The effect of mental workload on the intensity and emotional dynamics of perceived exertion

anales de psicología, 2013, ol. 29, nº 3 (oc ub e), 662-673 h p://dx.doi.o g/10.6018/analesps.29.3.175801 © Copy igh 2013: Se icio de Publicaciones de la Uni e sidad de Mu cia. Mu cia (España) ISSN edición imp esa: 0212-9728. ISSN edición web (h p:// e is as.um.es/analesps): 1695-2294 - 662 - The e ec o men al wo kload on he in ensi y and emo ional dynamics o pe cei ed exe ion Da id Cá denas1*, José C. Pe ales2, Luis J. Chi osa1, José Conde-González3, Daniel Aguila -Ma ínez1 y Se gio A aya4 1 Depa men o Physical Educa ion and Spo . Facul y o Spo Sciences. Uni e si y o G anada. Spain 2 Depa men o Expe imen al Psychology. Facul y o Psychology. Uni e si y o G anada. Spain 3 Depa men o Physical Educa ion and Spo . Facul y o Educa ion Sciences. Uni e si y o Se illa. Spain 4 Depa men o Physical Educa ion and Spo . Uni e si y A u o P a . (Iquique) Chile Tí ulo: El e ec o de la ca ga de abajo men al en la in ensidad y la dinámi- ca emocional del es ue zo pe cibido. Resumen: El es ue zo pe cibido, medido po escalas psicomé icas, ha demos ado se una he amien a álida pa a e alua la ca ga de en ena- mien o, que co elaciona al amen e con dimensiones isiológicas y mecáni- cas del es ue zo ísico. Sin emba go, poco se sabe sob e los co ela os emo- cionales del es ue zo, y cómo la pe cepción del es ue zo es in luenciada po la ca ga de abajo men al. En los dos expe imen os desc i os, se encon ó que la alo ación del es ue zo pe cibido (RPE) es aba signi ica i amen e in- luenciada po la ca ga men al (gene ada po medio de una a ea cogni i a, sin elación, pe o empo almen e supe pues a con la a ea ísica) du an e la ecupe ación ac i a después de habe alcanzado el ago amien o du an e el eje cicio, pe o no du an e el eje cicio inc emen al. Es impo an e des aca que el es ue zo pe cibido co elacionó ue emen e con los alo es emocio- nales / hedónicos de alencia, pe o no así con los alo es de la ac i ación. Es os hallazgos e ue zan el alo mo i acional del es ue zo pe cibido y su inculación con o os cons uc os psicológicos. Palab as cla es: Ca ga men al; es ue zo pe cibido; es ue zo ísico; a iga; eje cicio; mo i ación. Abs ac : Pe cei ed exe ion, as measu ed by psychome ic scales, has been p o en o be a alid ool o assess aining load, and o highly co e- la e wi h physiological and mechanical dimensions o physical e o . How- e e , li le is known abou he emo ional co ela es o exe ion, and how pe cei ed exe ion is in luenced by men al wo kload. In he wo expe i- men s epo ed he e, a ings o pe cei ed exe ion (RPE) we e ound o be signi ican ly in luenced by men al wo kload (gene a ed by means o a cog- ni i e ask, un ela ed o, bu empo ally o e lapping wi h he physical ask) du ing ac i e eco e y a e exhaus ing exe cise, bu no du ing inc emen al exe cise. Impo an ly, pe cei ed exe ion was ound o s ongly co ela e wi h epo ed emo ional/hedonic alence, bu no so igh ly wi h epo ed a ousal. These indings s eng hen he mo i a ional alue o pe cei ed ex- e ion, and i s linkage o o he psychological cons uc s Key wo ds: Men al wo kload; pe cei ed exe ion; physical e o ; a igue; exe cise; mo i a ion. In oduc ion Pe cei ed exe ion is a key concep in he psychology o ex- e cise, and may be de ined as he subjec i e in ensi y o e - o , s ain, discom o , and/o a igue ha is expe ienced du ing physical exe cise (Robe son & Noble 1997). Fo ex- ample, le us imagine an a hle e unning a ma a hon. Du ing he ace, he ma a hone will con inuously moni o he sense o e o , and will decide on-line how as o un in o de o each he inish line sho ly be o e exhaus ion. In gene al, people use pe cei ed exe ion o decide when o s op exe - cising. Ou ma a hone will p obably un beyond he poin a which she would ha e s opped in ci cums ances o he han compe i ion. Mo eo e , e o ole ance depends on psychological ac o s such as mood (A uza, Balagué & A - ie a, 1998), a ousal, and concomi an men al wo kload. Con- sequen ly, unde s anding hese ac o s is c ucial o imp o e decision making in a hle es (Schome , 1986). Un o una ely, despi e he ac ha pe cei ed exe ion is a psychological cons uc , li le esea ch has been done on how hese ac o s in luence pe cei ed exe ion, o on he ela ionship be ween pe cei ed exe ion and o he psychological cons uc s, in- cluding hose ha de ine he subjec i e a ec i e/emo ional s a e (e.g. Baden, McLean, Tucke , Noakes, & Hun e , 2005; Beniscelli & To eg osa, 2010). * Di ección pa a co espondencia [Co espondence add ess]: Da id Cá denas Vélez. Ca e e a Al aca s/n CP: 18071 G anada (Spain). E-mail: [email p o ec ed] In ace o he impo ance o pe cei ed exe ion, se e al psychome ic scales ha e been de eloped o measu e i . Ac- ually, he ela ionship be ween mechanical and physiological measu es o e o (mainly powe ou pu , lac a e concen a- ion, oxygen consump ion, and ca dio ascula indices), on he one side, and subjec i ely pe cei ed exe ion, on he o he , cons i u es a discipline by i sel : he psychophysics o e o (Bo g, 1982; Noble & Robe son, 1996; Bo g & Bo g, 2002; Bo g & Kaijse , 2006; see Faulkne , Pa i & Es on, 2008, o a ecen e iew). The p