Validation of the Portuguese version of impulsive-premeditated aggression scale in an inmate population

Feb ua y 2018 | Volume 9 | A icle 101 O iginal esea ch published: 05 Feb ua y 2018 doi: 10.3389/ psy .2018.00010 F on ie s in Psychia y | www. on ie sin.o g Edi ed by: Ka a ina Howne , Ka olinska Ins i u e (KI), Sweden Re iewed by: Giuseppe Sa o i, Uni e si à degli S udi di Pado a, I aly E ich Flamme , Z P Suedwue embe g, Ge many *Co espondence: Jacin o Cos a Aze edo [email p o ec ed] Special y sec ion: This a icle was submi ed o Fo ensic Psychia y, a sec ion o he jou nal F on ie s in Psychia y Recei ed: 25Oc obe 2017 Accep ed: 15Janua y2018 Published: 05Feb ua y2018 Ci a ion: Aze edoJC, Pais-Ribei oJL, CoelhoR and Figuei edo-B agaM (2018) Valida ion o he Po uguese Ve sion o Impulsi e–P emedi a ed Agg ession Scale in an Inma e Popula ion. F on . Psychia y 9:10. doi: 10.3389/ psy .2018.00010 Valida ion o he Po uguese Ve sion o impulsi e–P emedi a ed agg ession scale in an inma e Popula ion Jacin o Cos a Aze edo1,2*, José Luís Pais-Ribei o3, Rui Coelho1 and Ma ga ida Figuei edo-B aga1,2 1 Depa men o Neu oscience and Men al Heal h, Facul y o Medicine, Uni e si y o Po o, Po o, Po ugal, 2 I3S – Ins i u o de In es igação e Ino ação em Saúde, Po o, Po ugal, 3 Facul y o Psychology and Educa ional Sciences, Uni e si y o Po o, Po o, Po ugal Agg ession is one o he co e symp oms o an isocial pe sonali y diso de (ASPD) wi h he apeu ic and p ognos ic ele ance. ASPD is highly p e alen among inma es, being esponsible o ad e se e en s and ele a ed di ec and indi ec economic cos s o he c iminal jus ice sys em. The Impulsi e/P emedi a ed Agg ession Scale (IPAS) is a sel - epo ins umen ha cha ac e izes agg ession as ei he p edomina ely impulsi e o p emedi a ed. This s udy aims o de e mine he alidi y and eliabili y o he IPAS in a sample o Po uguese inma es. A o al o 240 inma es we e included in he s udy. A p incipal componen ac o analysis was pe o med so as o ob ain he cons uc alid- i y o he IPAS impulsi e agg ession (IA) and p emedi a ed agg ession (PM) subscales; in e nal consis ency was de e mined by C onbach’s alpha coe icien ; con e gen and di e gen alidi y o he subscales we e de e mined analyzing co ela ions wi h he Ba a Impulsi eness scale, 11 h e sion (BIS-11), and he Psychopa hic Checklis Re ised (PCL-R). The o a ed ma ix wi h wo ac o s accoun ed o 49.9% o o al a iance. IA subscale had 11 i ems and PM subscale had 10 i ems. The IA and PM subscales had a good C onbach’s alpha alues o 0.89 and 0.88, espec i ely. The IA subscale is co ela ed wi h BIS-11 a en ional, mo o , and non-planning impulsi eness dimensions (p<0.05). The PM subscale is co ela ed wi h BIS-11 a en ional, mo o impulsi eness dimensions (p<0.05). The PM subscale is co ela ed wi h PCL-R in e pe sonal, li es yle, and an isocial dimensions (p<0.05). The IA subscale is no co ela ed wi h PCL-R. The Po uguese ansla ed e sion o IPAS has adequa e psychome ic p ope ies, allowing he measu emen o impulsi e and p emedi a ed dimensions o agg ession. Keywo ds: agg ession, an isocial pe sonali y diso de , psychopa hy, impulsi i y, impulsi e–p emedi a ed agg ession scale, Ba a impulsi eness scale, 11 h e sion, Psychopa hic checklis e ised inT ODUcTiOn Agg ession ep esen s a public heal h issue. Ha ing a nega i e impac on indi iduals and socie y (1), i can be de ined as a “beha io al display in which physical o ce is used wi h he in en o ha m o damage ano he indi idual o objec ” (2). 2 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 A clinical de ini ion o agg essi e beha io should conside biological, en i onmen al, cul u al, and social a iables, and hei in e play o ac as p edisposi ion o elici ing ac o s (3). The ange o beha io s ha can be classi ied as agg essi e can a y om e bal agg ession o homicide, and i is impo an o cha ac e ize he le el o planning, he possible unde s anding o hypo he ical consequences, he p esence o us a ions, insul s, in e pe sonal a ack, h ea s, en i onmen al s esso s, and associ- a ed psychopa hologies. Empi ical li e a u e has p oposed a dicho omous agg ession classi ica ion, i.e., impulsi e agg ession (IA; eac i e, a ec i e, o non-planned), and p emedi a ed agg ession (PM; p oac i e, ins umen al, p eda o y, o con olled) (4). The exp ession “IA” e e s o uncon olled agg essi e ou bu s s ha a e ou o p opo ion o he p o oking e en , while “PM” desc ibes agg essi e beha io s ha a e planned, con olled, and/o goal- o ien ed (5). Acco ding o he agg ession ype, indi iduals may di e in social adjus men , c iminal beha io , emo ional unc ion, cogni i e pe o mance, au onomic esponse, and ea men ou come (6, 7). The IA ends o occu in he p esence o igge ing s imulus ha is in e p e ed as h ea o p o oca ion (8). This ype o agg ession in ol es a ec i e a ousal, leading o a apid and uncon olled beha io al esponse. I has been co ela ed wi h in o ma ion-p ocessing/neu ocogni i e impai men s, abusi e home backg ounds, an ang y/impulsi e/anxious pe sonali y, and high psychological s ess eac i i y (9). The PM, by con as , ends o be a planned beha io ha has a speci ic expec ed goal (8), and i can be explained as a lea ned beha io (10). I has been co ela ed