Motivational Power of the Happy Face | My Assignment Tutor

brainMotivational Power of the Happy FacesciencesArticleThe Motivational Power of the Happy FaceJana Nikitin 1,* and Alexandra M. Freund 21 Faculty of Psychology, Research Group Personality and Developmental Psychology, University of Basel,4055 Basel, Switzerland2 Department of Psychology, University Research Priority Program Dynamics of Healthy Aging,University of Zurich, 8050 Zurich, Switzerland; freund@psychologie.uzh.ch* Correspondence: jana.nikitin@unibas.ch; Tel.: +41-(0)61-207-0683Received: 28 November 2018; Accepted: 31 December 2018; Published: 7 January 2019Abstract: People who are cheerful have better social relationships. This might be the case becausehappy faces communicate an invitation to interact. Thus, happy faces might have a strongmotivational effect on others. We tested this hypothesis in a set of four studies. Study 1 (N = 94)showed that approach reactions to happy faces are faster than other reactions to happy or angryfaces. Study 2 (N = 99) found the same effect when comparing reactions to happy faces with reactionsto disgusted faces. Supporting the notion that this effect is related to motivation, habitual socialapproach motivation intensified the motivational effect of happy faces (Study 3, N = 82). Finally,Study 4 (N = 40) showed that the reaction-time asymmetry does not hold for categorization taskswithout approach and avoidance movements. These studies demonstrate that happy faces have astrong motivational power. They seem to activate approach reactions more strongly than angry ordisgusted faces activate avoidance reactions.Keywords: emotional faces; approach; avoidance; motivation; reaction times1. IntroductionPeople who smile are rated more favorably by others and interactions with them are expectedto be rewarding [1]. This might be one reason why cheerful people are happier [1], have a lowerdivorce rate [2], and live longer [3] than less cheerful people. Starting with the functional view ofemotions by Darwin [4], it has long been recognized that emotions have an interpersonal function.People who smile express positive social intentions that are essential for the creation and maintenanceof social bonds [5]. Smiling face and its frequent acoustic concomitant, laughter, are associated withbonding, agreement, and affection (for a summary, see Reference [6]). A happy face signals positiveemotions, as well as attachment availability, care, support, and credibility [7–9]. Recently, Tamir andHughes [10] argued that positive social signals such as smiling faces not only serve ultimate goals(e.g., forming strong bonds) but they are also rewarding in and off themselves. According to Tamir andHughes, this proximal value of positive social signals forms the foundation on which the complexitiesof human sociality are built. In line with this hypothesis, Yang and Urminsky [11] demonstrated thatanticipated positive affective reactions of social partners powerfully shape peoples’ behavior. In otherwords, people are highly motivated to evoke smile and happiness in others. In Yang and Urminsky’sstudies, participants decided which gift they gave to a person. Participants often chose to forgosatisfaction-maximizing gifts in favor of gifts that maximized anticipated positive emotional reactionsin the perceiver. These results confirmed the author’s “smile-seeking” hypothesis that peoples’ desireto induce, approach, and enjoy others’ spontaneous displays of affective reactions is innately valuableand rewarding, and consequently, a powerful motivator in social contexts (see also References [12–15]).Similar conclusions come from neuropsychological research that link the reward areas of the basalganglia with the perception of smiling faces [16]. The reward circuitry in the basal ganglia, in turn,Brain Sci. 2019, 9, 6; doi:10.3390/brainsci9010006 www.mdpi.com/journal/brainsciBrain Sci. 2019, 9, 6 2 of 18is associated with what Davidson [17] termed pre-goal attainment positive affect. In Davidson’s view,activation in subcortical reward centers supports an organism’s approach toward an appetitive goal.It seems that the activation of the reward system by a perceived smile serves appetitive behavior. Peoplewho smile show signals of prosocial intentions [18]. In most—if not all—cases, approaching positivesocial stimuli is adaptive [19]. Given that approaching a happy person implies approaching safety andthe possibility for affiliation, others might follow this invitation. In other words, a smiling face shouldmotivate others to approach [20–22]. In fact, previous studies have demonstrated that people showapproach behavioral tendencies towards smiling faces [20,21,23–25]. Although it has been arguedrepeatedly that these approach tendencies to happy faces have motivational underpinnings [10,15],to our knowledge the motivation hypothesis has not been directly tested. The present research aims atproviding support for the motivation hypothesis by comparing the motivational effect of happy facesto facial expressions of negative emotions (anger, disgust). We assumed that approach behavior tohappy faces is the default reaction.Approaching positive social stimuli such as happy faces should be generally adaptive. In contrast,it is less clear what the most adaptive reaction to negative social stimuli might be. For example, facesexpressing anger communicate aggressive tendencies, and thus, possible threat [26,27]. In some casesof encountering anger, approach behavior might be most appropriate (e.g., when fighting againstan aggressor is important for one’s social standing), but in other cases avoidance behavior mightbe more adaptive (e.g., when a fight might escalate aggression and likely leads to social or physicallosses). Thus, angry facial expressions might not automatically elicit approach or avoidance behavior.Mineka [28] suggested that behavioral responses to (social) threat are slowed down in order to avoiderroneous decisions that could have detrimental consequences for the individual. Given the difficultiesin deciding how to react best when confronted with anger, slowing down the reaction might give theperson time to select the best solution in a given situation. In other words, there might be no cleardefault reaction to the display of anger. Indirect support for this assumption comes from a studyby Roelofs, Hagenaars, and Stins [29]. Using a stabilometric force platform assessing the amount ofspontaneous body sway, Roelofs and colleagues found reduced behavioral reaction when participantsviewed angry compared to neutral and happy faces (for similar results, see Reference [25]). In addition,a recent meta-analysis [23] of 29 studies with 81 effects sizes found small- to medium-sized effects forthe compatibility of different positive and negative stimuli (faces, pictures, and words) in the approachand avoidance tendencies, respectively [30]. Interestingly, however, there was a tendency of facialexpressions to be less effective in initiating approach or avoidance behavior than all other stimuli.Particularly with respect to reactions to negative facial expressions such as anger, the results are mixed.Although there is good evidence that angry faces can be detected very quickly [31], they seem to beambiguous with regard to the action tendencies they evoke. Some studies found congruent effects ofangry faces and avoidance (compared to approach) behaviors [20,22,32], but other studies failed todetect any behavioral tendencies in reaction to angry faces [21,25,30,33,34].In summary, we hypothesize that approach of happy faces is the default reaction, whereasthere is no such default reaction to negative facial expressions. We test this hypothesis using avery basic motivated behavioral tendency to approach or avoid [35], measuring reaction times of armmovements [30]. The principle of this reaction-time task is that participants’ response speed to approachor avoid is affected by the compatibility between the response and the valence of the stimuli [36].We start by testing the robustness of approach reactions to happy faces compared to reactions tonegative facial expressions in Studies 1 and 2. We compare reactions towards happy faces with reactionsto faces expressing anger (Study 1) and disgust (Study 2). We expect that participants react fasterwith approach than avoidance to happy faces and that the speed of approach and avoidance reactionsto negative facial expressions does not significantly differ. Studies 3 and 4 were designed to test themotivational hypothesis more directly. Concretely, we test whether habitual approach motivationaccelerates approach reactions to happy faces (Study 3) and we rule out that the reaction-time effectsof happy faces are just categorization effects (Study 4).Brain Sci. 2019, 9, 6 3 of 182. Study 1Study 1 tested the hypothesis that approach reactions to happy faces are faster than avoidancereactions to happy faces and that there is no clear reaction tendency to angry faces. To test thishypothesis, we used a distance-regulation paradigm. In this paradigm, approach and avoidancemovements are associated with moving a manikin representing the self towards or away fromemotional faces (for a similar approach, see Reference [37]). We used the manikin task insteadof using arm flexion/arm extension as often used in the approach-avoidance task (AAT [30,38])because the manikin approach is less ambiguous with respect to the reference point than thearm-flexion/arm-extension approach [32,39]. Arm flexion can be a movement towards oneself(indicating approach behavior) but also away from a stimulus (indicating avoidance behavior).Similarly, arm extension can be a movement away from oneself but also towards a stimulus. In themanikin task, the movements are unambiguously associated to approach and avoidance, respectively.In addition, the manikin task is more sensitive to valence of a stimulus (such as happy and angry faces)than the arm-flexion/arm-extension task [40].2.1. MethodThis research