ecise cha ac e iza ion o he unc ions ha ela e he objec i e and subjec i e dimensions o exe ion has helped spo people o be e unde s and and o imp o e aining me hods (see, o example, Ma io & Lamb, 1996; Swee , Fos e , McGuigan, & B ice, 2004; Fos- e e al., 2001). The p esen wo k is aimed a in es iga ing (1) whe he , and how, pe cei ed exe ion is a ec ed by men al wo kload du ing exe cise (cycling) up o oli ional exhaus ion, and du ing ac i e eco e y. In o he wo ds, we in end o es he common in ui ions ha physical asks in ol ing concomi an complex in o ma ion p ocessing (as, o example, planning o decision-making) a e subjec i ely pe cei ed as mo e e - o ul han plainly physical asks, and ha cogni i e e o can hinde eco e y a e exe cise. And, (2) we explo e how he subjec i ely es ima ed a ec i e s a e o he exe cise is a ec ed by e o , and modula ed by men al wo kload. In ope a ional e ms, men ally loading asks a e hose e- qui ing he managemen o signi ican amoun s o in o ma- ion in a non-au oma ic way. These asks ha e been claimed o equi e a en ional esou ces and o in e e e wi h o he The e ec o men al wo kload on he in ensi y and emo ional dynamics o pe cei ed exe ion 663 anales de psicología, 2013, ol. 29, nº 3 (oc ub e) cogni i e asks (Kahneman, 1973), o mobilize cen al e- sou ces (Wickens, 1984), o o equi e he in ol emen o he cen al execu i e sys em (Baddeley, 2003) – he cogni i e mechanism necessa y o planning ac i i ies when he ask canno be au oma ized, o gene a es esponse con lic (Mi- yake & Shah, 1999) –. In addi ion, cogni i e wo kload has emo ional co ela es: he e o associa ed wi h cogni i e wo kload is hedonically nega i e, and inc eases a ousal (Baumeis e , Vohs & Tice, 2007; Wallace & Baumeis e , 2002). The in e es on he in e ac ion be ween cogni ion and physical e o is no new. A ew s udies ha e in es iga ed he ole o men al a igue12on pe cei ed exe ion, in con- secu i e men al-physical asks (Ma co a, S aiano, & Man- ning, 2009). Rela edly, some s udies ha e ocused on he opposi e causal di ec ion, ha is, on he e ec o physical e - o and a igue on cogni i e pe o mance (Sanab ia e al., 2011). To ou knowledge, howe e , ou s udy is he i s o expe imen ally manipula e he magni ude o men al wo k- load du ing and a e exe cise, and o check i s e ec on pe cei ed exe ion (and ela ed emo ional dimensions). To sum up, ou aims a e: (1) o quan i y he deg ee o which men al wo kload con ibu es o pe cei ed exe ion du ing exe cise and eco e y; and (2) o analyse he emo- ional dynamics esul ing om he combined e ec s o physical and cogni i e load. In Expe imen 1, men al wo k- load was manipula ed by p esen ing pa icipan s wi h cogni- i e asks du ing exe cise, un il exhaus ion, and du ing e- co e y. N-back and Go/No-go asks we e used o gene a e ei- he high o low men al wo kload. Load was manipula ed ac oss sessions, and compa ed o a con ol session wi hou any concomi an men al ask. Ra ings o pe cei ed exe ion we e egis e ed du ing each session, and compa ed ac oss sessions wi h di e en men al wo kload le els. In Expe i- men 2, wo men al wo kload condi ions, and one condi ion wi hou men al wo kload we e compa ed. This second ex- pe imen was ca ied ou o dissocia e he po en ial e ec o men al wo kload om he e ec o cogni i e pe o mance eedback. Simul aneously o RPE (Ra ing o Pe cei ed Exe - ion; Bo g, 1982), we eco ded wo emo ional dimensions (a ousal and hedonic alence) du ing he whole o each ses- sion, wi h he aim o in es iga ing he emo ional ea u es o concomi an physical and men al wo kload. 12Cogni i e wo kload is di ec ly p opo ional o he amoun o in o ma ion o be managed, and he complexi y o he ope a ions pe o med on i . We assume ha subjec i ely pe cei ed cogni i e e o pa allels cogni i e wo k- load, in a way simila o how pe cei ed exe ion pa allels physical load. Fol- lowing he analogy, we dis inguish be ween men al a igue (namely, a educ- ion in men al pe o mance, mainly, bu no only, a ibu able o sus ained cogni i e e o ) and cogni i e e o . Recen s udies ha e ac ually dissoci- a ed men al e o om men al a igue, by measu ing hei di e en ial e - ec s on eye mo ili y (Di S asi, An olí, & Cañas, 2011; Di S asi, Renne , S aeh , Helme , Velichko sky, Cañas, Ca ena, & Pannasch, 2010). Expe imen 1 Two al e na ing execu i e unc ion asks we e used o di- ec ly manipula e cogni i e wo kload, a h ee le els (High load, Low load, No load), du ing exe cise. Pe cei ed exe - ion and oxygen consump ion we e moni o ed du ing moun ing exe cise, a oli ional exhaus ion, and du ing ac- i e eco e y. The main wo king hypo hesis was ha he sense o physical e o (RPE) du ing exe ion and/o e- co e y would be di ec ly modula ed by he le el o cogni i e wo kload. Me hod Pa icipan s 18 pa icipan s (4 women and 14 men; a e age age 21.17 yea s; age ange 19-26 yea s) ook pa in Expe imen 1. All o hem we e Spo Sciences s uden s a he Uni e si y o G anada. In acco dance wi h e hical s anda ds, all o hem we e in o med o he expe imen ’s aims and condi ions, and signed an in o med consen o m. In addi ion, a medical doc o was a ailable in he building du ing he whole du a- ion o each expe imen al session, and was in o med when such