–25. Da is, M. H. 1980. ‘A Mul idimensional App oach o Indi idual Di e ences in Empa hy’, JSAS Ca alog o Selec ed Documen s in Psychology, 10, 85. Eckel, C. C. & G ossman, P. J. 1996. ‘Al uism in Anonymous Dic a o Games’, Games and Economic Beha io , 16, 181–91. Engel, C. 2011. ‘Dic a o Games: A Me a S udy’, Expe imen al Economics, 14, 583–610. Fishkin, J. S. & Luskin, R. 2005. ‘Expe imen ing wi h a Democ a ic Ideal: Delibe a i e Polling and Public Opinion’, Ac a Poli ica, 40, 284–98. Fong, C. M. 2007. ‘E idence om an Expe imen on Cha i y o Wel a e Recipien s: Recip oci y, Al uism and he Empa hic Responsi eness Hypo hesis’, Economic Jou nal,117,1008–24. Glynn, A. N. & Sen, M. 2015. ‘Iden i ying Judicial Empa hy: Does Ha ing Daugh e s Cause Judges o Rule o Women’s Issues?’, Ame ican Jou nal o Poli ical Science, 59(1), 37–54. Goodin, R. E. 2000. ‘Democ a ic Delibe a ion Wi hin’, Philosophy and Public A ai s, 29(1), 81–109. G € onlund, K. 2014. Ci izen Delibe a ion on Immig a ion: Su ey Da a 2012 [da ase ]. Ve sion 1.0 (2014-11-05). Finnish Social Science Da a A chi e [dis ibu o ]. A ailable online a : h p://u n. i/u n:nbn: i: sd:T-FSD2958 G € onlund, K., He ne, K. & Se € al€ a, M. 2015. ‘Does Encla e Delibe a ion Pola ize Opinions?’, Poli ical Beha io , 37(4), 995–1020. G € onlund, K., Se € al€ a, M. & He ne, K. 2010. ‘Delibe a ion and Ci ic Vi ue: Lessons om a Ci izen Delibe a ion Expe imen ’, Eu opean Poli ical Science Re iew, 2(1), 95–117. Habe mas, J. 1996. Be ween Fac s and No ms. Camb idge, MA: MIT P ess. Himmel oos, S., Rapeli, L. & G € onlund, K. 2017. ‘Talking wi h Like-minded People: Equali y and E icacy in Encla e Delibe a ion’, Social Science Jou nal, 54(2), 148–58. Ho man, M. L. 2000. Empa hy and Mo al De elopmen : Implica ions o Ca ing and Jus ice. Camb idge: Camb idge Uni e si y P ess. Ka powi z, C. F., Raphael, C. & Hammond IV, A. S. 2009. ‘Delibe a i e Democ acy and Inequali y: Two Chee s o Encla e Delibe a ion among he Disempowe ed’, Poli ics and Socie y, 37(4), 576–615. K ause, S. E. 2008. Ci il Passions: Mo al Sen imen and Democ a ic Delibe a ion. P ince on, NJ: P ince on Uni e si y P ess. Luskin, R. C., Fishkin, J. S. & Jowell, R. 2002. ‘Conside ed Opinions: Delibe a i e Polling in B i ain’, Poli ics and B i ish Jou nal o Poli ical Science, 32(3), 455–87. Mansb idge, J. 1994. ‘Using Powe /Figh ing Powe ’, Cons ella ions, 1(1), 53–73. Mead, G. H. 1934. Mind, Sel and Socie y. Ed. C. M. Mo is. Chicago, IL: Uni e si y o Chicago P ess. Me cie , H. & Landemo e, H. 2012. ‘Reasoning is o A guing: Unde s anding he Successes and Failu es o Delibe a ion’, Poli ical Psychology, 33(2), 243–58. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017 479 Mon ada, L. & Schneide , A. 1989. ‘Jus ice and Emo ional Reac ions o he Disad an aged’, Social Jus ice Resea ch, 3(4), 313–44. Mo ell, M. E. 2010. Empa hy and Democ acy. Uni e si y Pa k, PA: Pennsyl ania S a e Uni- e si y P ess. Mu z, D. C. 2002. ‘C oss-cu ing Social Ne wo ks: Tes ing Democ a ic Theo y in P ac ice’, Ame ican Poli ical Science Re iew, 96, 111–26. Mu z, D. C. 2006. Hea ing he O he Side: Delibe a i e Ve sus Pa icipa o y Democ acy.Cam- b idge: Camb idge Uni e si y P ess. Newman, B. J. e al. 2015. ‘Easing he Hea y Hand: Humani a ian Conce n, Empa hy and Opinion on Immig a ion’, B i ish Jou nal o Poli ical Science, 45(3), 583–607. Pagano, S. J. & Huo, Y. J. 2007. The Role o Mo al Emo ions in P edic ing Suppo o Poli i- cal Ac ions in Pos -wa I aq’, Poli ical Psychology, 28(2), 227–55. P es on, S. D. & De Waal, F. B. M. 2002. ‘Empa hy: I s Ul ima e and P oxima e Bases’, Beha io al and B ain Sciences, 25, 1–72. Rapeli, L. & Ka jalainen, M. 2015. ‘Who Will No delibe a e? A i ion in a Mul i-s age Ci i- zen Delibe a ion Expe imen ’, Quali y and Quan i y, 49(1), 407–22 Rawls, J. 1971. A Theo y o Jus ice. Ox o d: Ox o d Uni e si y P ess. Se € al€ a, M., G € onlund, K. & He ne, K. 2010. ‘Ci izen Delibe a ion on Nuclea Powe : A Com- pa ison o Two Decision-making Me hods’, Poli ical S udies, 58, 688–714. S andbe g, K., Himmel oos, S. & G € onlund, K. 2017. ‘Do Discussions in Like-minded G oups Necessa ily Lead o Mo e Ex eme Opinions? Delibe a i e Democ acy and G oup Pola iza- ion’ In e na ional Poli ical Science Re iew, DOI: 10.1177/0192512117692136. Suns ein, C. 2002. ‘The Law o G oup Pola iza ion’, Jou nal o Poli ical Philosophy,10(2), 175–95. Suns ein, C. 2007. Republic.com 2.0. P ince on, NJ: P ince on Uni e si y P ess. Suns ein, C. 2009. Going o Ex emes. How Like Minds Uni e and Di ide. Ox o d: Ox o d Uni- e si y P ess. Thompson, D. F. 2008. ‘Delibe a i e Democ a ic Theo y and Empi ical Poli ical Science’, Annual Re iew o Poli ical Science, 11, 497–520. Wal e , H. 2012. ‘Social Cogni i e Neu oscience o Empa hy: Concep s, Ci cui s and Genes’, Emo ion Re iew, 4(1), 9–17. V C2017 The Au ho s Scandina ian Poli ical S udies published by John Wiley & Sons L d on behal o No dic Poli ical Science Associa ion 480 Scandina ian Poli ical S udies, Vol. 40 – No. 4, 2017