was conducted with the ethical guidelines of the University of Zurich; all studieswere considered exempt from formal ethical review.2.1.1. ParticipantsStudents of the University of Zurich were recruited within a larger project investigating socialbehavior in a transition from parental home to a shared apartment (see References [41–43]). Participantsfirst completed an online questionnaire at home assessing sociodemographic variables and variablesnot relevant for the present research. Later, participants were invited to the lab for the reaction-timepart of the study. They were informed that they would react to pictures of faces of different facialexpressions. The sample consisted of 94 participants (82% female, age M = 23.93, SD = 4.33 years;socio-demographic data of 20 participants could not be assigned because of a wrong personalcode, that means these participants created a different personal code for the online questionnaireassessing the sociodemographic information and for the reaction-time part of the study, so thattheir sociodemographic information could not be assigned to their reaction-time data (Excludingthese participants from the main analysis did not change the results: we found a main effect offacial expression (F(1, 72) = 53.20, p < 0.001, ηp2 = 0.43), a main effect of movement direction(F(1, 22) = 26.87, p < 0.001, ηp2 = 0.27), and an expression × movement interaction, (F(1, 72) = 62.62,p < 0.001, ηp2 = 0.47). No other effects were statistically significant (ps ≥ 0.12).). The majority ofthe participants (55.5%) was in a stable relationship or married, 44.6% were single. We includedonly participants with good knowledge of the German language. There were no other inclusion orexclusion criteria. Participants were recruited via emails sent from the university administration,via announcements on a university-wide job portal, and via flyers that were distributed all over theuniversity. Participants gave written informed consent for participation. After participation, theywere debriefed (i.e., they received a written document shortly explaining the research questions andhypotheses of the study) and received 20 Swiss francs (approximately 17 €) or extra course credit.2.1.2. Stimuli and ProcedureFacial stimuli were chosen from the Lifespan Database of Adult Emotional Facial Stimuli [44].We selected pictures of 110 models (27 young males, 28 young females, 28 middle-aged males, and 27middle-aged females). One picture of each model clearly expressed happiness or anger, respectively.In order to delete peripheral information (such as hairdo, ears, neck) and to standardize the differentmodels as much as possible, pictures were cut vertically from the hairline to the chin and horizontallyat the cheekbones. Consequently, the picture width and length varied from 3.95” to 4.70”. For theBrain Sci. 2019, 9, 6 4 of 18highest possible uniformity, all pictures were gray scaled. We used the program DirectRT [45] forstimulus presentation, timing, and data collection.An experimental trial started with a blank screen presented for 100 ms and was followed by apresentation of a small white manikin on a gray background in the middle of the screen (the procedurefor an exemplary trial is shown in Figure 1). Participants were told that the manikin represents them,indicated by the superscription “I” (in German “Ich”) over its head. After another 500 ms, a picture ofa happy or an angry face appeared randomly on the left or the right of the manikin. In half of the trials,the task was to move the manikin as fast as possible towards happy face and away from angry face(congruent trials). In the other half of the trials, the movement was reversed (incongruent trials). Thetrials were organized in four blocks, with alternating congruent and incongruent trials. Approximatelyhalf of the participants (n = 45) started with the congruent, and half (n = 49) with the incongruentcondition. Half of the happy faces appeared on the left side, half of them on the right side. Half ofthem appeared in the congruent, half in the incongruent condition. The same was true for angry faces.The order of the presentation of the stimuli and their placement were randomized across participants.For the movements, participants used a joystick that was fixated on a board in the middle of the table.The manikin moved synchronously with the joystick movement.Brain Sci. 2019, 9, 6 4 of 18An experimental trial started with a blank screen presented for 100 ms and was followed by apresentation of a small white manikin on a gray background in the middle of the screen (theprocedure for an exemplary trial is shown in Figure 1). Participants were told that the manikinrepresents them, indicated by the superscription “I” (in German “Ich”) over its head. After another500 ms, a picture of a happy or an angry face appeared randomly on the left or the right of themanikin. In half of the trials, the task was to move the manikin as fast as possible towards happy faceand away from angry face (congruent trials). In the other half of the trials, the movement wasreversed (incongruent trials). The trials were organized in four blocks, with alternating congruentand incongruent trials. Approximately half of the participants (n = 45) started with the congruent,and half (n = 49) with the incongruent condition. Half of the happy faces appeared on the left side,half of them on the right side. Half of them appeared in the congruent, half in the incongruentcondition. The same was true for angry faces. The order of the presentation of the stimuli and theirplacement were randomized across participants. For the movements, participants used a joystick thatwas fixated on a board in the middle of the table. The manikin moved synchronously with the joystickmovement.Figure 1. Exemplary schematic presentation of a congruent trial (“Ich” = “I”) from Study 1, 2, and 3.2.2. Results and DicussionFrom a total of 20,680 trials, we excluded trials with incorrect responses, trials with reactiontimes above 1500 ms or below 200 ms, as well as responses that were more than two standarddeviations above or below each participant’s mean response latency. Overall, 15.1% trials wereexcluded. There were 17,561 trials remaining in the data file. The amount of the remaining trials didnot differ significantly between the four conditions, χ2(3, 17,561) = 1.28, p = 0.73. We ran all analyseswith log-transformed reaction times to correct for skewness in the distribution.We ran a 2 × 2 × 2 mixed-design ANOVA with facial expression (happy vs. angry) and movementdirection (approach vs. avoidance) as within-participants factors and block order (congruent first vs.incongruent first) as a between-participants factor. The analysis revealed a main effect of facialexpression (F(1, 92) = 80.71, p < 0.001, ηp2 = 0.47) and a main effect of movement direction (F(1, 92) =71.87, p < 0.001, ηp2 = 0.44). People reacted faster to happy (M = 673.60 ms, SD = 102.76 ms) than toangry faces (M = 699.99 ms, SD = 101.23 ms) and faster with approach (M = 675.72 ms, SD = 102.03ms) than with avoidance (M = 697.86 ms, SD = 101.58 ms). However, and in line with the hypotheses,this main effect was qualified by a significant expression × movement interaction, F(1, 92) = 37.57, pFigure 1. Exemplary schematic presentation of a congruent trial (“Ich” = “I”) from Study 1, 2, and 3.2.2. Results and DicussionFrom a total of 20,680 trials, we excluded trials with incorrect responses, trials with reaction timesabove 1500 ms or below 200 ms, as well as responses that were more than two standard deviationsabove or below each participant’s mean response latency. Overall, 15.1% trials were excluded. Therewere 17,561 trials remaining in the data file. The amount of the remaining trials did not differsignificantly between the four conditions, χ2(3, 17,561) = 1.28, p = 0.73. We ran all analyses withlog-transformed reaction times to correct for skewness in the distribution.We ran a 2 × 2 × 2 mixed-design ANOVA with facial expression (happy vs. angry) and movementdirection (approach vs. avoidance) as within-participants factors and block order (congruent firstvs. incongruent first) as a between-participants factor. The analysis revealed a main effect offacial expression (F(1, 92) = 80.71, p < 0.001, ηp2 = 0.47) and a main effect of movement direction(F(1, 92) = 71.87, p < 0.001, ηp2 = 0.44). People reacted faster to happy (M = 673.60 ms, SD = 102.76 ms)than to angry faces (M = 699.99 ms, SD = 101.23 ms) and faster with approach (M = 675.72 ms,Brain Sci. 2019, 9, 6 5 of 18SD = 102.03 ms) than with avoidance (M = 697.86 ms, SD = 101.58 ms). However, and in line withthe hypotheses, this main effect was qualified by a significant expression × movement interaction,F(1, 92) = 37.57, p < 0.001, ηp2 = 0.29 (There was neither a main effect of gender on the reaction times(p = 0.42), nor did gender moderate the effect of facial expression (p = 0.27), movement (p = 0.65),condition (p = 0.25), or the expression × movement interaction (p = 0.13) on the reaction times).As shown in Figure 2, approach reactions to happy faces were faster than avoidance reactions to happyfaces (t(93) = -9.58, p < 0.001, d = 1.99), they were faster than avoidance reactions to angry faces(t(93) = -11.12, p < 0.001, d = 2.31), and also faster than approach reactions to angry faces (t(98) = -9.17,p < 0.001, d = 1.90). Avoidance reactions to happy faces were not significantly different from avoidancereactions to angry faces (t(93) = 1.46, p = 0.15) and also not different from approach reactions toangry faces (t(93) = -1.05, p = 0.30). Avoidance reactions to angry faces were marginally faster thanapproach reactions to angry faces (t(93) = -1.90, p = 0.06, d = 0.39). The main effect of block order(F(1, 92) = 2.82, p = 0.10) as well as the interaction between block order, facial expression, and movementwere not statistically significant (F < 1).In sum, then, results of Study 1 supported the hypothesis of faster approach reactions to happyfaces