• No results found

Positive emotions and prosocial behavior

N/A
N/A
Protected

Academic year: 2021

Share "Positive emotions and prosocial behavior"

Copied!
22
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Graduate School of Psychology

R

ESEARCH

M

ASTER

S

P

SYCHOLOGY

I

NTERNSHIP

R

EPORT

Date: 24 / 06 / 2017 Student

Name : Sally A. M. Hogenboom

Student ID number : 11377909

Supervisor

Within ResMas (obligatory) : Dr. Disa A. Sauter Specialization : Social Psychology

Research center / location : University of Amsterdam

Number of credits (1 ec = 28 hrs) : 18 EC = 504 hours Ethics committee reference code : 2017-SP-7786

Prosocial behaviors are acts that benefit others. However, although prosocial behaviors are ubiquitous, they occur more often for in-group than out-group recipients. The present study builds on Cavanaugh, Bettman, and Luce’s findings (2015) that love uniquely increases prosocial behaviors towards out-group recipients. In a between-subjects design we attempted to induce one of three positive emotions (love, awe, and pride) or a neutral state through memory recall. It was predicted that while all positive emotion conditions (compared to a neutral state) increase helping of in-group recipients, participants will help out-group recipients longer after inductions of love or awe (compared to pride and neutral). Results indicated that participants were not susceptible to the emotion induction paradigm; participants did not report higher feelings of the target emotion compared to participants in other conditions. Furthermore, only a limited number of participants opted to help the recipient. The decision to help was not influenced by the membership of the recipient (in- or out-group) nor by experimental condition (positive emotions condition or neutral).

(2)

Individuals and organizations are less likely to help an out-group than an in-group member, a cause for concern especially when essential needs (e.g., food) are threatened by war or natural disasters (e.g., Cuddy, Rock, & Norton, 2007; Eisenberg, Eggum, Di Giunta, 2010). In 2016, The American Red Cross spent 4% of their budget on International relief (i.e. out-group), which is a fraction of the 12% they spent for Domestic relief purposes (i.e. in-group; American Red Cross, 2016). This is unsurprising given that at the individual level, prosocial behaviors are also performed far more often for in-group than for out-group recipients (e.g., Levine, Prosser, Evans, & Reicher, 2005; Telzer, Ichien, & Qu, 2015). Levine et al. demonstrated that Manchester United fans who saw a confederate falling down were much more likely to help him if the confederate was dressed in a Manchester United t-shirt (i.e., in-group; 92%) than if the confederate was dressed in a Liverpool t-shirt (i.e., out-group; 33%). These findings show that those in need are thus less likely to be helped if they belong to an out-group. The current study seeks to resolve this in-group bias in prosocial behavior, by building on previous findings that positive emotions are able to increase prosocial behaviors towards out-group recipients (e.g., Bartlett & DeSteno, 2006; Eisenberg, 1991).

Prosocial Behavior: Resolving In-group Biases in Helping or Sharing

Prosocial behavior is apparent in babies as young as 14-months old (Warneken & Tomasello, 2009) and benefits both the helper and the recipient (e.g., Aknin, Dunn, & Norton, 2012; Weinstein & Ryan, 2010). The array of prosocial behaviors can be divided in two separate domains; helping and sharing (e.g., Dunfield & Kuhlmeier, 2013; Imuta, Henry, Slaughter, Selcuk, & Ruffman, 2016; Warneken & Tomasello, 2009). Helping are acts that help someone else achieve his/her instrumental goals (e.g., handing the final piece of a puzzle). Sharing are acts of redistributing one’s own goods so that another person can also benefit (e.g., giving someone part of your lunch or monetary donations). Developmental psychologists distinguish between domains of prosocial behavior because children develop them at different ages, however, in adulthood the prosocial behaviors of sharing and helping are often used interchangeably (e.g., Piff, Dietze, Feinberg, Stancato, & Keltner, 2015). This is remarkable given that helping and sharing require different amounts of effort and that the effort required by a potential helper subsequently influences the decision to help (Bartlett & DeSteno, 2006).

Studies have sought to decrease the in-group bias in prosocial behavior in various ways, but were not always consistently successful (e.g., religious priming [Batara, Franco, Quiachon, & Sembrero, 2016], behavioral synchrony [Reddish, Tong, Jong, Lanman, &

(3)

Whitehouse, 2016]). One way through which researchers have consistently been able to increase prosocial behavior towards the out-group is through the induction of incidental positive emotions - emotions that are elicited by sources unrelated to the intergroup relation (e.g., Bartlett & DeSteno, 2006; Cavanaugh, Bettman, & Luce, 2015; Dasgupta, DeSteno, Williams, & Hunsinger, 2009; Piff et al., 2015). However, whilst in-group biases have successfully been decreased in the domain of prosocial sharing, these findings might not hold when more effort is required of a potential helper in domain of prosocial helping.

Cavanaugh and colleagues (2015) studied the effects of love, hope, pride, and compassion on monetary donations for national and international relief purposes (i.e., sharing with in- and out-group recipients). A series of four studies revealed that all positive emotions (compared to a neutral state) increased prosocial behavior towards the in-group. This is in line with other findings in the field (for a review see Eisenberg, 1991). However, Cavanaugh et al. showed that only incidental love, as compared to the other positive emotions, increased donations towards the out-group. This suggests that love can uniquely increase prosocial behavior towards out-group recipients when participants are asked to donate money. Will we still find this pattern of results when participants are asked to help an in- or out-group recipient by actively volunteering in a helping task instead of passively donating money?