an expe imen al session was being ca ied ou . A de i- b illa ion de ice was also a each. Two pa icipan s did no epo eaching 18 poin s in he epo ed exe ion scale and we e excluded om u he analyses (as his was a maximum e o es , we conside ed ha no eaching a pe cei ed exe ed e o close o he maximum implied no ha ing unde s ood he logic o he scale, o no ha ing exe ed he maximum possible e o du ing he es ). Appa a us and s imuli A Ca dioline® x 100 cycloe gome e was used o he physical e o ask. The s imuli o he cogni i e load asks we e p ojec ed on a sc een loca ed app oxima ely wo me- es in on o he pa icipan . S imuli p esen a ion was con- olled om a lap op loca ed on he le o he pa icipan . Responses o he cogni i e ask we e eco ded by using a wi eless nume ic keyboa d a ached o he cycloe gome e handle, and connec ed o he lap op con olling he ask. The pa icipan could espond by p essing any key o he keyboa d wi h he humb, while g asping he handle. Each s imulus was a numbe o le e (depending on he ask), and he pa icipan was asked o espond o by p essing/no p essing any key o he keyboa d. A dis inc i e sound was used as eedback o each esponse. P ocedu e and design Physical e o ask. Each session consis ed o h ee s ages (see Figu e 1). In he wa ming up s age, pa icipan s cycled on he cycloe gome e wi hou esis ance o 5 minu es. In he 664 Da id Cá denas Vélez e . al. anales de psicología, 2013, ol. 29, nº 3 (oc ub e) second s age, pa icipan s we e asked o cycle in inc emen al 2-minu e-long exe cise in ensi y le els (hence o h, e o le els) un il oli ional exhaus ion. Du ing his s age, pa ici- pan s we e asked o cycle a a pace o 70 pm. When pace de ia ed om ha e e ence, one o he expe imen e s wa ned he pa icipan o inc ease o dec ease he pace. In he i s e o le el in his s age, he le el o physical e o was se a a numbe o wa s calcula ed as hal he pa ici- pan ’s weigh (measu ed in kilog ams). In o he wo ds, a pe son weigh ing 70 kilog ams s a ed he e o s age mobi- lizing 35 wa s. In each successi e le el, he e o was in- c eased bodyweigh /2 wa s. Following he same example, a pe son weigh ing 70 kg mobilized 70 wa s in he second le el, 105 in he hi d one, and so on. A e exhaus ion, e- sis ance was emo ed, bu he pa icipan was asked o keep he pace a 70 pm du ing 8 minu es mo e ( eco e y s age). Figu e 1. The h ee s ages o he session. Cogni i e load asks. In he load condi ions, du ing inc emen al e o and eco e y, pa icipan s pe o med wo cogni i e load asks in an al e na ing ashion. The i s one is known as he N-back ask (see Owen, McMillan, Lai d, & Bullmo e, 2005). In ou e sion o he ask a se ies o digi s appea ed successi ely on he sc een, a a a e o one digi e e y 2500 miliseconds (wi h a s imulus du a ion o 2000 ms), and he pa icipan was asked o epo whe he he numbe on sc een was he same as he one p esen ed N posi ions ea - lie . Once he esponse was made (“p ess i yes, wi hhold e- sponse i no”) and he ollowing digi appea ed, he p e ious compa ison digi had o be disca ded and he con en o he wo king memo y upda ed wi h he new one. Only digi s 1 o 4 we e used in he p esen e sion o he ask. The digi ap- pea ing in each ial was andomly selec ed, which means ha , on a e age, he p esen digi was he same as he one N posi ions ea lie , and hus p essing was co ec , in 25% o he ials. The second ask is called he Go-No go ask (see Gómez, Ra cli , & Pe ea, 2007; Pe ales, Ve dejo-Ga cía, Moya, Lozano, & Pé ez-Ga cía, 2009). In his ask he pa icipan is p esen ed wi h one o wo s imuli in each ial ( o example, one o wo le e s), and is simply asked o espond when one o hem appea s on sc een (one o he le e s) and o wi h- hold he esponse in p esence o he o he . The i s s imu- lus is called he Go s imulus, and appea s in 3 ou o 4 ials; he o he one is he No-go s imulus and appea s in 1 ou o 4 ials. The a e o appea ance o s imuli and he du a ion o each ial we e he same in he wo asks. Bo h he N-back and he Go/No-go ask equi e he in- ol emen o he cen al execu i e sys em and hus gene a e signi ican cogni i e wo kload. Howe e , he deg ee o wo kload can be easily adjus ed o he expe imen e ’s needs. In he case o he N-back ask, he la ge N, he mo e de- manding he ask. No e ha N de e mines how many digi s he pa icipan needs o keep in wo king memo y and ope - a e wi h. 1-back asks a e subjec i ely pe cei ed as e y easy, 2-back ask as mode a ely demanding, and 3-back asks as e y di icul . Two le els o wo kload we e used in he p e- sen expe imen : 1-back (low load), and 3-back (high load). Simila ly, he wo kload gene a ed by Go/No-go asks can be a ied by manipula ing he complexi y o he e- sponse c i e ion in each ial. In he simple c i e ion (low load) e sion, he pa icipan is jus asked o espond o one le e and no o espond o he o he . In he double c i e- ion (high load) e sion, each o he wo le e s can appea wi hin a squa e o wi hou any o he s imulus p esen . In- s uc ions demand a esponse o one o he le e s when i appea s alone, bu no i ha same le e appea s wi hin he squa e; simila ly, he pa icipan is ins uc ed no o espond o he o he le e alone, bu o espond i ha le e appea s wi hin a squa e. Thus, he disc imina ion be ween he go and