compared to all other conditions. When participants were instructed to approach happy faces,they were faster than when they were instructed to avoid happy faces or to approach or avoid angryfaces. These results are in line with the assumption that approaching happy faces is the default—andtherefore the fastest—reaction and support previous findings on clear (default) approach tendenciestowards happy faces [20,21,23–25]. Our results also indicate that there does not seem to be a strongdefault reaction to angry faces, although people are by trend faster when avoiding angry faces thanwhen approaching them. Thus, our results are in line with other studies that failed to find clear(unambiguous) behavioral tendencies to angry faces [21,25,30,33,34].Brain Sci. 2019, 9, 6 5 of 18< 0.001, ηp2 = 0.29 (There was neither a main effect of gender on the reaction times (p = 0.42), nor didgender moderate the effect of facial expression (p = 0.27), movement (p = 0.65), condition (p = 0.25),or the expression × movement interaction (p = 0.13) on the reaction times). As shown in Figure 2,approach reactions to happy faces were faster than avoidance reactions to happy faces (t(93) = -9.58,p < 0.001, d = 1.99), they were faster than avoidance reactions to angry faces (t(93) = -11.12, p < 0.001,d = 2.31), and also faster than approach reactions to angry faces (t(98) = -9.17, p < 0.001, d = 1.90).Avoidance reactions to happy faces were not significantly different from avoidance reactions to angryfaces (t(93) = 1.46, p = 0.15) and also not different from approach reactions to angry faces (t(93) = -1.05,p = 0.30). Avoidance reactions to angry faces were marginally faster than approach reactions to angryfaces (t(93) = -1.90, p = 0.06, d = 0.39). The main effect of block order (F(1, 92) = 2.82, p = 0.10) as wellas the interaction between block order, facial expression, and movement were not statisticallysignificant (F < 1).In sum, then, results of Study 1 supported the hypothesis of faster approach reactions to happyfaces compared to all other conditions. When participants were instructed to approach happy faces,they were faster than when they were instructed to avoid happy faces or to approach or avoid angryfaces. These results are in line with the assumption that approaching happy faces is the default––andtherefore the fastest––reaction and support previous findings on clear (default) approach tendenciestowards happy faces [20,21,23–25]. Our results also indicate that there does not seem to be a strongdefault reaction to angry faces, although people are by trend faster when avoiding angry faces thanwhen approaching them. Thus, our results are in line with other studies that failed to find clear(unambiguous) behavioral tendencies to angry faces [21,25,30,33,34].Figure 2. Mean reaction times (RTs) to happy and angry faces from Study 1. Error bars representstandard error of the mean.3. Study 2Study 2 aimed at replicating the results of Study 1. In addition, Study 2 used different negativefacial expressions to rule out the possibility that the reaction-time effects of Study 1 are driven by thecomparison of happy and angry faces. We chose faces expressing disgust. Different to angry faces,disgusted faces do not express a direct threat towards the perceiver; however, they express a clearlynegative emotion [46]. Thus, we expect that––similarly to angry faces––there is no default reaction tofaces expressing disgust. In line with this hypothesis, Seidel and colleagues [21] found no significanttendency to approach or avoid faces expressing disgust.620640660680700720740760Happy AngryRT (ms)Approach AvoidanceFigure 2. Mean reaction times (RTs) to happy and angry faces from Study 1. Error bars representstandard error of the mean.3. Study 2Study 2 aimed at replicating the results of Study 1. In addition, Study 2 used different negativefacial expressions to rule out the possibility that the reaction-time effects of Study 1 are driven by thecomparison of happy and angry faces. We chose faces expressing disgust. Different to angry faces,disgusted faces do not express a direct threat towards the perceiver; however, they express a clearlynegative emotion [46]. Thus, we expect that—similarly to angry faces—there is no default reaction tofaces expressing disgust. In line with this hypothesis, Seidel and colleagues [21] found no significanttendency to approach or avoid faces expressing disgust.Brain Sci. 2019, 9, 6 6 of 183.1. Method3.1.1. ParticipantsWe recruited students of the University of Zurich to participate in a lab study entitled “Reactionsto Faces.” Participants first completed an online questionnaire at home assessing sociodemographicvariables and variables not relevant for the present research. Later, participants were invited to the labfor the reaction-time part of the study. They were informed that they would react to pictures of facesof different facial expressions. The sample consisted of 99 participants (77% female; age M = 24.00,SD = 3.00 years; socio-demographic data of three participants could not be assigned because of awrong personal code). The majority of the participants (67.7%) were single, 32.2% was in a stablerelationship or married. We included only participants with good knowledge of the German language.There were no other inclusion or exclusion criteria. Participants were recruited via announcements ona university-wide job portal and via flyers that were distributed all over the university. Participantsgave written informed consent for participation. After participation, they were debriefed (i.e., theyreceived a written document shortly explaining the research questions and hypotheses of the study)and received 20 Swiss francs (approximately 17 €) or extra course credit.3.1.2. Stimuli and ProcedureWe selected pictures of the same 110 young models as in Study 1. One picture of each modelclearly expressed happiness or disgust, respectively. The procedure was the same as in Study 1.Approximately half of the participants (n = 49) started with the congruent, half (n = 50) with theincongruent condition.3.2. Results and DicussionFrom a total of 21,780 facial-expression trials, we excluded 8.7% trials using the same criteria as inStudy 1. There were 19,888 trials remaining in the data file. The amount of the excluded trials did notdiffer significantly between the four conditions, χ2(3, N = 21,780) = 2.06, p = 0.56.We ran a 2 × 2 × 2 mixed-design ANOVA with facial expression (happy vs. disgusted) andmovement direction (approach vs. avoidance) as within-participants factors and block order (congruentfirst vs. incongruent first) as a between-participants factor. The analysis revealed a main effect offacial expression (F(1, 97) = 5.19, p = 0.03, ηp2 = 0.05) and a main effect of movement direction(F(1, 97) = 54.69, p < 0.001, ηp2 = 0.36). People reacted slightly faster to happy (M = 709.61 ms,SD = 77.44 ms) than to disgusted faces (M = 713.63 ms, SD = 74.53 ms) and faster with approach(M = 694.0 ms, SD = 75.38 ms) than with avoidance (M = 729.24 ms, SD = 77.30 ms). However, and inline with the hypotheses, this main effect was qualified by a significant expression × movementinteraction (F(1, 97) = 151.39, p < 0.001, ηp2 = 0.61). As shown in Figure 3, approach reactions to happyfaces were faster than avoidance reactions to happy faces (t(98) = -13.07, p < 0.001, d = 2.64), they werefaster than avoidance reactions to disgusted faces (t(98) = -7.06, p < 0.001, d = 1.43) and also fasterthan approach reactions to disgusted faces (t(98) = -10.21, p < 0.001, d = 2.06). In contrast, avoidancereactions to happy faces were slower than avoidance reactions to disgusted faces (t(98) = 5.28, p < 0.001,d = 1.07) and also slower than approach reactions to disgusted faces (t(98) = 8.44, p < 0.001, d = 1.71).Approach and avoidance reactions to disgusted faces did not significantly differ, t(98) = 1.51, p = 0.14.The main effect of block order (F < 1) as well as the interaction between block order, facial expression,and movement were not statistically significant, F(1, 97) = 1.04, p = 0.31. (Although there was a maineffect of gender on the reaction times (F(1, 92) = 4.13, p = 0.045), suggesting that males reacted faster(M = 684.33, SD = 65.21) than females (M = 720.34, SD = 75.30), gender did not moderate the effect offacial expression (p = 0.58), movement (p = 0.68), condition (p = 0.73), or the expression × movementinteraction (p = 0.71) on the reaction times.)Brain Sci. Brain Sci.2019 2019,,99, 6 , 6 7 of 18 7 of 18Figure 3. Mean reaction times (RTs) to happy and disgusted faces from Study 2. Error bars representstandard error of the mean.In sum, then, Study 2 replicated the finding of Study 1 that approach reactions to happy facesare faster than any other reaction. Importantly, Study 2 showed that this effect is not limited to thecomparison of happy and angry faces but extends to the comparison of happy and disgusted faces.