wi h poo pa en al con ol, lack o a ec , psychopa hic pe sonali y, and low physiological a ousal (9). The classi ica ion o an agg essi e ac o a pa e n o agg es- si e beha io allows ea men selec ion and iolence isk managemen . The IA has a good esponse o pha macologic ea men when mood s abilize s and/o an ipsycho ics a e p esc ibed (11). In he case o PM, on he con a y, he esponse o pha macological ea men is gene ally insu icien (12), and o ensic/beha io al s a egies a e ecommended (7). Fu he mo e, indi iduals wi h PM ha e a highe isk o iolen c iminal ecidi ism han hose wi h IA (13). The e is a link be ween agg essi e beha io and psychia ic diso de s (14). Indi iduals wi h dep ession, anxie y, psychosis, subs ance use diso de s, hype ac i e a en ion de ici diso de , and pe sonali y diso de s p esen a highe equency o agg essi e ac s (15). Impulsi e agg ession has been linked o psychia ic diseases, such as anxie y, dep ession, pe sonali y diso de s, and subs ance use diso de s (16). PM has been associa ed wi h psychopa hic pe sonali y ai s (17). In pa icula , an isocial pe sonali y diso de (ASPD) is cha - ac e ized by signi ican i i abili y, agi a ion, impulsi eness, and hos ili y, and agg ession ep esen s one o i s co e symp oms (18). ASPD is highly p e alen among inma es. I is, hus, esponsible o ad e se e en s and ele a ed di ec and indi ec economic cos s in he c iminal jus ice sys em (19). An isocial pe sonali y diso de is he only acknowledged psychia ic diso de ha con e s an inc eased isk o bo h IA and PM (20–22), wi h se e al s udies epo ing psychopa hy and psychopa hic ai s as isk ac o s o PM (23). Psychopa hic Checklis Re ised (PCL-R) is he gold s and- a d o psychopa hy assessmen and diagnosis, and sepa a es psychopa hic ai s in o ou dimensions: (1) in e pe sonal, (2) a ec i e, (3) li es yle, and (4) an isocial (24). Impulsi i y has been ela ed o agg ession and is a symp om o ASPD (25). Impulsi i y can be a necessa y ac o o agg ession de elopmen (26). I can be de ined “as a p edisposi ion owa d apid, unplanned eac ions o in e nal o ex e nal s imuli wi hou ega d o he nega i e consequences o hese eac ions o he impulsi e indi iduals o o o he s” (27). Impulsi i y can be measu ed wi h Ba a Impulsi eness Scale, 11 h e sion (BIS-11), a sel - epo psychome ic ins umen ha sepa a e impulsi i y in o h ee componen s: (1) ac ing on he spu o he momen (mo o ac i a ion), (2) no ocusing on he ask a hand (a en ion), and (3) no planning and hinking ca e ully (lack o planning) (28). Indi iduals wi h sco es highe han 72 a e conside ed highly impulsi e (29). Pa ien s wi h subs ance use diso de s, dep ession, bipola diso de , a en ion de ici /hype ac i i y diso de (AD/HD), as well as suicide a emp e s and c iminal o ende s end o ha e highe BIS-11 o al sco es (29). Despi e he e idence suppo ing he no ion ha pe sons who display an impulsi e agg essi e beha io a e dis inc om hei coun e pa s who p esen a p emedi a ed beha io , he clini- cal use ulness and gene aliza ion is ques ioned because mos s udies included iolen inca ce a ed o ende s, o psychia ic inpa ien s. Wi hin he inca ce a ed popula ion, howe e , agg ession is a ele an daily clinical conce n. The e o e, he s udy o agg es- sion in o ensic psychia ic se ings should be encou aged in o de o ind and p omo e he bes clinical p ac ices (30, 31). I is impo an o use alida ed ins umen s in he e alua ion o agg essi e ac s in p isone s in o de o imp o e medical in e - en ions (32). In Po ugal, we do no ha e a alida ed agg ession ca ego iza ion ins umen . The e a e wo psychome ic ins umen s ha allow agg es- sion ca ego iza ion: he impulsi e–p emedi a ed agg ession scale (IPAS) and he eac i e–p oac i e agg ession ques ionnai e (2, 33, 34). In he o iginal alida ion s udy o IPAS, he sample comp ised 93 physically agg essi e men ec ui ed in he communi y wi h a mean age o 35.9yea s, whose p incipal agg ession ype was physical assaul (n=80). The agg ession was measu ed h ough he Li e ime His o y o Agg ession, Buss Pe y Agg ession Ques ionnai e, and he Agg ession In e iew. The p incipal componen analysis iden i ied wo ac o s (IA and PM), which accoun ed o 16.56 and 14.03% o o al a iance, espec i ely. The in e nal consis ency measu ed by C onbach’s alpha was 0.77 o he IA ac o and 0.82 o he PM ac o . Sensi i i y o he IA scale was 0.96 and speci ici y was 0.50. Fo he PM scale, sensi i - i y and speci ici y we e 0.60 and 0.96, espec i ely (2). All o he alida ion s udies o IPAS epo ed iden ical esul s in p incipal componen analysis, wi h wo ac o s IA and PM. TaBle 1 | Valida ion s udies o Impulsi e/P emedi a ed Agg ession Scale (IPAS). e e ence sample Mean agea P incipal agg ession agg ession measu es in e nal consis- encyb To al ex plained a iancec nType O igin iadPMeia PM S an o d e al. (2) 93 men Physically agg essi e men Communi y 35.9 86% physical assaul LHAQ , BPAQg, AIh 0.77 0.82 16.56 14.03 Kockle e al. (35) 86 men P isone s Inpa ien Fo ensic hospi al 38 83% con ic ed o iolen c imesi, 37% non-guil y by eason o insani y nsj0.81 0.72 20 13 Conne e al. (36) 61 women, 60 men Pa ien s in ea men o opia e dependence Ou pa ien Psychia ic Se ice 41.9 ns LHAQ, BPAQ 0.74 0.75 ns Ma hias e al. (37) 