he no-go s imuli canno be made on he basis o one ea u e alone ( he iden i y o he le e o he p esence o he squa e), bu only on he basis o he combina ion o bo h. In he p esen expe imen , each pa icipan did he whole ask h ee imes (one o each cogni i e load condi- ion, wi h a sepa a ion o , a leas , 72 hou s be ween wo consecu i e ones). In one o hem, he e o es was ca ied ou wi hou any concomi an men al wo kload ask (No load condi ion); in ano he one wi h he low load asks (Low load condi ion); and, in he emaining one, wi h he high load asks (High load condi ion). The o de o condi ions was coun e balanced. Load was he main independen a iable. In he wo load condi ions, he ype o ask (N-back, Go/No-go) al e na ed be ween blocks. The du a ion o each cogni i e load ask block coincided wi h he du a ion o each inc emen al e o block (2’). Du ing eco e y (8’), blocks we e sepa a ed only by pe cei ed exe ion assessmen s, and he cogni i e asks kep on al e na ing e e y wo minu es. The e ec o men al wo kload on he in ensi y and emo ional dynamics o pe cei ed exe ion 665 anales de psicología, 2013, ol. 29, nº 3 (oc ub e) Reco ding o pe cei ed exe ion. App oxima ely ou seconds we e equi ed o ep og am he cycloe gome e be ween e o le els. While one expe imen e did his, a second one ques- ioned he pa icipan abou his/he le el o pe cei ed exe - ion ( his one also checked ha he pedalling pace emained a 70 pm). The RPE 6-20 scale23(Bo g, 1998) was used o collec local, cen al, and o al sco es o pe cei ed exe ion. Be o e each session, each pa icipan was amilia ised wi h he scale. S anda dised ins uc ions we e p o ided on how o epo eelings o exe ion du ing exe cise in he h ee di - e en subscales, ollowing he p ocedu e desc ibed by Swank, S einel, and Moo e (2003). Pe cei ed exe ion a ings we e collec ed a e e e y wo minu es (a e e e y cogni i e ask block, and so exac ly be- o e he e o le el was inc eased), a exhaus ion, and a e e e y wo minu es du ing eco e y. The numbe o measu - ing poin s du ing exe cise a ied om one pa icipan o an- o he , as he numbe o e o le els pe o med be o e ex- haus ion depended on hei i ness le el. Reco ding o oxygen consump ion and hea a e. Oxygen consump- ion was eco ded o all pa icipan s du ing he e o and eco e y s ages. A K4 b2 ® gas composi ion analyze was used. The gas analyze was submi ed o he manu ac u e ’s o ices o echnical e ision be o e he expe imen began ( echnical e isions a e ecommended o his model e e y 6 mon hs), and was indi idually calib a ed o each session, ollowing he manu ac u e ’s ins uc ion manual. The in- s umen gene a es an obse a ion poin o each inhal- ing/exhaling cycle in ml/min·kg, compu ed om he di e - ence be ween concen a ions o oxygen in inhaled and ex- haled ai . Hea a e was moni o ed by using a Pola ® w is de- ice. BPM measu es as eco ded by his de ice, and oxygen consump ion measu es nea exhaus ion, we e used o con- di ion compa abili y checks only. Online eco dings o comple e sessions, be o e and a e exhaus ion can be p o- ided by he co esponding au ho on demand. Resul s Condi ion compa abili y checks E o le el escala ed un il he pa icipan epo ed being unable o exe cise any longe a he same le el. As no ed abo e, only he pa icipan s who eached 18 poin s o mo e in he RPE-To al subscale we e conside ed o u he analyses. Mean RPE a he exhaus ion poin did no a y ac oss condi ions, which ensu es ha , a his poin , he h ee condi- 23The RPE 6-20 scale (Noble & Robe son, 1996) has been cus oma ily used in he las decades o assess pe cei ed exe ion. Al hough some ecen e i- dence has shown a ce ain supe io i y o CR10 scales (Bo g, & Kaijse , 2006) o e he RPE 6-20 scale, such a po en ial supe io i y has mo e o do wi h he unc ions ela ing pe iphe al physiological indexes o pe cei ed exe ion, han wi h sensi i i y o he ool o i s cons uc a iable. ions we e equa ed in subjec i ely pe cei ed exe ion. A 3 (Load: high load, low load, no load) x 3 (RPE subscale: local, cen al, o al) wi hin-subjec ANOVA did no yield any sig- ni ican e ec . Mean (SE) RPE sco es o he high-load, he low-load, and he no-load condi ions we e 19.48 (.30), 19.56 (.25), and 19.54 (.27) [F<1]. The Subscale x Load in e ac ion was no signi ican ei he [F(4, 60)=2.03; MSE=.17, p=.10]. Howe e , he e ec o he subscale was a he close o sig- ni icance. Mean (SE) sco es o he local, cen al, and o al scales we e 19.60 (.15), 19.42 (.19), and 19.56 (.13) [F(2, 30)=3.02; MSE=.15; p=.064]. On he o he hand, hea a e was moni o ed du ing e - o and eco e y. A exhaus ion, mean (SE) pe cen ages o e he indi idually es ima ed maximum hea a e (com- pu ed as 220 minus age) o he h ee condi ions we e 92.44% (1.54), 95.11% (1.00), and 93.90% (1.74) [F(2, 30)=2.29, MSE=12.50; p=.12]. This ensu es, i s , ha pa - icipan s we e ac ually close o exhaus ion when hey e- po ed being exhaus ed, and, second, ha he h ee condi- ions we e equa ed in ha momen . As will be discussed la e , any di e ence ound om ha poin beyond could be a ibu ed o he men al load manipula ion. Fo each pa icipan and each condi ion, we a e aged VO2 measu es o consecu i e 30 s. in e als, co esponding o app oxima ely 10 minu es ( he las 30 second in e