(To test whether approach and avoidance reactions to emotional faces differed between Study 1 and2, we computed difference scores between the log-transformed reaction times of avoidance reactionsminus approach reactions to happy, angry, and disgusted faces. A mixed-design ANOVA testedwhether the difference scores differed dependent on the interaction between Valence (positive vs.negative facial expression) and Study (Study 1 vs. Study 2). There was a main effect of Study (F(1,191)= 7.12, p = 0.008, ηp2 = 0.04) and a main effect of Valence (F(1,191) = 83.70, p < 0.001, ηp2 = 0.04). Thedifference between approach and avoidance movements was generally larger in Study 2 (M = 35.24ms, SD = 29.58 ms) than in Study 1 (M = 22.14 ms, SD = 28.72 ms) and the difference was generallylarger for happy (M = 64.09 ms, SD = 61.26 ms) than for angry/disgusted faces (M = -6.37 ms, SD =66.76 ms). Importantly, there was no significant valence × study interaction, F(1,191) = 1.91, p = 0.17,suggesting that the pattern of responses to happy vs. angry and happy vs. disgusted faces iscomparable in both studies.)4. Study 3Study 3 tested the hypothesis that the faster approach reactions to happy faces have motivationalroots. To that end, Study 3 investigated whether habitual social approach motivation, i.e., the generaltendency to approach positive social outcomes [47], intensifies the approach effect of happy faces.This should be the case because habitual social approach motivation energizes and directs behaviortowards positive social information such as smiling faces (for summaries, see References [48,49]). Infact, habitual social approach motivation enhances the exposure to positive social events infriendships [50], active approach behavior in social interactions with strangers [51], and extravertedbehavior in interactions with roommates [41], suggesting appetitive behavioral tendencies towardssocial rewards. As behavioral tendencies of habitual motives are assumed to be embodied andautomatic [15,52,53], habitual social approach motivation should be associated with very basicbehavioral approach tendencies towards happy faces as measured in the present research.Study 3 again used happy and angry faces. As Study 1 and 2 demonstrated that the reactiontime differences between reactions to happy and negative facial expressions are driven by fasterapproach reactions to happy faces than other reactions to angry or disgusted faces, Study 3 used onlyapproach movements to happy faces and avoidance movements to angry faces. Our hypothesis was620640660680700720740760Happy DisgustRT (ms)Approach AvoidanceFigure 3. Mean reaction times (RTs) to happy and disgusted faces from Study 2. Error bars representstandard error of the mean.In sum, then, Study 2 replicated the finding of Study 1 that approach reactions to happy facesare faster than any other reaction. Importantly, Study 2 showed that this effect is not limited tothe comparison of happy and angry faces but extends to the comparison of happy and disgustedfaces. (To test whether approach and avoidance reactions to emotional faces differed between Study 1and 2, we computed difference scores between the log-transformed reaction times of avoidancereactions minus approach reactions to happy, angry, and disgusted faces. A mixed-design ANOVAtested whether the difference scores differed dependent on the interaction between Valence (positivevs. negative facial expression) and Study (Study 1 vs. Study 2). There was a main effect of Study(F(1,191) = 7.12, p = 0.008, ηp2 = 0.04) and a main effect of Valence (F(1,191) = 83.70, p < 0.001,ηp2 = 0.04). The difference between approach and avoidance movements was generally larger inStudy 2 (M = 35.24 ms, SD = 29.58 ms) than in Study 1 (M = 22.14 ms, SD = 28.72 ms) and thedifference was generally larger for happy (M = 64.09 ms, SD = 61.26 ms) than for angry/disgustedfaces (M = -6.37 ms, SD = 66.76 ms). Importantly, there was no significant valence × study interaction,F(1,191) = 1.91, p = 0.17, suggesting that the pattern of responses to happy vs. angry and happy vs.disgusted faces is comparable in both studies.)4. Study 3Study 3 tested the hypothesis that the faster approach reactions to happy faces have motivationalroots. To that end, Study 3 investigated whether habitual social approach motivation, i.e., the generaltendency to approach positive social outcomes [47], intensifies the approach effect of happy faces. Thisshould be the case because habitual social approach motivation energizes and directs behavior towardspositive social information such as smiling faces (for summaries, see References [48,49]). In fact, habitualsocial approach motivation enhances the exposure to positive social events in friendships [50], activeapproach behavior in social interactions with strangers [51], and extraverted behavior in interactionswith roommates [41], suggesting appetitive behavioral tendencies towards social rewards. As behavioraltendencies of habitual motives are assumed to be embodied and automatic [15,52,53], habitual socialapproach motivation should be associated with very basic behavioral approach tendencies towardshappy faces as measured in the present research.Study 3 again used happy and angry faces. As Study 1 and 2 demonstrated that the reaction-timedifferences between reactions to happy and negative facial expressions are driven by faster approachreactions to happy faces than other reactions to angry or disgusted faces, Study 3 used only approachmovements to happy faces and avoidance movements to angry faces. Our hypothesis was that theBrain Sci. 2019, 9, 6 8 of 18higher the habitual social approach motivation is, the faster are approach reactions to happy faces.As habitual approach motivation regulates responses to positive but not to negative social information(e.g., [54]), we did not expect habitual approach motivation to be associated with reaction times toangry faces.4.1. Method4.1.1. ParticipantsAs in Study 2, we recruited students of the University of Zurich to participate in a lab studyentitled “Reactions to Faces.” Participants first completed an online questionnaire at home assessingsociodemographic variables, habitual social approach motivation, and variables not relevant for thepresent research. Later, participants were invited to the lab for the reaction-time part of the study.They were informed that they would react to pictures of faces of different facial expressions. Thesample consisted of 82 participants (73% females, age M = 26.77, SD = 6.29 years). Approximatelyhalf of the participants (48.8%) were single, 50.0% was in a stable relationship or married, and oneperson was divorced. We included only participants with good knowledge of the German language.There were no other inclusion or exclusion criteria. Participants were recruited via announcementson a university-wide job portal and via flyers that were distributed all over the university. The studywas run in the laboratories of the University of Zurich. Participants gave written informed consentfor participation. After participation, they were debriefed (i.e., they received a written documentshortly explaining the research questions and hypotheses of the study) and received 20 Swiss francs(approximately 17 €) or extra course credit.4.1.2. Stimuli and ProcedureWe used the same stimuli and the identical procedure as in Study 1 but included only thehappy-approach and angry-avoidance conditions.4.1.3. Assessment of Habitual Social Approach MotivationThe Affiliation Tendency Scale [55], German version in Reference [56], was used to assess habitualsocial approach motivation. This scale consists of 25 self-descriptive items (e.g., “I like to make asmany friends as I can”). Responses were given on a rating scale ranging from 0 (strongly disagree) to 6(strongly agree). The internal consistency of the scale was Cronbach’s a = 0.74 (M = 3.88, SD = 0.55).4.2. Results and DicussionFrom a total of 18,040 trials, we excluded 5.96% of the trials using the same exclusion criteria as inStudy 1 and 2. There were 16,966 trials remaining in the data file. The number of remaining trials didnot significantly differ between the two conditions, χ2(1, N = 16,966) = 0.44, p = 0.51. All analyses wererun with log-transformed reaction times to correct for skewness in the distribution.The average reaction times to happy faces (M = 625.62 ms, SD = 87.13 ms) were significantly fasterthan the average reaction times to angry faces (M = 652.64 ms, SD = 82.78 ms), t(81) = -9.25, p < 0.001,d = 2.06. Bivariate correlations of reaction times and habitual social approach motivation supportedthe motivational hypothesis: The higher the habitual social approach motivation was, the faster wereapproach reactions to happy faces (r = -0.28, p = 0.01). The correlation between social motivation andavoidance reactions to angry faces was not statistically significant (r = -0.14, p = 0.22). Importantly,social approach motivation was correlated with the difference in reaction times to happy and angryfaces (r = -0.31, p = 0.004). The higher the social approach motivation was, the faster participantsreacted to happy compared to angry faces (see Figure 4). (There was neither a main effect of gender onthe reaction times (p = 0.18), nor did gender moderate the effect of facial expression (p = 0.52) on thereaction times. There were also no significant moderation effects of gender on the associations betweenBrain Sci. 2019, 9, 6 9 of 18habitual approach motivation and the reaction times to happy and angry faces and their difference(all ps ≥ 0.16).)Taken together, Study 3 supported the hypothesis that habitual social motivation acceleratesapproach reactions to happy faces. Happy faces seem to have a particular motivational power forpeople who are generally highly motivated to approach positive social outcomes, supporting themotivational explanation of approach reactions to happy faces.Brain Sci. 2019, 9, 6 9 of 18associations between habitual approach motivation and the reaction times to happy and angry facesand their difference (all ps ≥ 0.16).)Taken together, Study 3 supported the hypothesis that habitual social motivation acceleratesapproach reactions to happy faces. Happy faces seem to have a particular motivational power forpeople who are generally highly motivated to approach positive social outcomes, supporting themotivational explanation of approach reactions to happy faces.Figure 4. Bivariate correlation between habitual social approach motivation and the reaction-timedifference between approach reactions to happy and avoidance reactions to angry faces (Study 3).Positive values represent slower reactions to happy than to angry faces, negative values representfaster reactions to happy than to angry faces.5. Study 4Study 4 was conducted to ensure that the reaction-time effects of happy faces are not justcategorization effects but are motivational in nature. In other words, we wanted to rule out thatparticipants react faster to happy compared to angry faces because happy faces are faster recognizedor processed [57]. This would be the case when