24 gi ls, 42 boys Adolescen s wi h conduc diso de Communi y 14.5 ns LHAQ, BPAQ 0.82 0.78 18.5 15.6 S an o d e al. (38) 113 men Men con ic ed o domes ic iolence Communi y 36 Domes ic iolence LHAQ 0.75 0.86 ns Haden e al. (39) 213 women, 127 men S uden s Communi y 19.06 ns BPAQ 0.77 0.81 24 12 Kuyck e al. (40) 149 men, 70 women P isone s P ison 50% be ween 25 and 39 75% con ic ed o iolen c imes ns 0.93 0.90 ns Chen e al. (41) 389 women, 262 men S uden s Communi y 26.1 ns BPAQ 0.70 0.66 12.9 11.9 Romans e al. (42) 114 women, 49 men Psychia ic pa ien s Ou pa ien Psychia ic Se ice 25.8 ns OASk0.85 0.76 23.3 10.1 aYea s. bC onbach’s alpha. c%. dImpulsi e agg ession ac o . eP emedi a ed Agg ession ac o . Li e ime His o y o Agg ession. gBuss Pe y Agg ession Ques ionnai e. hAgg ession In e iew. iAgg a a ed assaul , ba e y on a law o ice , sexual assaul , mu de , a med obbe y, a emp ed mu de , and manslaugh e . jNo speci ied. kO e Agg ession Scale. 3 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 In e nal consis ency coe icien s a ied be ween 0.70 and 0.93 o IA and be ween 0.66 and 0.90 o PM (Table1). We in end o alida e he Po uguese e sion o he IPAS in inma es and o de e mine i s psychome ic p ope ies. Thus, we expec o eplica e he pu po ed wo- ac o model o IPAS; we p edic ha he subscales o he IPAS and he BIS-11 will be posi i ely co ela ed; we p edic ha he PM subscales o he IPAS—unlike he IA ones—will be co ela ed wi h he PCL-R. This s udy is a pa o a la ge s udy aiming o cha ac e ize biological and psychosocial ac o s ha p edic he agg ession ype in young adul male inma es. MaTe ials anD MeThODs Ou sample was collec ed a wo peni en ia y ins i u ions in he No h o Po ugal. The esea ch p o ocol was o mally app o ed by he E hics Commi ee o Cen o Hospi ala de São João, Po o, Po ugal (Documen numbe 48.14), and by he hos ing ins i u ion, he Gene al Di ec ion o P oba ion and P ison Se ices. Pa icipa ion was olun a y and he e was no ewa d o i . The pa icipan s we e all Po uguese. W i en in o med consen was ob ained a e he p ocedu es we e explained o he pa icipan s, in acco dance wi h he Decla a ion o Helsinki. Pa icipan s we e included i hey we e o e 18yea s old, had a pe sonal his o y o agg ession in he pas 6mon hs, had been e e ed o he clinical se ices o agg essions owa d o he inma es, we e able o ead, and i hey we e capable o p o iding hei w i en in o med consen . Socio-demog aphic cha ac e is ics and o ensic his o y we e collec ed om in e iews and clinical eco ds. We applied h ee psychome - ic ins umen s: he IPAS, he BIS-11, and he PCL-R. This sample does no in end o ep esen all he Po uguese inma e popula ion, as he me hods in ol e a non-p obabilis ic sample app oach. T anscul u al adap a ion and Valida ion In o de o add ess he anscul u al adap a ion and alida ion o IPAS, an ini ial ansla ion was ca ied ou by a g oup o Po uguese expe s. The back- ansla ion o his i s ansla ed e sion was hen pe o med by a na i e English-speaking expe . The i s g oup analyzed he seman ic and idioma ic co espond- ence o he ansla ion and he back- ansla ion, and pe o med a inal syn hesis ansla ion (43). 4 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 Psychome ic ins umen s Impulsi e–P emedi a ed Agg ession Scale The IPAS is a 30-i em sel - epo ques ionnai e used o classi y agg essi e ac s occu ing o e he p e ious 6mon hs. I ems a e sco ed wi hin a i e-poin Like scale (1=S ongly Disag ee o 5=S ongly Ag ee). The scale di e en ia es wo ac o s: IA and PM, wi h he weigh ed sco e allowing o he ca ego iza ion o he ype o agg ession. The indi idual’s le el o IA and PM was ob ained h ough he sum o 20 o he 30 i ems o he IPAS: he IA i ems (eigh i ems: 3, 5, 7, 8, 9, 21, 24, and 26) and he PM i ems (12 i ems: 1, 2, 6, 10, 11, 12, 14, 16, 17, 20, 29, and 30) (2). Ba a Impulsi eness Scale-11 To de e mine he con e gen alidi y o he IA and he PA subscales o he IPAS, we ha e u ilized BIS-11. The BIS-11 is a sel - epo ques ionnai e o assessing gene al impulsi eness. The cu en scale e sion con ains 30 i ems ha a e coded om 1 ( a ely/ne e ) o 4 (almos always/always). The le el o impul- si eness is calcula ed by summing up he sco es o each i em. All i ems we e de ined as iden i ying impulsi eness wi hin he s uc u e o ela ed pe sonali y ai s. The ac o analysis e ealed h ee componen s as ollows: “a en ional impulsi eness,” “mo o impulsi eness,” and “non-planning impulsi eness.” Ins ead o a unique o al sco e, he s udy o he indi idual con ibu ion o each componen is ecommended. Findings epo an accep able/high in e nal consis ency (C onbach’s alpha 0.79–0.82) o he scale when applied o o ensic samples (28). The s uc u al p ope ies o he BIS-11 we e eplica ed in Po uguese speaking samples (44). Psychopa hy Checklis Re ised To de e mine he di e gen alidi y o he IA and he PA sub- scales o he IPAS, we ha e applied he PCL-R. The PCL-R is he gold s anda d measu e o psychopa hy, ga he ing in o ma ion om eco ds and a semi-s uc u ed in e iew (24, 25). The 20 i ems a e sco ed as absen (0), p esen o some deg ee (1), o ully p esen (2), ha ing a maximum o al sco e o 40 poin s. This ins umen may be conside ed a ou - ac o model comp is- ing in e pe sonal, a ec i e, li es