al is incomple e o some pa icipan s who exhaus ed du ing ha in e al; a e aging was used o s abilize he measu e). Oxy- gen consump ion g adually inc eased du ing exe cise, almos linea ly wi h espec o he physical e o exe ed du ing he ask. Only he e ec o in e al was signi ican , F(19, 285)=316.78, MSE=9.00, p<.001. Nei he he e ec o Load no he Block x Load in e ac ion was close o signi icance (F<1 in bo h cases)3.4 And inally, we also collec ed oxygen consump ion, a e - aged o he las 30 seconds p io o exhaus ion. Mean (SE) VO2 alues, measu ed in ml/kg·min we e 43.15 (2.85), 44.97 (2.65), and 44.83 (2.98), o he high-load, he low-load, and he no load condi ions, espec i ely. These alues did no di e signi ican ly among hem (F<1). Th ee eco dings om wo pa icipan s seemed clea ly anomalous (VO2<25), so we also ca ied ou he same analysis disca ding hose pa icipan s. Mean (SE) VO2 alues, we e hen 44.51 (2.76), 47.12 (2.50), and 48.38 (2.00). The e ec o Cogni i e load emained non-signi ican [F(2, 26)=1.08, MSE=50.58, p=.36]. Cogni i e in ol emen check In bo h he N-back and he Go/No-go asks, a ge s imuli a e hose o which a esponse is igh , and dis ac- 34Ou cycloe gome e does no p o ide an independen es ima e o powe ou pu . Howe e , gi en he close in a-indi idual ela ionship be ween powe ou pu and oxygen consump ion, compa able cu es o oxygen con- sump ion ac oss condi ions ensu e ha he absence o di e ences among condi ions in RPE du ing moun ing e o is no due o di e ences in objec- i e physical e o . 666 Da id Cá denas Vélez e . al. anales de psicología, 2013, ol. 29, nº 3 (oc ub e) o s a e hose no mee ing he c i e ion o esponding. I he pa icipan succeeds o some deg ee in he cogni i e ask, he a e o esponding mus be highe o a ge s han o dis ac o s. The hi a e (h) is cus oma ily de ined as he a e o esponding o a ge s, whe eas he alse ala m a e ( ) is he a e o esponding o dis ac o s. The e a e a numbe o disc iminabil y measu es based on he di e ence be ween hi and alse ala m a es; he e, we will use a cus oma ily ac- cep ed non-pa ame ic sco e (see Pe ales, Ca ena, Shanks, & González, 2005): he di e ence o a csines [DA=acos( )-acos(h)]. We compu ed disc iminabili y (DA) o 30-second in e - als ac oss he whole ask. Gi en ha all pa icipan s in all condi ions comple ed a leas 5 e o le els, we ha e 20 dis- c iminabili y sco es o all o hem, plus ano he 16 o he 8-minu e eco e y s age. Fo he sake o simplici y, and in o de o s abilize measu emen , we a e aged disc iminabili y sco es o he ou 30-second in e als wi hin each 2-minu e e o le el. In addi ion, we analyzed Go/No-go and N-back oge he . The same disc iminabili y analysis can be applied o he wo asks, so he ype o ask mus be conside ed a coun e balanced ac o . Figu e 2 displays a e aged dis- c iminabili y le els o each 2-minu e le el du ing e o ( op panel) and eco e y (bo om panel), o he wo Load condi- ions (high-load and low-load; ob iously he e was no cogni- i e ask o check in he no-load condi ion). Du ing inc e- men al e o , he Cogni i e load (high, low) x Le el (1-5) ANOVA yielded a signi ican e ec o Load [F(1, 30)=102.09, MSE=.20, p<.01]. Nei he he e ec o Le el no he Load x Le el in e ac ion we e signi ican [F(4, 120)=1.70, MSE=.08, p=.15, and F(4, 120)=1.62, MSE=.08, p=.17, espec i ely]. Du ing eco e y, he e was again a main e ec o Cogni- i e Load [F(1, 30)=67.33, MSE=.21, p<.01], wi h he ex- pec ed be e pe o mance in he low-load condi ion. Addi- ionally, pe o mance mode a ely imp o ed wi h p og essi e eco e y [F(3, 90)=3.08, MSE=.09, p=.03]. The Cogni i e load x Reco e y le el in e ac ion did no each signi icance [F(3, 90)=1.58, MSE=.09, p=.20]. In summa y, disc iminabili y in he wo le els was well abo e ze o o all measu emen poin s, which is in e p e - able as a signi ican in ol emen o pa icipan s wi h he ask du ing exe cise. Addi ionally, as expec ed, pe o mance was signi ican ly be e o he low-load ask. In nei he o he wo load condi ions did pe o mance wo sen wi h inc e- men al exe ion al hough in bo h o hem, i mode a ely im- p o ed du ing eco e y. Time o exhaus ion Mean ime o exhaus ion did no di e ac oss Cogni i e load condi ions. Pa icipan s exe cised du ing 774 (23.02), 752 (28.45), and 765 (36.44) seconds, in he high-, low-, and no-load condi ions, espec i ely [F<1]. In o he wo ds, cog- ni i e wo kload did no accele a e exhaus ion. Figu e 2. A e aged disc iminabili y le els o each 2-minu e le el du ing e - o ( op panel) and eco e y (bo om panel), o he wo Load condi ions (high-load and low-load). RPE du ing inc emen al e o As no ed abo e, all pa icipan s in all condi ions comple ed a leas i e e o le els. Consequen ly, we ha e i e RPE measu emen poin s du ing e o (plus he ini ial measu emen aken be ween he end o he wa ming up s age and he inc e- men al e o s age). An E o le el (0-5) x Cogni i e load (high-load, low-load, no-load) x RPE subscale (Local, Cen al, To al) wi hin-subjec ANOVA was ca ied ou on he ou i s RPE measu emen s o all pa icipan s. Figu e 3 shows how RPE escala es wi h e o o he h ee cogni i e load condi ions and he h ee subscales. The e ec o E o le el was ob iously s ongly signi ican , wi h RPE g adu- ally inc