participants would react faster to happy than to angryfaces even when they do not execute any approach behavior. To test this possibility, Study 4 used thesame stimuli as Study 1 and 3 (i.e., happy and angry faces) but instead of reacting with approach andavoidance, the task was to categorize the facial expressions by pressing a response key. We expectedno significant differences in the reaction times between reactions to happy and angry faces as thesereactions were not associated with approach or avoidance behavior. In addition, we tested whetherhabitual social approach motivation affects the reaction times. We expected that because the task isnot related to approach reactions, habitual social approach motivation would not be associated withthe reaction times.5.1. Method5.1.1. ParticipantsWe recruited students of the University of Zurich to participate in a lab study entitled “Reactionsto Faces.” Participants first completed an online questionnaire at home assessing sociodemographicvariables, habitual social approach motivation, and variables not relevant for the present research.Later, participants were invited to the lab for the reaction-time part of the study. They were informedthat they would react to pictures of faces of different facial expressions. The sample consisted of N =40 young adults (73% female, age M = 25.30, SD = 4.42 years). The majority of the participants (62.5%)were single, 37.5% was in a stable relationship or married. We included only participants with good-0.25-0.2-0.15-0.1-0.0500.050.10.153 3.5 4 4.5 5 5.5 6 6.5RT to happy – angry faces (logtransformed)Dispositional social approach motivationr = -0.31, p = 0.004Figure 4. Bivariate correlation between habitual social approach motivation and the reaction-timedifference between approach reactions to happy and avoidance reactions to angry faces (Study 3).Positive values represent slower reactions to happy than to angry faces, negative values representfaster reactions to happy than to angry faces.5. Study 4Study 4 was conducted to ensure that the reaction-time effects of happy faces are not justcategorization effects but are motivational in nature. In other words, we wanted to rule out thatparticipants react faster to happy compared to angry faces because happy faces are faster recognizedor processed [57]. This would be the case when participants would react faster to happy than to angryfaces even when they do not execute any approach behavior. To test this possibility, Study 4 used thesame stimuli as Study 1 and 3 (i.e., happy and angry faces) but instead of reacting with approach andavoidance, the task was to categorize the facial expressions by pressing a response key. We expectedno significant differences in the reaction times between reactions to happy and angry faces as thesereactions were not associated with approach or avoidance behavior. In addition, we tested whetherhabitual social approach motivation affects the reaction times. We expected that because the task is notrelated to approach reactions, habitual social approach motivation would not be associated with thereaction times.5.1. Method5.1.1. ParticipantsWe recruited students of the University of Zurich to participate in a lab study entitled “Reactionsto Faces.” Participants first completed an online questionnaire at home assessing sociodemographicvariables, habitual social approach motivation, and variables not relevant for the present research.Later, participants were invited to the lab for the reaction-time part of the study. They were informedthat they would react to pictures of faces of different facial expressions. The sample consisted ofN = 40 young adults (73% female, age M = 25.30, SD = 4.42 years). The majority of the participantsBrain Sci. 2019, 9, 6 10 of 18(62.5%) were single, 37.5% was in a stable relationship or married. We included only participantswith good knowledge of the German language. There were no other inclusion or exclusion criteria.Participants were recruited via announcements on a university-wide job portal and via flyers thatwere distributed all over the university. Participants gave written informed consent for participation.After participation, they were debriefed and received 20 Swiss francs (approximately 17 €) or extracourse credit.5.1.2. Stimuli and ProcedureWe used the same stimuli as in Study 1 and 3. An experimental trial started with a blankscreen presented for 100 ms and was followed by a presentation of a picture of a happy or an angryface in the middle of the screen. The task was to press a choice button with the left or right indexfinger, respectively. Index fingers were placed on the buttons throughout the experiment. Half of theparticipants (n = 20) pressed the left key for happy and the right key for angry faces, the other half(n = 20) responded in the reversed way.5.1.3. Assessment of Habitual Social Approach MotivationThe Affiliation Tendency Scale was again used to assess habitual social approach motivation. Theinternal consistency of the scale was Cronbach’s a = 0.74 (M = 3.71, SD = 0.57).5.2. Results and DicussionFrom a total of 8800 facial-expression trials, we excluded 4.2% of the trials using the same exclusioncriteria as in the previous studies. There were 8428 trials remaining in the data file. The amount ofremaining trials in the two conditions did not significantly differ, χ2(1, N = 8428) = 0.01, p = 0.91. Allanalyses were run with log-transformed reaction times to correct for skewness in the distribution.We ran a mixed-design ANOVA with facial expression (happy vs. angry) as a within-participantsfactor and the response-key allocation (positive left/negative right vs. positive right/negative left) asa between-participants factor. There was no main effect of facial expression (F < 1). Reaction timesto happy faces (M = 495.94 ms, SD = 74.46 ms) and reaction times to angry faces (M = 492.97 ms,SD = 74.61 ms) did not differ significantly. There was also no effect of the response-key allocation (F < 1)or an interaction effect of facial expression and response-key allocation, F(1, 38) = 1.55, p = 0.22. Inaddition, habitual social approach motivation did not correlate with reaction times to happy (r = -0.01,p = 0.97) or angry faces (r = -0.06, p = 0.71). There was also no significant correlation between socialapproach motivation and the difference in the reactions to happy and angry faces (r = 0.14, p = 0.38).(There was neither a main effect of gender on the reaction times (p = 0.17), nor did gender moderate theeffect of facial expression (p = 0.66), response-key allocation (p = 0.18), or their interaction (p = 0.91) onthe reaction times. There were also no significant moderation effects (conducted with the programmPROCESS, 91) of gender on the associations between habitual approach motivation and the reactiontimes to happy and angry faces and their difference (all ps ≥ 0.21).)Study 4 further supported the notion that faster reactions to happy faces have motivationalroots. Categorization of emotional faces without approach and avoidance movements did not revealthe reaction-time differences that we observed in Studies 1–3. In addition, habitual social approachmotivation was not associated with the reaction times in Study 4. We maintain that this is thecase because pressing a response key does not reflect an approach of the happy face. (Because it isproblematic to accept a Null-hypothesis, we contrasted results of Study 4 with the previous threestudies. To this end, we calculated difference scores between the log-transformed reaction times toangry/disgusted minus happy faces for all four studies (we used only the approach movements tohappy and avoidance movements to angry/disgusted faces from Study 1 and 2) and compared themin a univariate ANOVA with Study (1–4) as a between-participants factor. There was a significanteffect of the factor Study, F(3,311) = 19.63, p < 0.001, ηp2 = 0.04. The post-hoc tests (LSD) revealedsignificant differences between Study 4 and all other three studies (all ps < 0.001). No other differencesBrain Sci. 2019, 9, 6 11 of 18were statistically significant (all ps > 0.05). Thus, the differences between happy and angry/disgustedfaces in Studies 1, 2, and 3 cannot be interpreted as a result of simple categorization.)6. General DiscussionSince Harlow’s [58] classic studies and the advent of attachment theory [59,60], belongingness hasbeen acknowledged as one of the most fundamental human motives [61]. The present studies appliedthis motivational hypothesis to study the reactions to emotional faces. Study 1 and 2 supported therobustness of approach reactions to happy faces. Approach reactions to happy faces were faster thanany other reaction to happy, angry, or disgusted faces. Study 3 demonstrated that habitual socialapproach motivation intensifies the motivational power of happy faces. Importantly, results of Study4 indicate that this effect might be specific to approach reactions. Simple categorization withoutapproach did not reveal the happy-face advantage in reaction times, supporting the notion that theeffect is driven by motivational factors. In the following, we discuss the results of the present studiesand their theoretical and practical implications.6.1. Happy FacesIn summary, people react very fast to happy faces with approach behavior. This is not onlythe result of the present studies but is supported without exception by previous research thatinvestigated approach and avoidance reactions to happy faces [23]. This might be the case becauseapproaching a happy person might often have benefits and rarely costs. Humans rely on otherpeople, and therefore, have to approach them. The high survival value of other people is arguablythe reason why belongingness is one of the most fundamental human needs and why we strive forpositive social encounters most of the time [61]. A smile signals positive social intentions [5] andthe possibility to satisfy our desire for belongingness. This might be the reason why a happy facehas a strong