yle, and an isocial dimensions (45). The in e pe sonal and a ec i e dimensions join ly se e as second-o de ac o s ep esen ing he co e ai s o he psycho- pa hic pe sonali y. Li es yle and an isocial ace s o m a supe - o dina e ac o o social de iance (46). The wo second-o de ac o s (Fac o 1 and Fac o 2) a e concomi an wi h he o iginal ac o s uc u e epo ed o he i s edi ion o he PCL-R (47). The s uc u al p ope ies o he PCL-R we e eplica ed in Po uguese samples o he s anda d p o ocol, including eco d e iew and s uc u ed in e iew (48). The PCL-R was applied by an expe ienced psychia is (Jacin o Cos a Aze edo). s a is ical analyses S a is ical analyses we e conduc ed ollowing he me hodology used in he p e ious IPAS alida ion s udies in adul s. The p in- cipal componen analysis was conduc ed wi h no assump ions ega ding he numbe o po en ial ac o s. Lau enschlage ’s (49) ables we e used o de e mine he h eshold o signi ican ac o s (minimum eigen alue o 2.0) and i em ac o load- ings (minimum eigen alue o 0.40). In o de o assess ac o s uc u e and maximize he eliabili y o ac o in e p e a ion, an Oblimin o a ion was used so as o ob ain an oblique ac- o solu ion o he o iginal ac o s (50). In e nal consis ency was measu ed using C onbach’s alpha (51). Con e gen and di e gen alidi y o he IPAS was es ed by examining he Pea son’s p oduc -momen co ela ions wi h he s anda d- ized measu es o impulsi i y and pe sonali y (BIS-11 and PCL-R). Con i ma o y ac o ial analysis was es ed by calcula - ing goodness-o - i indice, obus me hod (52). The analyses we e ca ied ou using IBM SPSS S a is ics o Windows, Ve sion 22.0 (A monk, NY, USA: IBM Co p.), and wi h he S uc u al Equa ion P og am, EQS e sion 6.1 (53). esUlTs sample Desc ip ion The sample comp ised 240 inma es wi h a mean (±SD) age o 35.4±8.4yea s; a o al o 67.1% (n=161) o he pa icipan s we e single, and 58.3% did no ha e any child en (n=140). The mean educa ion le el o he sample was 6.8±3.2yea s. The mean ime in p ison ( ime spen behind ba s a he ime o assessmen ) o he sample was 109.3±70.6mon hs. A o al o 45.8% (n=110) o he inma es had been con ic ed o iolen c imes (physical assaul , mu de , a emp ed mu de ). Valida ion p ocedu es we e pe o med in he o al sample (n=240). P incipal componen analysis We ha e pe o med a p incipal componen analysis explo ing he o iginal 30 i ems. Fi s , we inspec ed he Kaise –Meye –Olkin Measu e o Sampling Adequacy (KMO=0.87) and eached he conclusion ha i was an app op ia e alue. Second, we achie ed a s a is ically signi ican Ba le ’s Tes o Sphe ici y (p<0.01). Thi d, we ex ac ed wo ac o s (wi h an oblimin o a ion) wi h eigen alues o 7.19 (Fac o 1, IA) and 3.29 (Fac o 2, PM) accoun ing o 34.27 and 15.65% o o al a iance, espec i ely. The IA ac o was composed o en i ems: 30, 27, 22, 9, 24, 15, 26, 4, 7, and 13. The PM ac o was composed o ele en i ems: 6, 14, 29, 28, 2, 23, 12, 16, 20, 10, and 1. The i ems 3, 5, 8, 11, 17, 18, 19, 21, and 25 we e excluded om he analysis as hey exhibi ed componen loadings in e io o 0.40 (Table2). con i ma o y Fac o ial analysis To es he hypo hesis o wo dimensions, we use he con i ma o y ac o analysis, o es he goodness-o - i indice, obus me hod. Resul s showed a compa a i e i index o 0.90, and a oo mean squa e e o o app oxima ion (RMSEA) o 0.06, 90% con idence in e al (0.05, 0.08). in e nal consis ency The in e nal consis ency es o he dimensions ob ained in he ac o analysis wi h he p esen sample e ealed a good C onbach’s alpha alue o he IA ac o (0.89), and a good C onbach’s alpha alue o he PM ac o (0.88) (Table2). Reliabili y was also es ed ega ding he o iginal e sion o he IPAS subscales. C onbach’s alpha o he IA o iginal subscale was accep able (0.78), and i was good o he PM subscale (0.85). TaBle 3 | Pea son co ela ions coe icien s be ween Impulsi e/P emedi a ed Agg ession Scale (IPAS), Ba a Impulsi eness Scale, 11 h e sion (BIS-11), and Psychopa hic Checklis Re ised (PCL-R). iPas subscales Measu es subscales ia PM BIS-11 To al sco e 0.21** 0.20** A en ional 0.14* 0.18* Mo o 0.26** 0.27** Non-planning 0.15* 0.06 PCL-R To al sco e 0.01 0.29** In e pe sonal −0.14 0.24* A ec i e −0.14 0.16* Li es yle −0.07 0.22* An isocial −0.07 0.21* The s udy used 240 pa icipan s o co ela ions be ween IPAS and BIS-11; and a subsample o 134 pa icipan s o co ela ions be ween IPAS and PCL-R. IA, impulsi e agg ession; PM, p emedi a ed agg ession. *p<0.05; **p<0.01. TaBle 2 | Fac o loadings o Impulsi e/P emedi a ed Agg ession Scale (IPAS). iPas i ems impulsi e P emedi a ed 30. Any hing could ha e se me o p io o he inciden s 0.63 – 27. I was in a bad mood he day o he inciden 0.66 – 22. I was con used du ing he ac s 0.79 – 9. I eel I los con ol o my empe du ing he ac s 0.80 – 24. My beha io was oo ex eme o he le el o p o oca ion 0.72 – 15. I became agi a ed o emo ionally upse p io o he ac s 0.72 – 26. I conside he ac s o ha e been impulsi e 0.66 – 4. I ypically el guil y a e he agg essi e ac s 0.73 – 7. I usually can’ ecall he de ails o he inciden s well 0.50 – 13. I eel some o he inciden s wen oo a 0.72 – 6. I eel my ac ions we e necessa y o ge wha I wan ed – 0.58 14. I hink he o he pe son dese ed wha happened o hem du ing some o he inciden s – 