easing wi h physical e o , F(5, 75)=212.72, MSE=9.1, p<.01. Cogni i e load, howe e , did no ha e any di ec o in- e ac i e e ec on RPE [F<1; F(4, 60)=1.04; MSE=.70; p=.39; F(10, 150)=1.53; MSE=2.60; p=.13; and F(20, 300)=1.51, MSE=.20; p=.08, o he ma ginal Load e ec , he Load x Le el in e ac ion, he Load x Subscale in e ac ion, and he 3-way in- e ac ion, espec i ely]. The e ec o men al wo kload on he in ensi y and emo ional dynamics o pe cei ed exe ion 667 anales de psicología, 2013, ol. 29, nº 3 (oc ub e) Figu e 3. RPE o he h ee cogni i e load condi ions and he h ee subscales. RPE sco es di e ed be ween subscales [F(2, 30)=5.95, MSE=1.50, p=.01], wi h local and o al subscales yielding highe sco es han he cen al one. Mean (SE) sco es o he local, cen al, and o al subscales we e 11.41 (1.29), 11.06 (1.43), and 11.27 (1.47), espec i ely. Planned compa isons yielded a signi ican di e ence be ween he local and he cen al sco e (p=.02), as well as be ween he cen al and he o al sco e (p<.01), bu no be ween he local and he o al sco e (p=.16). In addi ion, Subscale in e ac ed wi h E o le el [F(10, 150)=3.00; MSE=.30; p<.01]. As can be seen in Figu e 3, pe cei ed exe ion accumula ed as e locally han cen ally, and o al e o seems o be based mo e di ec ly on local e o han in ca dio- espi a o y dis ess. RPE du ing eco e y A 3 (Load) x 5 (Le el: exhaus ion, eco e y 1, eco e y 2, eco e y 3, eco e y 4) x 3 (Subscale) wi hin-subjec ANOVA on RPE sco es yielded a signi ican e ec o Load, F(2, 30)=3.25, MSE=19.6, p=.05. Planned compa isons showed a signi ican di e ence be ween high and no-load (p=.03). As can be seen in Figu e 4 ( op panel), low load mean RPE was be ween he high- and he low-load condi- ions, and was no signi ican ly di e en om any o he wo. Mean (SE) RPE alues we e 12.49 (2.31), 12.07 (1.90), and 11.47 (1.65) o he high-load, low-load, and no-load condi ions espec i ely. T end analyses showed ha he e - ec o Load on RPE was exclusi ely linea (p=.03; F<1 o he o he componen s). Ob iously, he e ec o E o le el was s ongly signi i- can , F(4, 60)=222.73; MSE=13.2; p<.01. Mo e in e es ing a e he e ec s o Subscale, F(2, 30)=12.31; MSE=2.60; p<.01, and he Le el x Subscale in e ac ion, F(8, 120)=2.70, MSE=1.00; p<.01. Mean (SE) RPEs we e 12.40 (1.87), 11.68 (1.74), and 11.94 (1.77) o he local, cen al, and o al sub- scales, espec i ely. Planned compa isons yielded signi ican di e ences be ween he local and he cen al subscales (p<.01), and be ween he cen al and he o al subscale (p<.01), bu also be ween he local and he o al subscale (p<.01). As i can be seen in Figu e 4 (bo om panel), he lo- cal and o al subscales we e he ones mos a ec ed by e o ; howe e he in e ac ion is due o he ac ha local RPE e- mained highe han cen al RPE, om exhaus ion o he end o eco e y, whe eas he o al RPE dec eased mo e apidly and ma ched cen al RPE by he end o eco e y ( o he sake o b e i y, we do no analyze his in e ac ion in de ail). Expe imen 2 Expe imen 1 showed a limi ed bu ele an e ec o cogni- i e wo kload on pe cei ed exe ion. This e ec did no each signi icance du ing inc emen al e o , bu did each i du ing eco e y. Visually, he high-load condi ion seemed o boos pe cei ed exe ion a he beginning o he ask, bu ha di e ence anished as exe cise con inued. Impo an ly, i clea ly eappea ed (and eached signi icance) du ing eco - e y. On he o he hand, men al wo kload did no in e ac wi h he ype o subscale. In o he wo ds, al hough he h ee subscales showed sligh ly di e en dynamics, hese emained una ec ed by he cogni i e wo kload manipula ion. Ne e heless, he e ec o cogni i e wo kload can be explained in an al e na i e way. Highly loading cogni i e asks a e no only mo e cogni i ely demanding, bu hey also elici mo e e o s, which make hem e en mo e a ousing and, po en ially, hedonically nega i e. In Expe imen 2 we ied o eplica e he pa e n o e ec s ound in Expe imen 1, bu a i icially manipula ed he balance be ween posi i e and nega i e eedback o esponses in he cogni i e ask, wi hou ac ually manipula ing he di icul y o he ask. Ad- di ionally, he emo ional s a e o he pa icipan du ing he whole ask was moni o ed. 668 Da id Cá denas Vélez e . al. anales de psicología, 2013, ol. 29, nº 3 (oc ub e) Figu e 4. Mean RPE sco es du ing exe cise and eco e y. Me hod Pa icipan s, appa a us and asks 18 pa icipan s (9 women and 9 men; a e age age 21.72 yea s; age ange 18-26 yea s) ook pa in he expe imen . Pa icipan ec ui ing and appa a us we e iden ical o he ones in Expe imen 1. The cogni i e ask was a alse sequence- acking es . Pa icipan s we e p esen ed wi h a se ies o digi s, one pe ial (wi h du a ion and in e -s imuli in e als iden ical o he ones in Expe imen 1). They we e old ha digi gene a ion ollowed a hidden pa e n, and hei ask was o disco e ha pa e n. They we e also old ha no all he digi s, bu only a high p opo ion o hem ollowed he pa e n, so o each ial, hey we e asked o indica e whe he he cu en digi ollowed he pa e n o no (“p ess i yes, wi hhold you esponse i no”). In ac , he e was no pa e n o dis- co e , and he eedback was p og ammed depending on he expe imen al condi ion. In he Load/Nega i e eedback condi ion, pa icipan s we