motivational power to approach. This conclusion is most strongly supported by Study3 of the present research, that showed a positive association between the strength of habitual socialapproach motivation and speed of approach reactions to happy faces, and Study 4, that showed nosuch differences in a categorization task.Further indirect support for this motivational interpretation comes from studies investigatingpatients with depression. For example, Derntl and colleagues [33] found less amygdala activation(indicating reduced reactivity in response to positive emotional stimuli; [62]) in patients withmajor depression compared to healthy controls during approach movements toward happy faces.Additionally, more pronounced depressive symptoms in the patient group were accompanied bylower levels of amygdala activation. As depressive symptoms are negatively linked to appetitivemotivation [63], these results suggest that people with low levels of appetitive motivation are lesssensitive to positive social signs (for similar results, see Reference [24]). Moreover, research on socialanxiety shows that highly socially anxious persons show avoidance tendencies to smiling faces [30].This might be the case because socially anxious persons do not expect benevolent intentions of theirinteraction partners [64], and thus avoid social interaction partners even when the interaction partnersexpress positive emotions [65].These findings do not only support the motivational explanation of approach of happy faces,they also demonstrate the boundaries of the fundamentality of these reactions. Although we assumethat most people search positive social interactions most of the time, there are also exceptions. Theseexceptions are explained either by individual differences such as low appetitive motivation associatedwith depressive symptoms [24,33] or possibly negative evaluation of happy faces associated withsocial anxiety [30,65]. However, there are also situational factors that reduce approach tendenciesto happy faces, particularly when the smiling face signals dominance, Schadenfreude, or shameand embarrassment [14,66,67]. This might be the case in situations that are hostile or competitive.In support of this hypothesis, Paulus and Wentura [22] found that smile in members of an outgroup(e.g., Middle-Eastern men for White-Caucasian participants) are associated with faster avoidanceBrain Sci. 2019, 9, 6 12 of 18than approach reactions. The authors concluded that a smile of an outgroup member is evaluatedmore probably as dominance, arrogance or Schadenfreude rather than friendliness and cooperation,and thus lead to avoidance reactions (for similar results, see Reference [68]). Future research shouldinvestigate more systematically, in which situations and in whom smiling faces do (not) lead to strongapproach tendencies.6.2. Negative Facial ExpressionsAn interaction with an angry person or a person expressing disgust is ambivalent. If someonesignalizes anger directed to us, we might experience a conflict between the need to belong and the wishto avoid a negative social encounter. This might slow down our reactions. The same might be true fordisgusted faces. Accordingly, studies investigating approach and avoidance tendencies towards angry(and disgusted) faces provide mixed results [21,25,30,33,34]. Even in studies that find faster avoidancethan approach responses to angry faces (e.g., [20]), responses to anger expressions are typically slowerthan responses to other emotional expressions (see also [21]). The present studies support previousfindings indicating that there is no default behavioral reaction to negative facial expressions such asanger or disgust. In contrast to negative words [37,69], negative attitude objects [35], negative auditorystimuli [70], negative affective pictures [71], or fear-related animals such as spiders [38], negative facialexpressions do not seem to elicit a default avoidance reaction. Instead, negative facial expressions,and particularly angry facial expressions, seem to lead to slower reactions, irrespective of approach oravoidance [20,21].This might be the case because angry facial expressions activate neural circuits involved inbehavioral suppression [72], and thus lead to behavioral inhibition in the perceiver [25,29]. Suchinhibition might reflect an orienting response during which we prepare an appropriate reaction.As Phaf and colleagues [23] put it, angry faces seem to require further interpretation to elicit eitherapproach, when evoking anger in the perceiver (e.g., [34]), or avoidance, when evoking fear (e.g., [32].Similarly, Seidel and colleagues [21] argued that disgust in facial expression can signal differentmessages that request either approach (e.g., food-offence disgust) or avoidance (e.g., individual-relateddisgust) [73]. As Winkielman and colleagues [15] discussed, the social environment is complex, and itdemands context-appropriate responses. It is an interesting direction for future research to modulatecontext variables to test directly, which variables elicit approach and which avoidance behavior as areaction to negative facial expressions [22].6.3. Limitations and Future DirectionsOne limitation of the present studies is the artificial setting and task. Looking at still faces at acomputer screen and reacting to them using a joystick bears no resemblance to real-life situationsand carries no real-life relevance. A natural setting would provide more contextual information suchas verbal and non-verbal behavior, the type of relationship, or whether the environment is safe ornot. Such contextual information might be particularly helpful for deciding how to react when beingconfronted with a negative facial expression. Therefore, with more contextual information, the reactionto negative facial expressions might come as readily as the reaction to a happy face. On the other side,the context-free setting emphasizes the motivational power of happy faces. Even when there is nocontextual information, people’s approach reactions to happy faces are faster than any other reactionto happy, angry, or disgusted faces.Another critical point of our studies is that we cannot disentangle the effect of valence from theeffect of arousal. Happy faces are lower in arousal than angry or disgusted faces (e.g., 77). Thus, peoplemight react slower to angry and disgusted faces than to happy faces because angry and disgustedfaces are more arousing. Contradicting this assumption, Robinson and colleagues [74] found a fasterprocessing of highly arousing negative stimuli compared to negative stimuli low in arousal. Similarly,Lang [75] provided evidence that negative and highly arousing stimuli elicit behavioral responseswith a larger amplitude than would be expected only on the basis of their valence. Thus, if arousalBrain Sci. 2019, 9, 6 13 of 18had any effect on the current findings, it should lead to faster reactions to angry or disgusted facesthan to happy faces. In contrast, we find faster reactions to happy faces than to negative emotionalexpressions, speaking for the effect of valence rather than the effect of arousal.Further, the present studies are based on samples with considerably more females than males,which can be partly explained by the fact that the studies were run in the labs of the departmentof psychology and many of the participants were psychology students who are, in majority, female.It is not clear whether the present findings can be generalized to other populations. Note, that wetested possible moderation effects of gender on the present findings and found no evidence for suchmoderation effects. We caution, however, that the male samples were too small to draw reliableconclusions from these additional analyses. In fact, there is some evidence for gender differencesin social motivation: for example, women rate themselves as higher in warmth and empathy thanmen [76] and they are more other-oriented and more cautious than men [77]. This could explainsome findings of the present research (e.g., faster approach reactions to happy faces and slowerreactions to negative facial expressions). Although gender differences both in socially relevant (such asempathy; [78]) as well as general psychological factors [79] are relatively small, the present researchshould be nevertheless replicated in studies with more balanced gender distributions.In the present studies, we did not measure the ambiguity of the reactions to angry faces directlybut tested a possible consequence of this ambiguity, namely longer reaction times to angry as comparedto happy faces. Further studies need to test the link between the difficulty to choose the appropriatereaction facing an angry person and reaction times directly. One possibility is to use the “mousetracking” procedure that allows investigating the process of decision-making [80].Finally, some studies using the categorization task found different results than Study 4 of thepresent research [57,81]. These studies found faster categorization of happy faces than of negativefacial expressions. These results were explained by (1) a general bias of expecting a higher base rate ofpositive over negative events, including facial expressions [57]; (2) a match between one’s (positive)mood [82] and the encountering of a (positive) facial expression in others; and (3) an asymmetrybetween the number of positive (one: smiling face) and negative facial expressions of basic emotions(four: anger, sadness, disgust, fear; [57]). There are differences in the method of previous categorizationstudies and of the current Study 4 that might account for the different findings. First, we used a highnumber of models for the facial expressions so that each model was presented only two (Studies 3and 4: happy and angry facial expression) or four times (Studies 1 and 2: happy and angry/disgustedfacial expression in the approach and the avoidance condition). In contrast, previous categorizationstudies used a small number of models or even schematic