0.65 29. I am glad some o he inciden s occu ed – 0.64 28. The ac s we e a “ elease” and I el be e a e wa d – 0.75 2. I el my ou bu s s we e jus i ied – 0.70 23. P io o he inciden s I knew an al e ca ion was going o occu – 0.55 12. I wan ed some o he inciden s o occu – 0.80 16. The ac s led o powe o e o he s o imp o ed social s a us o me – 0.72 20. Some o he ac s we e a emp s a e enge – 0.75 10. Some imes I pu posely delayed he ac s un il a la e ime – 0.63 1. I planned when and whe e my ange was exp essed – 0.64 Eigen alues 7.19 3.29 Va iance (%) 34.27 15.65 C onbach’s alpha 0.89 0.88 5 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 cap u e impo an aspec s o he cons uc , as well as i s ele ance o he p oposed use. The p esen esul s show app op ia e alidi y o he Po uguese e sion o he IPAS. The i ems o he ins umen we e adap ed acco ding o he o iginal e sion (2). In ou sample, IPAS was able o demons a e he p esence o wo ac o s: impulsi e and PM. All he au ho s who s udied he alidi y o IPAS we e able o educe he a iabili y o he scale o wo ac o s: IA and PM ac o s. This is in line wi h he li e a u e on human agg ession which a gues ha agg ession is no a single cons uc and should be unde s ood as being composed o wo dimensions o ca ego- ized in o wo di e en ypes: impulsi e o PM. In he analysis o he explained a iance desc ibed by o he au ho s, we ha e obse ed ha he IA ac o can explain be ween 12.9% (41) and 24% (39) o he o al IPAS a iance; and ha he PM ac o can accoun o be ween 10.3% (42) and 15.6% (37) o he o al IPAS a iance. We ha e ob ained highe alues o explained a iance (34.3% o he IA ac o and 15.7% o he PM ac o ). In he analysis o he IPAS in e nal consis ency desc ibed by o he au ho s, we ha e obse ed ha i a ies be ween 0.70 (41) and 0.93 (40) o he IA ac o ; and be ween 0.66 (41) and 0.90 (40) o he PM ac o . Ou alues o in e nal consis ency and a iance can be explained by he homogenei y o he sample. Acco ding o Ben le and Douglas (52), when he goodness- o - i and adjus ed goodness-o - i indexes a e highe han 0.90, he analyses indica e adequa e i o he models. Also, acco ding o Ben le and Douglas, when he RMSEA is less han. 0.10, he analysis indica es adequa e i o he models. In ou s udy, we ha e assessed he con e gen alidi y h ough he analysis o Pea son p oduc -momen co ela ions coe icien s be ween IPAS and BIS-11. We had signi ican co ela ions be ween he IA subscale and he BIS-11 mo o ac i a ion, a en- ion, and non-planning dimensions. We had signi ican co ela- ions be ween he PM subscale and he BIS-11 mo o ac i a ion, a en ion dimensions, bu no o non-planning dimension. We ha e ound a sligh ly highe co ela ion be ween he o al sco e o BIS-11 and IA han wi h he PM subscale o IPAS. Ou con e gen and Di e gen Validi y We ha e s udied he con e gence be ween he IPAS and he BIS-11. The s a is ically signi ican co ela ions be ween he wo scales (IPAS and BIS-11) and be ween i s dimensions a e de ailed in Tabl e3, showing co ela ions wi h a en ional, mo o , and non-planning dimensions (2, 5, 28). Co ela ions we e s a is ically signi ican , sugges ing he exis ence o con e gence be ween he wo scales. We ha e s udied di e gen alidi y h ough he analysis o co - ela ions be ween IPAS and PCL-R in a subsample o 134 inma es. The s a is ically signi ican co ela ions be ween he wo scales (IPAS and PCL-R) and be ween i s dimensions a e de ailed in Table 3. The IA subscale is no co ela ed wi h PCL-R. The PM subscale is co ela ed wi h PCL-R in e pe sonal, li es yle, and an isocial dimensions (p<0.05). DiscUssiOn Acco ding o AERA-APA-NCME (54), heo e ical and empi ical pieces o e idence ha suppo he in e p e a ions o es sco es a e undamen al so as o indica e he deg ee o which sco es 6 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 esul s we e line wi h Ma hias and colleagues who epo ed signi ican co ela ions be ween he IPAS subscales and BIS-11’s o al sco e (co ela ions coe icien s o 0.39 o AI and 0.26 o PM) (37). Howe e , S an o d and colleagues ha e epo ed highe sig- ni ican co ela ions be ween PM and he o al sco e o BIS-11 (co ela ion coe icien s o 0.21 o AI and 0.38 o PM) (2). Ou da a e eal signi ican co ela ions be ween he h ee dimensions o impulsi i y measu ed by BIS-11and IA subscale o IPAS. Simila esul s we e epo ed by Chen and colleagues ha ob ained signi ican alues o co ela ions be ween IA and he h ee dimensions o BIS-11 (co ela ion coe icien s o 0.15 o a en ional impulsi eness, 0.21 o mo o impulsi eness, and 0.12 o non-planning impulsi eness and AI) (41). In he case o PM subscale o IPAS, we did no obse e signi ican co ela ions wi h he non-planning dimension o BIS-11. The same was epo ed by Chen and colleagues, who only epo ed a signi ican co ela ion o 0.19 be ween PM and a en ional impulsi eness (41). We can explain hese esul s in h ee possible ways. Fi s , i is assumed ha impulsi i y is ela ed agg ession, no doing any disc imina ion be ween dimensions and ype o agg ession (33). These esul s a e in acco dance wi h he hypo hesis ha impul- si i y is ela ed and can be a p edisposing ac o o agg ession (26). On he o he hand, we should no e ha he e a e impulsi e indi iduals who a e no agg essi e, and he e o e, impulsi