e gi en alse posi i e eedback (a dis inc i e sound) in one ou h o he ials, and nega i e eedback in he o he h ee ou hs. In he Load/Posi i e eedback condi ion, on he o he hand, hey we e gi en posi- i e eedback in h ee ou hs o he ials, and nega i e eedback in he o he ou h. Du ing he ask, none o he pa icipan s epo ed ha ing disco e ed ha eedback was ac ually un ela ed o he ask. In he con ol no-load condi- ion, no cogni i e ask was p esen ed concomi an ly o he physical ask. Apa om he a iables desc ibed in Expe imen 1 (RPE subscales, ime o exhaus ion, hea a e, and oxygen consump ion), we also collec ed sco es o emo ional s a e. The Sel -assessmen manikin (SAM-Spanish e sion, Mol ó e al., 1999) is a sel - epo scale in which pa icipan s a e asked o assess hei own emo ional s a e, elici ed by some en i- onmen al e en , in h ee dimensions: Valence, A ousal, and Dominance. Only he alence and a ousal scales we e used in his expe imen . The esponse is gi en by choosing one o 5 icons displayed ho izon ally on a shee (o he in e media e poin be ween wo o hem). The icons ep esen p og es- si e emo ional s a es ( om low o high a ousal, and om nega i e o posi i e alence). The esponse is coded as a alue be ween 1 and 9. Pa icipan s we e asked o epo hei emo ional s a e in he wo dimensions a e each wo- minu e e o le el. Condi ion compa abili y checks As desc ibed o Expe imen 1, only hose pa icipan s who epo ed o al pe cei ed exe ion sco es close o he maximum (RPE-To al >= 18) we e aken in o accoun , which le 12 pa icipan s o all u he analyses. Again, we checked o possible di e ences be ween Load condi ions a he exhaus ion poin . Mean (SE) RPE alues we e 19.61 (.27), 18.97 (.55), and 19.41 (.26) o he High Load, he Low Load, and he No Load condi ions, e- spec i ely. As happened in Expe imen 1, hese alues we e close o he maximum (20) and did no di e be ween hem F(2, 22)=2.73, MSE=1.41, p=.09. RPEs o he h ee sub- scales di e ed be ween hem a his poin , F(2, 22)=3.67, MSE=.19, p<.05. Planned compa isons showed a signi ican di e ence be ween he Local and he To al scale, (p=.04), bu no be ween he Local and he Cen al scale, (p=.24). The one be ween he Cen al and he To al scale was e y close o signi icance (p=.053). Mean (SD) RPE alues o he h ee scales we e 19.19 (.37), 19.33 (.26), and 19.47 (.21) o he local, he cen al, and he o al scales, espec i ely. In his case, unlike wha was obse ed in Expe imen 1, i was he o al RPE subscale, bu no he local RPE subscale, he one ha mos s ongly e lec ed pe cei ed exe ion. A exhaus ion, mean (SE) pe cen ages o e he indi idu- ally es ima ed maximum hea a e (compu ed as 220 minus age) o he h ee condi ions we e 92.21% (1.63), 93.45% (1.60), and 93.72% (1.09). The e ec o Feedback/load was a om signi icance (F<1). Again, his ensu es ha pa ici- The e ec o men al wo kload on he in ensi y and emo ional dynamics o pe cei ed exe ion 669 anales de psicología, 2013, ol. 29, nº 3 (oc ub e) pan s we e close o exhaus ion by he end o he inc emen al e o s ages, bu also ha he h ee condi ions we e equa ed a his poin . As expec ed, oxygen consump ion inc eased du ing exe - ion, F(18, 198)=290.11, MSE=6.40, p<.001. A di e ence wi h Expe imen 1, oxygen consump ion inc eased a di e - en a es o di e en Load condi ions, F(36, 396)=2.31, MSE=2.80, p<.0014.5 Mean (SE) VO2 alues, measu ed in ml/kg·min and a - e aged o he 30 seconds p io o exhaus ion we e 50.49 (1.69), 46.48 (3.56), and 48.39 (2.38), o he high-load, he low-load, and he no load condi ions, espec i ely. These alues did no signi ican ly di e among hem (F<1). One eco ding om one pa icipan seemed clea ly anomalous (VO2<25), so we also ca ied ou he same analysis dis e- ga ding ha pa icipan . Mean (SE) VO2 alues, we e hen 50.75 (1.83), 48.44 (3.25), and 47.51 (2.42). The e ec o Cogni i e load emained non-signi ican (F<1). Time o exhaus ion Time o exhaus ion yielded di e ences among he h ee condi ions, bu no in he di ec ion expec ed. Mean (SD) imes o exhaus ion o he Load/nega i e eedback, he Load/Posi i e eedback, and he No Load condi ions we e 885.5 (38.26), 787.5 (33.97), and 752.5 (30.68) seconds, e- spec i ely. In o he wo ds, people in he Load/nega i e eedback condi ion cycled o longe han in he o he wo condi ions [F(2, 22)=22.06; MSE=2586, p<.01] (no e, how- e e , ha his di e ence is no e lec ed by oxygen con- sump ion a es, and does no a ec RPE a ings a he ex- haus ion poin ; see Foo no e 4). RPE du ing inc emen al e o Figu e 5 shows mean RPE sco es o he h ee RPE sub- scales du ing inc emen al e o . All pa icipan s comple ed a leas ou e o le els in he h ee condi ions. A 3 (Load/Feedback: Load-Nega i e eedback, Load-Posi i e eedback, No load) x 5 (Le el: Wa ming up plus ou le els) x 3 (Subscale: L, C, T) wi hin-subjec ANOVA only yielded a signi icance e ec o e o le el, wi h RPE almos linea ly inc easing wi h le el. Mean (SE) RPE alues o he h ee condi ions we e 12.00 (1.33), 12.66 (1.66), and 12.72 (1.13), o he Load/nega i e eedback, he Load/posi i e eed- back, and he No-load condi ions, espec i ely [F(2, 22)=1.99; MSE=14.26, p=.16]. Unlike Expe imen 1, RPEs 45Howe e , oxygen consump ion did no ma ch Load. Ac ually, he condi- ion