facial expressions that were presented manytimes [57,81]. This could have led to learning effects that might be stronger with respect to happythan to negative facial expressions. In addition, previous categorization studies did not compare thefindings from the categorization task with possible results of an approach-avoidance task. Thus, theycannot rule out additional acceleration of responses to happy faces when happy faces are associatedwith approach reactions. Future research is needed to test these possible differences directly.7. ConclusionsThe present research investigated how basic behavioral reactions might help us to understand thefunctionality of social behavior. Happy faces signal attachment availability and people react readily tothis signal. By their motivational power, happy faces help to satisfy the need to belong of both thesmiling person and the perceiver. Thus, an intensive smile seems to have an important interpersonalfunction. The present findings lead to several important conclusions within this framework.First, they show that the reaction-time advantage to happy compared to negative facial expressionsis driven by motivational factors. Although it has been repeatedly found that people react fasterto happy faces than to negative facial expressions, we still know very little about why this isthe case. The present studies illustrate the importance of motivational factors in the happy-facebehavioral advantage.Brain Sci. 2019, 9, 6 14 of 18Second, the present studies indicate that there is no default behavioral reaction to negativefacial expressions such as anger or disgust. In contrast to negative words [37,69], negative attitudeobjects [35], negative auditory stimuli [70], negative affective pictures [71], or fear-related animals suchas spiders [38], negative facial expressions do not seem to elicit a default avoidance reaction (see alsoReference [30]). Instead, negative facial expressions seem to lead to slower reactions, irrespective ofapproach or avoidance. This might be the case because negative facial expressions are associatedwith behavioral inhibition [29]. Such inhibition might reflect an orienting response during which weprepare an appropriate reaction.Third, the present research provides support for the notion that intervention studies thatare based on the approach-avoidance task might train motivation (and not simply associationsbetween stimuli and reactions). Such interventions use for example the adoption of approach-typepostures [83], execution of approach-type movements towards rewarding choices [84] or positivefacial expressions [85], and avoidance-type movements that lead to more controlled (less impulsive)information processing [86]. One of these studies has also demonstrated that socially anxiousparticipants who were trained to approach smiling faces displayed more social approach behaviorsduring a subsequent social interaction compared to participants in the control group [87]. The presentresearch provides support for the motivational explanation of these trainings, and thus contributes tobetter understanding of the mechanisms that underlie such training effects.Author Contributions: The studies were conceptualized by both authors and conducted by J.N. J.N. analyzed thedata. Both authors were involved in writing.Funding: This research was funded by the Swiss National Science Foundation, grant number 100014_126868/1and the Suzanne and Hans Biäsch Foundation for Applied Psychology, grant number 2012/12.Acknowledgments: We wish to thank Angela Ruckstuhl, Julia Becker, Anna Gunsch, and Anaïs Hofmann fortheir help with data collection. Moreover, we appreciate the time and effort the study participants invested inthis research.Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of thestudy; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision topublish the results.References1. Harker, L.; Keltner, D. Expressions of positive emotion in women’s college yearbook pictures and theirrelationship to personality and life outcomes across adulthood. J. Pers. Soc. Psychol. 2001, 80, 112–124.[CrossRef] [PubMed]2. Hertenstein, M.J.; Hansel, C.A.; Butts, A.M.; Hile, S.N. Smile intensity in photographs predicts divorce laterin life. Motiv. Emot. 2009, 33, 99–105. [CrossRef]3. Abel, M.L.; Kruger, E.L. Smile intensity in photographs predicts longevity. Psychol. Sci. 2010, 21, 542–544.[CrossRef] [PubMed]4. Darwin, C. The Expression of Emotions in Man and Animals; University of Chicago Press: Chicago, IL, USA, 1872.5. Fridlund, A.J. The behavioral ecology view of smiling and other facial expressions. In An Empirical Reflectionon the Smile; Abel, M.H., Ed.; Mellen studies in psychology, Vol. 4; Edwin Mellen Press: Lewiston, NY, USA,2002; pp. 45–82. ISBN 0-7734-7002-6.6. Scott, S.K.; Lavan, N.; Chen, S.; McGettigan, C. The social life of laughter. Trends Cogn. Sci. 2014, 18, 618–620.[CrossRef] [PubMed]7. Dewitte, M.; De Houwer, J. Adult attachment and attention to positive and negative emotional faceexpressions. J. Res. Pers. 2008, 42, 498–505. [CrossRef]8. Horstmann, G. What do facial expressions convey: Feeling states, behavioral intentions, or action requests?Emotion 2003, 3, 150–166. [CrossRef]9. Reed, L.I.; Stratton, R.; Rambeas, J.D. Face value and cheap talk: How smiles can increase or cecrease thecredibility of our words. Evol. Psychol. 2018, 16, 147470491881440. [CrossRef]10. Tamir, D.I.; Hughes, B.L. Social rewards: From basic social building blocks to complex social behavior.Perspect. Psychol. Sci. 2018, 13, 700–717. [CrossRef]Brain Sci. 2019, 9, 6 15 of 1811. Yang, A.X.; Urminsky, O. The smile-seeking hypothesis: How immediate affective reactions motivate andreward gift giving. Psychol. Sci. 2018, 29, 1221–1233. [CrossRef]12. Tsukiura, T.; Cabeza, R. Orbitofrontal and hippocampal contributions to memory for face–name associations:The rewarding power of a smile. Neuropsychologia 2008, 46, 2310–2319. [CrossRef]13. Trevarthen, C. Conversations with a two-month-old. In Parent-Infant Psychodynamics: Wild Things, Mirrors andGhosts; Raphael-Leff, J., Ed.; Whurr Publishers: Philadelphia, PA, USA, 2003; pp. 25–34. ISBN 1-86156-346-9.14. Niedenthal, P.M.; Mermillod, M.; Maringer, M.; Hess, U. The Simulation of Smiles (SIMS) model: Embodiedsimulation and the meaning of facial expression. Behav. Brain Sci. 2010, 33, 417–433. [CrossRef] [PubMed]15. Winkielman, P.; Niedenthal, P.; Wielgosz, J.; Eelen, J.; Kavanagh, L.C. Embodiment of cognition and emotion.In APA Handbook of Personality and Social Psychology, Volume 1: Attitudes and Social Cognition; AmericanPsychological Association: Washington, DC, USA, 2015; pp. 151–175.16. Chakrabarti, B.; Bullmore, E.; Baron-Cohen, S. Empathizing with basic emotions: Common and discreteneural substrates. Soc. Neurosci. 2006, 1, 364–384. [CrossRef]17. Davidson, R.J. The neuropsychology of emotion and affective style. In Handbook of Emotions; Lewis, M.,Haviland, J.M., Eds.; Guilford Press: New York, NY, USA, 1993; pp. 143–154. ISBN 0-89862-988-8.18. Becker, D.V.; Srinivasan, N. The vividness of the happy face. Curr. Dir. Psychol. Sci. 2014, 23, 189–194.[CrossRef]19. Bryant, G.A.; Fessler, D.M.T.; Fusaroli, R.; Clint, E.; Amir, D.; Chávez, B.; Denton, K.K.; Díaz, C.; Duran, L.T.;Fan´covi´cová, J.; et al. The perception of spontaneous and volitional laughter across 21 societies. Psychol. Sci.2018, 29, 1515–1525. [CrossRef] [PubMed]20. Marsh, A.A.; Ambady, N.; Kleck, R.E. The effects of fear and anger facial expressions on approach- andavoidance-related behaviors. Emotion 2005, 5, 119–124. [CrossRef] [PubMed]21. Seidel, E.-M.; Habel, U.; Kirschner, M.; Gur, R.C.; Derntl, B. The impact of facial emotional expressionson behavioral tendencies in women and men. J. Exp. Psychol. Hum. Percept. Perform. 2010, 36, 500–507.[CrossRef]22. Paulus, A.; Wentura, D. Threatening joy: Approach and avoidance reactions to emotions are influenced bythe group membership of the expresser. Cogn. Emot. 2014, 28, 656–677. [CrossRef] [PubMed]23. Phaf, R.H.; Mohr, S.E.; Rotteveel, M.; Wicherts, J.M. Approach, avoidance, and affect: A meta-analysis ofapproach-avoidance tendencies in manual reaction time tasks. Front. Psychol. 2014, 5, 378. [CrossRef]24. Radke, S.; Güths, F.; André, J.A.; Müller, B.W.; de Bruijn, E.R.A. In action or inaction? Socialapproach–avoidance tendencies in major depression. Psychiatry Res. 2014, 219, 513–517. [CrossRef] [PubMed]25. Stins, J.F.; Roelofs, K.; Villan, J.; Kooijman, K.; Hagenaars, M.A.; Beek, P.J. Walk to me when I smile, step backwhen I’am angry: Emotional faces modulate whole-body approach‚Äìavoidance behaviors. Exp. Brain Res.2011, 212, 603–611. [CrossRef] [PubMed]26. Adams, R.B.; Gordon, H.L.; Baird, A.A.; Ambady, N.; Kleck, R.E. Effects of gaze on amygdala sensitivity toanger and fear faces. Science 2003, 300, 1536. [CrossRef] [PubMed]27. Wicker, B.; Perrett, D.I.; Baron-Cohen, S.; Decety, J. Being the target of another’s emotion: A PET study.Neuropsychologia 2003, 41, 139–146. [CrossRef]28. Mineka, S. Evolutionary memories, emotional processing, and the emotional disorders. Psychol. Learn. Motiv.1992, 28, 161–206. [CrossRef]29. Roelofs, K.; Hagenaars, M.A.; Stins, J. Facing freeze: Social threat induces bodily freeze in humans.Psychol. Sci. 2010, 21, 1575–1581. [CrossRef] [PubMed]30. Heuer, K.; Rinck, M.; Becker, E.S. Avoidance of emotional facial expressions in social anxiety: TheApproach-Avoidance Task. Behav. Res. Ther. 2007, 45, 2990–3001. [CrossRef] [PubMed]31. Eastwood, J.D.; Smilek, D.; Merikle, P.M. Differential attentional guidance by unattended faces expressingpositive and negative emotion. Percept. Psychophys. 2001, 63, 1004–1013. [CrossRef]32. Rotteveel, M.; Phaf, R. Automatic affective evaluation does not automatically predispose for arm flexion andextension. Emotion 2004, 4, 156–172. [CrossRef] [PubMed]33. Derntl, B.; Seidel, E.-M.; Eickhoff, S.B.; Kellermann, T.; Gur, R.C.; Schneider, F.; Habel, U. Neural correlatesof social approach and withdrawal in patients with major depression. Soc. Neurosci. 