i y is no he only ac o necessa y o he de elopmen o agg essi e- ness (55). Second, since ou sample is composed o inma es, we can assume a high p e alence o indi iduals wi h subs ance use diso de s and ASPD (56). In hese ypes o pa hology, he e a e a high exp ession o impulsi i y (57, 58). Thus, we can explain he co ela ions be ween subscales o IPAS and BIS-11 subscales because hey a e indi iduals wi h g ea e impulsi i y con e ed by unde lying psychopa hology. The hi d possible explana ion, ela ed o he non-planning dimension o impulsi i y, is pe haps hose impulsi e indi iduals who main ain planning abili y a e hose who can de elop and lea n p emedi a ed agg essi e beha io . These indings suppo he de ini ion o PM in which indi iduals wi h PM plan hei ac ion and, hus, ha e ewe impai men s in ac ion planning impulsi i y dimension (9). The non-planning dimension o impulsi i y is ela ed o wo king memo y (59) and execu i e unc ions, namely wi h he subscale o s a egic planning o he execu i e unc ion index (60). These esul s suppo he need o use o app op ia ed psy- chome ic scales o he e alua ion o agg ession, in o de o go beyond he assump ion ha indi iduals wi h highe le els o impulsi i y a e mo e likely o display ex e nalizing beha io and hus ha ing a g ea e p obabili y o showing agg essi e beha io . This was he a ionale o he use o BIS-11 o he e alua ion o con e gen alidi y wi h IPAS. We ha e e alua ed he di e gen alidi y by analyzing co - ela ions be ween AI and PM ac o s and PCL-R. As a esul , we ha e ob ained signi ican co ela ions o PM and PCL-R in e pe sonal, li es yle, and an isocial dimensions, and no signi ican co ela ions o AI and PCL-R. S an o d and col- leagues, in 2008, e alua ed co ela ions be ween Psychopa hic Pe sonali y In en o y (PPI) and he AI and PM ac o s, ha ing ob ained signi ican ly highe alues o PPI in he indi iduals ca ego ized as p emedi a ed agg esso s, ha is, ha ing highe PM sco es (38). These da a a e in ag eemen wi h he li e a u e on psychopa- hy and he ype o agg essi eness ex e nalized by indi iduals wi h highe exp ession o psychopa hic pe sonali y ai s. Psychopa hy seems o be a isk ac o o PM, mainly in o ensic samples (23). Rega ding he gene aliza ion o esul s, some limi a ions should be conside ed. Al hough we ha e wo ked wi h a la ge sample, i is exclusi ely composed by men, he e o e we do no know i agg essi e women will beha e in he same manne . The pa icipan s a e subjec o long sen ences, making i di icul o gene alize he esul s so as o include inma es wi h sho e sen- ences. Also, pa icipa ion in his kind o esea ch in a o ensic acili y con ex can change he way inma es espond o he IPAS. We mus ake in o conside a ion ha he legal ci cums ances o pa icipan s can modi y he ype o answe s. This sample was ob ained in a p ison o con ic ed indi iduals, in No he n Po ugal. We do no know whe he he scale psychome ic cha ac e is ics a e he same in o he o ensic samples. In u u e esea ch, i may be use ul o conside ools ha can ex e nally quan i y he indi idual ac s o agg ession, such as he Modi ied- O e Agg ession Scale. eThics sTaTeMenT The esea ch p o ocol was o mally app o ed by he E hics Commi ee o Cen o Hospi ala de São João, Po o, Po ugal (Documen numbe 48.14). W i en in o med consen was ob ained a e he p ocedu es had been explained o pa icipan s, in acco dance wi h he Decla a ion o Helsinki. aUThO cOnT iBUTiOns JA and MF-B con ibu ed o he concep ion o he s udy. The da a analysis was ca ied ou by JA and JP. All au ho s con- ibu ed o he in e p e a ion o he da a. JA and MF-B w o e he manusc ip . All au ho s e ised he con en c i ically and app o ed he inal e sion. acKnOWleDgMenTs We would like o hank Le ícia Mal a, MD, and Nu se João Pin o o hei collabo a ion in he acquisi ion o he da a. We would like o hank he Po uguese Gene al Di ec o a e o P ison Se ices o au ho izing he esea ch. FUnDing The au ho s had no unding in he p epa a ion o his manusc ip . 7 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 eFe ences 1. Sie e LJ. Neu obiology o agg ession and iolence. Am J Psychia y (2008) 165(4):429–42. doi:10.1176/appi.ajp.2008.07111774 2. S an o d MS, Hous on RJ, Ma hias CW, Villema e e-Pi man NR, Hel i z LE, Conklin SM. Cha ac e izing agg essi e beha io . Assessmen (2003) 10(2): 183–90. doi:10.1177/1073191103010002009 3. Fel hous AR, Lake SL, Rundle BK, S an o d MS. Pha maco he apy o impulsi e agg ession: a quali y compa ison o con olled s udies. In J Law Psychia y (2013) 36(3–4):258–63. doi:10.1016/j.ijlp.2013.04.017 4. Vi iello B, S o DM. Sub ypes o agg ession and hei ele ance o child psychia y. J Am Acad Child Adolesc Psychia y (1997) 36(3):307–15. doi:10.1097/00004583-199703000-00008 5. Ba a ES, S an o d MS, Dowdy L, Liebman MJ, Ken TA. Impulsi e and p emedi a ed agg ession: a ac o analysis o sel - epo ed ac s. Psychia y Res (1999) 86(2):163–73. doi:10.1016/S0165-1781(99)00024-4 6. Meloy JR. Empi ical basis and o ensic applica ion o a ec i e and p eda o y iolence. Aus N Z J Psychia y (2006) 40(6–7):539–47. doi:10.1080/j.1440- 1614.2006.01837.x 7. Meye JM, Cummings MA, P oc o G, S ahl SM. Psychopha macology o pe sis en iolence and agg ession. Psychia Clin No h Am (2016) 39(4):541–56. doi:10.1016/j.psc.2016.07.012 8. Be kowi z L. Agg ession: I s Causes, Consequences, and Con ol. New Yo k, NY: McG aw-Hill (1993). 