wi h no load was he one in which he inc ease o oxygen consump ion was ( isually) mos ma ked. Acco ding o LSD pos -hoc compa isons, only he Load/posi i e eedback condi ion signi ican ly de ia ed om he No- load condi ion (oxygen consump ion o his condi ion was sligh ly lowe in measu emen in e als 18 and 19). Following he logic desc ibed abo e, his di e ence migh ha e masked a po en ial RPE di e ence be ween he No- load and he Load/nega i e eedback condi ions du ing moun ing e o . A he exhaus ion poin , howe e , all di e ences disappea , so ha all e ec s a e his poin a e ully eliable. o he h ee subscales did no di e be ween hem [F(2, 22)=1.77; MSE=2.47, p=.19], no in e ac ed wi h Load/Feedback (F<1). Figu e 5. Mean RPE sco es o he h ee RPE subscales du ing inc emen al e o . SAM- alence and SAM-a ousal du ing inc emen al e o Changes in emo ional s a e alence and a ousal, as meas- u ed by he SAM scale a e displayed in Figu e 6. Repo ed alence dec eased almos linea ly wi h e o , e lec ing he in ui ion ha inc emen al e o is el as hedonically nega- i e, F(4, 44)=39.74, MSE=1.46, p<.01. Mean (SE) alence alues we e 7.94 (.48), 7.46 (.48), 6.61 (.38), 5.72 (.38), and 4.83 (.52), o he wa ming-up, and he ou e o le els, e- spec i ely. E o le el did no in e ac wi h Load/Feedback (F<1) Figu e 6. Changes in emo ional alence (do ed line) and a ousal (non- do ed line), du ing exe cise, as measu ed by he SAM scale 670 Da id Cá denas Vélez e . al. anales de psicología, 2013, ol. 29, nº 3 (oc ub e) A ousal nea ly inc eased wi h E o le el. Mean (SE) a ousal alues ac oss le els we e 2.78 (.73), 3.31 (.78), 4.14 (.59), 5.08 (.45), and 6.56 (.37) o he wa ming up, and he 4 e o le els, espec i ely [F(4, 44)=63.39, MSE=1.28, p<.01]. The e ec o Load and he Load x Le el in e ac ion we e e y a om signi icance (bo h F<=1). RPE du ing eco e y Figu e 7 displays mean RPE measu es om exhaus ion poin o he end o eco e y. Apa om he ob ious e ec o Le el (Exhaus ion, Reco e y 1-4) [F(4, 44)=319.59, MSE=7.20, p<.01], wi h RPE sco es g adually declining du - ing eco e y, he e was a s ongly signi ican Load/Feedback x Le el in e ac ion [F(8, 88)=7.38, MSE=2.68, p<.01]. The in e ac ion was due o he di e en a es o eco e y in he di e en Feedback/Load condi ions. Mo e speci ically, i he analysis is es ic ed o he exhaus ion poin and he i s eco e y s age, he e is a signi ican e ec o Load [F(2, 22)=4.51, MSE=5.04, p=.02]. Mean (SE) RPE sco es e- s ic ed o his segmen we e 16.32 (.89), 15.32 (.64), and 15.35 (.37), o he Load/Nega i e eedback, he Load/posi i e eedback, and he No-load condi ions, e- spec i ely. RPE sco es a e la ge o he Load/nega i e eedback condi ion han o he o he wo [F(1, 11)=5.48, MSE=8.28, p=.04], and hus his e ec con ains a signi ican quad a ic componen [F(1, 11)=5.06, MSE=2.64; p=.046]. Figu e 7. Mean RPE sco es om exhaus ion poin o he end o eco e y. I he analysis is ex ended o he second eco e y s age, mean (SE) RPE sco es ( es ic ed o his segmen ) a e 14.34 (1.01), 13.69 (.73), and 13.37 (1.21), o he Load/Nega i e eedback, he Load/Posi i e Feedback, and he No load condi ions, espec i ely. The main e ec o Load emained signi ican [F(2, 22)=3.44; MSE=7.73; p=.05], bu he quad- a ic componen anished (F<1). In o he wo ds, he e is a signi ican e ec o Load/Feedback, mos ly a ibu able o he i s s ages o e- co e y. In con as wi h wha happened in Expe imen 1, his e ec ended o anish as eco e y p og essed. Unlike Expe imen 1, howe e , he e was no e ec o he subscale ype [F<1 o he ma ginal e ec o subscale; mean (SE) alues we e 11.61 (1.23), 11.56 (.97), and 11.57 (.96) o he local, cen al, and o al subscales, espec i ely]. The subscale was e y a om in e ac ing wi h any o he o he ac o s [F(4, 44)=1.80, MSE=.48, p=.15; MSE(8, 88)=1.66, MSE=.26, p=.12; and F(16, 176)=1.16; MSE=.24, p=.30, o he Feedback/Load x Subscale, Le el x Subscale, and he h ee-way in e ac ions, espec i ely]. SAM alence and a ousal du ing eco e y Figu e 8 displays alence and a ousal alues o he h ee Load/ eedback condi ions, om exhaus ion poin o he end o he eco e y in e al (Exhaus ion, Reco e y 1-4). As expec ed, alence became g adually mo e posi i e du ing eco e y [F(4, 44)=45.03, MSE=3.49, p<.01]. Mo e in e es ingly, he e was a s ong main e ec o Load/Feedback [F(2, 22)=7.54, MSE=2.15, p<.01]. Mean (SD) alence alues we e 5.50 (.55), 6.12 (.47), and 6.53 (.37) o he Load/nega i e eedback, he Load/posi i e eed- back, and he No-load condi ions, espec i ely. T end analy- ses showed only he linea componen o be signi ican . In o he wo ds, he Load/Nega i e eedback condi ion elici ed a less posi i e alence han he No-load condi ion, wi h he posi i e eedback condi ion loca ed be ween hem [F(1, 11)=18.98, MSE=1.69, p<.01]. Figu e 8. Valence and a ousal alues o he h ee Load/ eedback condi- ions, om exhaus ion poin o he end o he eco e y in e al. A ousal, on he o he hand, g adually dec eased du ing eco e y [F(4, 44)=31.12, MSE=5.17, p<.01]. Load/ eedback, howe e , did no exe any signi ican e ec on a ousal [F(2, 22)=2.36, MSE=2.35, p=.11], no in e ac ed wi h eco e y le el [F(8, 88)=.13, MSE=1.54, p=.35]. On o he wo ds, ou p ocedu e was e ec i e a in luencing he