2011, 6, 482–501.[CrossRef] [PubMed]34. Wilkowski, B.M.; Meier, B.P. Bring it on: Angry facial expressions potentiate approach-motivated motorbehavior. J. Pers. Soc. Psychol. 2010, 98, 201–210. [CrossRef]Brain Sci. 2019, 9, 6 16 of 1835. Chen, M.; Bargh, J.A. Consequences of automatic evaluation: Immediate behavioral predispositions toapproach or avoid the stimulus. Personal. Soc. Psychol. Bull. 1999, 25, 215–224. [CrossRef]36. De Houwer, J. What are implicit measures and why are we using them? In Handbook of Implicit Cognitionand Addiction; Wiers, R.W., Stacy, A.W., Eds.; Sage Publications: Thousand Oaks, CA, USA, 2006; pp. 11–28.ISBN 1-4129-0974-0.37. De Houwer, J.; Crombez, G.; Baeyens, F.; Hermans, D. On the generality of the affective Simon effect.Cogn. Emot. 2001, 15, 189–206. [CrossRef]38. Rinck, M.; Becker, E.S. Approach and avoidance in fear of spiders. J. Behav. Ther. Exp. Psychiatry 2007, 38,105–120. [CrossRef]39. Eder, A.B.; Rothermund, K. When do motor behaviors (mis)match affective stimuli? An evaluative codingview of approach and avoidance reactions. J. Exp. Psychol. Gen. 2008, 137, 262–281. [CrossRef] [PubMed]40. Krieglmeyer, R.; Deutsch, R. Comparing measures of approach–avoidance behaviour: The manikin task vs.two versions of the joystick task. Cogn. Emot. 2010, 24, 810–828. [CrossRef]41. Nikitin, J.; Burgermeister, L.C.; Freund, A.M. The role of age and social motivation in developmentaltransitions in young and old adulthood. Front. Dev. Psychol. 2012, 3, 366. [CrossRef] [PubMed]42. Nikitin, J.; Freund, A.M. Social motives predict loneliness during a developmental transition. Swiss J. Psychol.2017, 76. [CrossRef]43. Nikitin, J.; Freund, A.M. The indirect nature of social motives: The relation of social approach and avoidancemotives with likeability via Extraversion and Agreeableness. J. Pers. 2015, 83, 97–105. [CrossRef] [PubMed]44. Ebner, N.C.; Riediger, M.; Lindenberger, U. FACES—A database of facial expressions in young, middle-aged,and older women and men: Development and validation. Behav. Res. Methods 2010, 42, 351–362. [CrossRef][PubMed]45. Jarvis, B. DirectRT (Version 2004.3.24); Empirisoft Corporation: New York, NY, USA, 2004.46. Fridlund, A.J. The new ethology of human facial expressions. In The Psychology of Facial Expression;Russell, J.A., Fernández-Dols, J.M., Eds.; Studies in Emotion and Social Interaction, 2nd Series; CambridgeUniversity Press: New York, NY, USA, 1997; pp. 103–129. ISBN 0-521-49667-5.47. Mehrabian, A. Evidence bearing on the Affiliative Tendency (MAFF) and Sensitivity to Rejection (MSR)scales. Curr. Psychol. 1994, 13, 97–116. [CrossRef]48. Gable, S.L.; Berkman, E.T. Making connections and avoiding loneliness: Approach and avoidance socialmotives and goals. In Handbook of Approach and Avoidance Motivation; Elliot, A.J., Ed.; Psychology Press:New York, NY, USA, 2008; pp. 204–216.49. Nikitin, J.; Schoch, S. Social approach and avoidance motivations. In A Handbook of Solitude: PsychologicalPerspectives on Social Isolation, Social Withdrawal, and Being Alone; Coplan, R.J., Bowker, J.C., Eds.;Wiley-Blackwell: West Sussex, UK, 2014; pp. 202–223.50. Gable, S.L. Approach and avoidance social motives and goals. J. Pers. 2006, 74, 175–222. [CrossRef] [PubMed]51. Nikitin, J.; Freund, A.M. When wanting and fearing go together: The effect of co-occurring social approachand avoidance motivation on behavior, affect, and cognition. Eur. J. Soc. Psychol. 2010, 40, 783–804. [CrossRef]52. Bargh, J.A. The automaticity of everyday life. In The Automaticity of Everyday Life: Advances in Social Cognition,Vol 10 Advances in Social Cognition, Vol 10; Wyer, R.S., Jr., Ed.; Lawrence Erlbaum Associates: Mahwah, NJ,USA, 1997; pp. 1–61.53. Shah, J. Automatic for the people: How representations of significant others implicitly affect goal pursuit.J. Personal. Soc. Psychol. 2003, 84, 661–681. [CrossRef]54. Schoch, S.; Nikitin, J.; Freund, A. Why do(n’t) you like me? The role of social approach and avoidancemotives in attributions following social acceptance and rejection. Motiv. Emot. 2015, 39, 1–13. [CrossRef]55. Mehrabian, A. The development and validation of measures of affiliative tendency and sensitivity to rejection.Educ. Psychol. Meas. 1970, 30, 417–428. [CrossRef]56. Sokolowski, K. Kognitionen und Emotionen in Anschlussthematischen Situationen [Cognitions andEmotions in Affiliation-Thematic Situations]. Ph.D. Thesis, Bergische Universität, Wuppertal, Germany, 1986.57. Leppänen, J.M.; Hietanen, J.K. Positive facial expressions are recognized faster than negative facialexpressions, but why? Psychol. Res. 2004, 69, 22–29. [CrossRef]58. Harlow, H.F. The nature of love. Am. Psychol. 1958, 13, 673–685. [CrossRef]Brain Sci. 2019, 9, 6 17 of 1859. Ainsworth, M.D.S.; Bell, S.M.; Stayton, D.J. Infant-mother attachment and social development: Socialisationas a product of reciprocal responsiveness to signals. In The Integration of a Child Into a Social World;Richards, M.J.M., Ed.; Cambridge University Press: London, UK, 1974; pp. 99–135.60. Bowlby, J. Attachment; Basic Books: New York, NY, USA, 1969.61. Baumeister, R.F.; Leary, M.R. The need to belong: Desire for interpersonal attachments as a fundamentalhuman motivation. Psychol. Bull. 1995, 117, 497–529. [CrossRef]62. Suslow, T.; Konrad, C.; Kugel, H.; Rumstadt, D.; Zwitserlood, P.; Schöning, S.; Ohrmann, P.; Bauer, J.;Pyka, M.; Kersting, A.; et al. Automatic mood-congruent amygdala responses to masked facial expressionsin major depression. Biol. Psychiatry 2010, 67, 155–160. [CrossRef]63. Kasch, K.L.; Rottenberg, J.; Arnow, B.A.; Gotlib, I.H. Behavioral activation and inhibition systems and theseverity and course of depression. J. Abnorm. Psychol. 2002, 111, 589–597. [CrossRef]64. Clark, D.M.; Wells, A. A cognitive model of social phobia. In Social Phobia: Diagnosis, Assessment, andTreatment; Guilford Press: New York, NY, USA, 1995; pp. 69–133.65. Mansell, W.; Clark, D.M.; Ehlers, A.; Chen, Y.-P. Social anxiety and attention away from emotional faces.Cogn. Emot. 1999, 13, 673–690. [CrossRef]66. Ambadar, Z.; Cohn, J.F.; Reed, L.I. All smiles are not created equal: Morphology and timing of smilesperceived as amused, polite, and embarrassed/nervous. J. Nonverbal Behav. 2009, 33, 17–34. [CrossRef][PubMed]67. Keltner, D. Signs of appeasement: Evidence for the distinct displays of embarrassment, amusement, andshame. J. Pers. Soc. Psychol. 1995, 68, 441–454. [CrossRef]68. Otten, M.; Mann, L.; van Berkum, J.J.A.; Jonas, K.J. No laughing matter: How the presence of laughingwitnesses changes the perception of insults. Soc. Neurosci. 2017, 12, 182–193. [CrossRef] [PubMed]69. Solarz, A.K. Latency of instrumental responses as a function of compatibility with the meaning of elicitingverbal signs. J. Exp. Psychol. 1960, 59, 239–245. [CrossRef]70. De Houwer, J.; Eelen, P. An affective variant of the Simon paradigm. Cogn. Emot. 1998, 12, 45–61. [CrossRef]71. Lavender, T.; Hommel, B. Affect and action: Towards an event-coding account. Cogn. Emot. 2007, 21,1270–1296. [CrossRef]72. Blair, R.J.R.; Morris, J.S.; Frith, C.D.; Perrett, D.I.; Dolan, R.J. Dissociable neural responses to facial expressionsof sadness and anger. Brain 1999, 122, 883–893. [CrossRef] [PubMed]73. Rozin, P.; Lowery, L.; Ebert, R. Varieties of disgust faces and the structure of disgust. J. Pers. Soc. Psychol.1994, 66, 870–881. [CrossRef] [PubMed]74. Robinson, M.D.; Storbeck, J.; Meier, B.P.; Kirkeby, B.S. Watch out! That could be dangerous: Valence-arousalinteractions in evaluative processing. Personal. Soc. Psychol. Bull. 2004, 30, 1472–1484. [CrossRef] [PubMed]75. Lang, P.J. The emotion probe: Studies of motivation and attention. Am. Psychol. 1995, 50, 372–385. [CrossRef]76. Costa, P.T.; Terracciano, A.; McCrae, R.R. Gender differences in personality traits across cultures: Robust andsurprising findings. J. Pers. Soc. Psychol. 2001, 81, 322–331. [CrossRef]77. Meyers-Levy, J.; Loken, B. Revisiting gender differences: What we know and what lies ahead.J. Consum. Psychol. 2015, 25, 129–149. [CrossRef]78. Baez, S.; Flichtentrei, D.; Prats, M.; Mastandueno, R.; García, A.M.; Cetkovich, M.; Ibáñez, A. Men, women . . .who cares? A population-based study on sex differences and gender roles in empathy and moral cognition.PLoS ONE 2017, 12, e0179336. [CrossRef] [PubMed]79. Hyde, J.S. The Gender Similarities Hypothesis. Am. Psychol. 2005, 60, 581–592. [CrossRef] [PubMed]80. Schulte-Mecklenbeck, M.; Kuehberger, A.; Ranyard, R. (Eds.) A Handbook of Process Tracing Methods forDecision Research 2011; Psychology Press: Hove, UK, 2011.81. Leppänen, J.M.; Tenhunen, M.; Hietanen, J.K. Faster choice-reaction times to positive than to negative facialexpressions: The role of cognitive and motor processes. J. Psychophysiol. 2003, 17, 113–123. [CrossRef]82. Diener, E.; Diener, C. Most people are happy. Psychol. Sci. 1996, 7, 181–185. [CrossRef]83. Harmon-Jones, E.; Gable, P.A.; Price, T.F. Does negative affect always narrow and positive affect alwaysbroaden the mind? Considering the influence of motivational intensity on cognitive scope. Curr. Dir.Psychol. Sci. 2013, 22, 301–307. [CrossRef]84. Van Den Bergh, B.; Schmitt, J.; Warlop, L. Embodied myopia. J. Mark. Res. 2011, 48, 1033–1044. [CrossRef]85. Nikitin, J.; Schoch, S.; Freund, A.M. The role of age and motivation for the experience of social acceptanceand rejection. Dev. Psychol. 2014, 50, 1943–1950. [CrossRef]Brain Sci. 2019, 9, 6 18 of 1886. Koch, S.; Holland, R.W.; Hengstler, M.; van Knippenberg, A. Body locomotion as regulatory process.Psychol. Sci. 2009, 20, 549–550. [CrossRef]87. Taylor, C.T.; Amir, N. Modifying automatic approach action tendencies in individuals with elevated socialanxiety symptoms. Behav. Res. Ther. 2012, 50, 529–536. [CrossRef] [PubMed]© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

QUALITY: 100% ORIGINAL PAPER – NO PLAGIARISM – CUSTOM PAPER

Leave a Reply

Your email address will not be published. Required fields are marked *