9. Cima M, Raine A. Dis inc cha ac e is ics o psychopa hy ela e o di e en sub ypes o agg ession. Pe s Indi id Di (2009) 47(8):835–40. doi:10.1016/j. paid.2009.06.031 10. Bandu a A. Social Founda ions o Though and Ac ion: A Social Cogni i e Theo y. Englewood Cli s, NJ: P en ice-Hall, Inc (1986). 11. Lane SD, Kjome KL, Moelle FG. Neu opsychia y o agg ession. Neu ol Clin (2011) 29(1):49–64, ii. doi:10.1016/j.ncl.2010.10.006 12. Khali a N, Duggan C, S o e s J, Huband N, Völlm BA, Fe i e M, e  al. Pha macological in e en ions o an isocial pe sonali y diso de . Coch ane Da abase Sys Re (2010) (8):CD007667. doi:10.1002/14651858.CD007667. pub2 13. Swogge MT, Walsh Z, Ch is ie M, P iddy BM, Conne KR. Impulsi e e sus p emedi a ed agg ession in he p edic ion o iolen c iminal ecidi ism. Agg ess Beha (2015) 41(4):346–52. doi:10.1002/ab.21549 14. Fazel S, Gula i G, Linsell L. Schizoph enia and iolence: sys ema ic e iew and me a-analysis. PLoS Med (2009) 2009(6):e1000120. doi:10.1371/jou nal. pmed.1000120 15. Rampling J, Fu ado V, Winspe C, Ma waha S, Lucca G, Li anou M, e al. Non-pha macological in e en ions o educing agg ession and iolence in se ious men al illness: a sys ema ic e iew and na a i e syn hesis. Eu Psychia y (2016) 34:17–28. doi:10.1016/j.eu psy.2016.01.2422 16. Swogge MT, Walsh Z, Hous on RJ, Cashman-B own S, Conne KR. Psychopa hy and axis I psychia ic diso de s among c iminal o ende s: ela ionships o impulsi e and p emedi a ed agg ession. Agg ess Beha (2010) 36:45–53. doi:10.1002/ab.20330 17. Woodwo h M, Po e S. In cold blood: cha ac e is ics o c iminal homicides as a unc ion o psychopa hy. J Abno m Psychol (2002) 111(3):436–45. doi:10.1037/0021-843X.111.3.436 18. Ame ican Psychia ic Associa ion. Diagnos ic and S a is ical Manual o Men al Diso de s. 5 h ed. Washing on, DC: Ame ican Psychia ic P ess (2013). 19. Mo an P. The epidemiology o an isocial pe sonali y diso de . Soc Psychia y Psychia Epidemiol (1999) 34(5):231–42. doi:10.1007/s001270050138 20. Co nell DG, Wa en J, Hawk G, S a o d E, O am G, Pine D. Psychopa hy in ins umen al and eac i e iolen o ende s. J Consul Clin Psychol (1996) 64(4):783–90. doi:10.1037/0022-006X.64.4.783 21. F ick PJ, Co nell AH, Ba y CT, Bodin SD, Dane HE. Callous-unemo ional ai s and conduc p oblems in he p edic ion o conduc p oblem se e i y, agg ession, and sel - epo o delinquency. J Abno m Child Psychol (2003) 31(4):457–70. doi:10.1023/A:1023899703866 22. Walsh Z, Swogge MT, Kosson DS. Psychopa hy and ins umen al iolence: ace le el ela ionships. J Pe s Diso d (2009) 23(4):416–24. doi:10.1521/pedi. 2009.23.4.416 23. Blais J, Solodukhin E, Fo h AE. A me a-analysis explo ing he ela ionship be ween psychopa hy and ins umen al e sus eac i e iolence. C im Jus ice Beha (2014) 41(7):797–821. doi:10.1177/0093854813519629 24. Ha e RD. Manual o he Re ised Psychopa hy Checklis , 2nd ed. To on o, ON: Mul i-Heal h Sys ems (2003). 25. Hous on RJ, S an o d MS, Villema e e-Pi man NR, Conklin SM, Hel i z LE. Neu obiological co ela es and clinical implica ions o agg essi e sub ypes. J Fo ensic Neu opsychol (2003) 3:67–87. doi:10.1300/J151 03n04_05 26. Hen ges RF, Shaw DS, Wang MT. Ea ly childhood pa en ing and child impul- si i y as p ecu so s o agg ession, subs ance use, and isky sexual beha io in adolescence and ea ly adul hood. De Psychopa hol (2017):1–15. doi:10.1017/ S0954579417001596 27. Moelle FG, Ba a ES, Doughe y DM, Schmi z JM, Swann AC. Psychia ic aspec s o impulsi i y. Am J Psychia y (2001) 158:1783–93. doi:10.1176/ appi.ajp.158.11.1783 28. Pa on JH, S an o d MS, Ba a ES. Fac o s uc u e o he Ba a Impulsi eness Scale. J Clin Psychol (1995) 51(6):768–74. doi:10.1002/1097- 4679(199511)51:6<768::AID-JCLP2270510607>3.0.CO;2-1 29. S an o d MS, Ma hias CW, Doughe y DM, Lake SL, Ande son NE, Pa on JH. Fi y yea s o he Ba a Impulsi eness Scale: an upda e and e iew. Pe s Indi id Di (2009) 47(5):385–95. doi:10.1016/j.paid.2009.04.008 30. Wo ld Heal h O ganiza ion. Wo ld Repo on Violence and Heal h. Gene a, Swi ze land: Wo ld Heal h O ganiza ion (2002). 31. Ame ican Psychia ic Associa ion. Psychia ic Se ices in Co ec ional Facili ies. A ling on, VA: Ame ican Psychia ic Pub (2015). 567 p. 32. Kendall T, Pilling S, Ty e P, Duggan C, Bu beck R, Meade N, e al. Guidelines: bo de line and an isocial pe sonali y diso de s: summa y o NICE guidance. BMJ (2009) 338(7689):293–5. doi:10.1136/bmj.b93 33. Su is A, Lind L, Emme G, Bo man PD, Kashne M, Ba a ES. Measu es o agg essi e beha io : o e iew o clinical and esea ch ins umen s. Agg ess Violen Beha (2004) 9(2):165–227. doi:10.1016/S1359-1789(03)00012-0 34. Raine A, Dodge K, Loebe R, Ga zke-Kopp L, Lynam D, Reynolds C, e al. The eac i e–p oac i e agg ession ques ionnai e: di e en ial co ela es o eac i e and p oac i e agg ession in adolescen boys. Agg ess Beha (2006) 32(2):159–71. doi:10.1002/ab.20115 35. Kockle TR, S an o d MS, Nelson CE, Meloy JR, San o d K. Cha ac e izing agg essi e beha io in a o ensic popula ion. Am J O hopsychia y (2006) 76(1):80–5. doi:10.1037/0002-9432.76.1.80 36. Conne KR, Hous on RJ, Swo s LM, Meld um S. Reliabili y o he Impulsi e-P emedi a ed Agg ession Scale (IPAS) in ea ed opia e-dependen indi i duals. Addic Beha (2007) 32(3):655–9. doi:10.1016/j.addbeh.2006. 06.026 37. Ma hias CW, S an o d MS, Ma sh DM, F ick PJ, Moelle FG, Swann AC, e  al. Cha ac e izing agg essi e beha io wi h he Impulsi e/P emedi a ed Agg ession Scale among adolescen s wi h conduc diso de . Psychia y Res (2007) 151(3):231–42. doi:10.1016/j.psych es.2006.11.001 38. S an o d MS, Hous on RJ, Bald idge RM. Compa ison o impulsi e and p emedi a ed pe pe a o s o in ima e pa ne iolence. Beha Sci Law (2008) 26(6):709–22. doi:10.1002/bsl.808 39. Haden SC, Sca pa A, S an o d MS. Valida ion o he Impulsi e/P emedi a ed Agg ession Scale in college s uden s. J Agg ess Mal ea T auma (2008) 17(3):352–73. doi:10.1080/10926770802406783 40. Kuyck WG, de Beu s E, Ba end eg M, an den B ink W. Psychome ic e alua ion o he Du ch e sion o he Impulsi e/P emedi a ed Agg ession Scale (IPAS) in male and emale p isone s. In J Fo ensic Men al Heal h (2013) 12(3):172–9. doi:10.1080/14999013.2013.819396 41. Chen FR, Yang Y, Qian M. Chinese e sion o Impulsi e/P emedi a ed Agg ession Scale: alida ion and i s psychome ic p ope ies. J Agg ess Mal- ea T auma (2013) 22(2):175–91. doi:10.1080/10926771.2013.741664 42. Romans L, F esán A, Sen íes H, Sa mien o E, Be langa C, Robles-Ga cía R, e al. Valida ion o he Impulsi e/P emedi a ed Agg ession Scale in Mexican psychia ic pa ien s. No d J Psychia y (2015) 69(5):397–402. doi:10.3109/08 039488.2014.994033 43. Guillemin F, Bomba die C, Bea on D. C oss-cul u al adap a ion o heal h- ela ed quali y o li e measu es: li e a u e e iew and p oposed guidelines. J Clin Epidemiol (1993) 46(12):1417–32. doi:10.1016/0895-4356(93)90142-N 44. on Diemen L, Szobo CM, Kessle F, Pechansky F. Adap a ion and cons uc alida ion o he Ba a Impulsi eness Scale (BIS 11) o B azilian Po uguese o use in adolescen s. Re B as Psiquia (2007) 29(2):153–6. doi:10.1590/ S1516-44462006005000020 45. Ha e D. Psychopa hy as a isk ac o o iolence. Psychia Q (1999) 70(3): 181–97. doi:10.1023/A:1022094925150 8 Aze edo e al. IPAS Validi y in Po uguese Inma es F on ie s in Psychia y | www. on ie sin.o g Feb ua y 2018 | Volume 9 | A icle 10 46. Ha e D, Neumann CS. The PCL-R assessmen o psychopa hy. In: Pa ick CJ, edi o . Handbook o Psychopa hy New Yo k: Guil o d P ess(2006). 58–88. 47. Ha e D, Neumann CS. Psychopa hy as a clinical and empi ical con- s uc . Annu Re Clin Psychol (2008) 4:217–46. doi:10.1146/annu e . clinpsy.3.022806.091452 48. Gonçal es R. Psicopa ia e p ocessos adap a i os à p isão: da in e enção pa a a p e enção. B aga: Uni e sidade do Minho (1999). 49. Lau enschlage GJ. A compa ison o al e na i es o conduc ing Mon e Ca lo analyses o de e mining pa allel analysis c i e ia. Mul i a ia e Beha Res (1989) 24:365–95. doi:10.1207/s15327906mb 2403_6 50. Jackson JE. Oblimin o a ion. In: A mi age P, Col on T, edi o s. Encyclopedia o Bios a is ics. John Wiley & Sons, L d (2005). 6 p. 51. Nunnally JC, Be ns ein IH. Psychome ic Theo y. 3 d ed. New Yo k: McG aw- Hill (1994). 52. Ben le P, Douglas G. Signi icance es s and goodness o i in he analysis o co a iance s uc u es. Psychol Bull (1980) 88:588–606. doi:10.1037/ 0033-2909.88.3.588 53. Ben le PM. EQS 6.1 S uc u al Equa ions P og am. Encino, CA: Mul i a ia e So wa e (2004). 54. Ame ican Educa ional Resea ch Associa ion, Ame ican Psychological Associa ion, Na ional Council on Measu emen in Educa ion. S anda ds o Educa ional and Psychological Tes ing. Washing on, DC: Ame ican Educa ional Resea ch Associa ion (1999). 55. Sca pazza C, Selli o M, di Pelleg ino G. Now o no -now? The in luence o alexi hymia on in e empo al decision-making. B ain Cogn (2017) 114:20–8. doi:10.1016/j.bandc.2017.03.001 56. Fazel S, Danesh J. Se ious men al diso de in 23000 p isone s: a sys ema ic e iew o 62 su eys. Lance (2002) 359(9306):545–50. doi:10.1016/ S0140-6736(02)07740-1 57. K eek MJ, Nielsen DA, Bu elman ER, LaFo ge KS. Gene ic in luences on impulsi i y, isk aking, s ess esponsi i y and ulne abili y o d ug abuse and addic ion. Na Neu osci (2005) 8(11):1450–7. doi:10.1038/nn1583 58. Mann FD, Engelha d L, B iley DA, G o zinge AD, Pa e son MW, Tacke JL, e al. Sensa ion seeking and impulsi e ai s as pe sonali y endopheno ypes o an isocial beha io : e idence om wo independen samples. Pe s Indi id Di (2017) 105:30–9. doi:10.1016/j.paid.2016.09.018 59. Whi ney P, Jameson T, Hinson JM. Impulsi eness and execu i e con ol o wo king memo y. Pe s Indi id Di (2004) 37:417–28. doi:10.1016/j. paid.2003.09.013 60. Spinella M. Sel - a ed execu i e unc ion: de elopmen o he execu i e unc- ion index. In J Neu osci (2005) 115:649–67. doi:10.1080/00207450590524304 Con lic o In e es S a emen : The au ho s epo no con lic s o in e es . The au ho s alone a e esponsible o he con en and w i ing o he pape . Copy igh © 2018 Aze edo, Pais-Ribei o, Coelho and Figuei edo-B aga. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed, p o ided he o iginal au ho (s) and he copy igh owne a e c edi ed and ha he o iginal publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic p ac ice. No use, dis ibu ion o ep oduc ion is pe mi ed which does no comply wi h hese e ms.