Disentangling Positive Emotions and Their Behavioral Consequences

Cavanaugh et al. (2015) demonstrated that the positive emotion love (compared to compassion, pride, and hope) uniquely elicited prosocial sharing of money with out-group recipients. In recent years, the need to differentiate positive emotions (e.g., love from pride) based on their behavioral consequences has been stressed by many researchers (e.g., Campos, Shiota, Keltner, Gonzaga, & Goetz, 2013; Fredrickson, 2004; Lerner & Keltner, 2000; Sauter, 2010). The current study will contribute to this work by establishing whether prosocial behavior towards out-group members can only be increased by love or whether it can also be elicited by another positive emotion. Directly comparing similar positive emotions and their associated behavioral responses will further the understanding of the functions of positive emotions.

The current study will explore whether, in addition to love, awe can increase prosocial behavior towards out-group members. Awe is the feeling of “… being diminished in the presence of something greater than the self, and the motivation to be good to others” (Piff et

(4)

1991). Consequently, positive emotions that strengthen feelings of relatedness to others (e.g., love) are stronger predictors of prosocial behavior than emotions that focus on the self (e.g., pride, Cavanaugh et al., 2015). Piff and colleagues (2015) showed that experiences of incidental awe were associated with the ‘motivation to be good to others’ and increased both prosocial sharing and prosocial helping. However, all recipients belonged to the participant’s in-group. The current study, therefore, will explore the possibility that awe increases prosocial behavior not only towards the in-group, but also the out-group.

Current Study

To summarize, we built on the Cavanaugh et al. (2015) studies to assess whether awe, in addition to love, can increase prosocial behavior towards out-group members when effortful helping is required. Based on Cavanaugh et al. (2015) and Piff et al. (2015) findings we predict that (a) love, awe, and pride (compared to a neutral state) will increase helping of an in-group recipient in a between-subject design, but that (b) only love and awe (compared to neutral and pride) will increase helping of out-group recipients.

Methods

Participants

All participants were Dutch students at the University of Amsterdam and completed the experiment in return for course credit. After exclusion (see ‘Participant Exclusion’ section for details), the data of 79 participants (25 men, 54 women, 0 other) were included in the analyses. All participants were between 18 and 29 years old (M = 20.7, SD = 2.11).

Procedure

At the start of the experiment in Qualtrics, participants were randomly assigned to an emotion condition (love, awe, pride, or neutral) and a beneficiary condition (in-group or out-group). After reading and signing the informed consent form, participants received instructions about the emotion induction task. The true purpose of the experiment was masked by stating that participants’ entries would be used to create an emotion lexicon. Participants recalled three emotionally salient memories and rated each of these memories on intensity. Participants then rated their current state on 10 items measuring both specific (e.g., awe) and general feelings (e.g., arousal). Participants then provided their demographics (gender, age, nationality) and were notified of reaching the apparent end of the experiment. Participants were then provided with the opportunity to help the beneficiary researchers (in-group or

(5)

out-group). If participants chose to engage in the helping task, they could decide after ach trial if they wished to continue helping. If participants chose not to help the beneficiaries, or if they stopped the helping task, a short explanation of the true purpose of the experiment was provided, along with the possibility to leave a comment.

Materials

Emotion Induction. Each target emotions love, awe, pride, or a neutral state was

induced by asking participants to recall three (emotionally) salient memories in which they experienced the target emotion, using the method developed by Piff et al. (2015). We matched their method to create an instruction for love based on Cavanaugh and colleagues’ definition of companionate love (2015, p.659). The online Appendix* contains an overview of the instructions that were used for each of the conditions throughout the experiment.

The participants were required to write approximately five sentences (400-450 characters) for each memory before they were able to proceed. Participants were instructed to not use gibberish, or the same line or character multiple times.

Manipulation Checks. Participants rated the extent to which they felt love, awe, pride,

amusement, joy, compassion, and contentment (presented in a randomized order), in order to establish whether the target emotions were successfully induced. All items were rated on a 7-point Likert scale ranging from 1 ‘does not describe my feelings’ to 7 ‘clearly describes my feelings’. The four non-target emotions were included to de-emphasize the focus emotions. All emotion terms were accompanied by a definition, taken from Fokkinga, Sauter, Fisher, and Desmet (2017; see online Appendix*). For example, “Awe was defined as the feeling when you encounter something that is greater or more powerful than yourself. You feel insignificant and need a moment to adjust.”

A second set of manipulations checks was included to ensure that each positive emotion condition was matched on arousal, and general positive and negative affect. In addition, an intensity rating of each recalled memory was included (“How intense was the event you recalled for you?”) to ensure that all conditions were matched on memory intensity. Memory intensity was rated on a 7-point Likert scale from 1 ‘not at all’ to 7 ‘a great deal’. Current arousal and positive and negative affect were rated in a fixed order on 7-point Likert scales ranging from 1 ‘does not describe my feelings’ to 7 ‘clearly describes my feelings’.

(6)

Helping task. Participants were asked to engage in an apparently unrelated helping

task, which would benefit either an in-group researcher (University of Amsterdam) or an group researcher (Goethe University Frankfurt). A German researcher was used as the out-group beneficiary as it provided the opportunity for a double out-out-group manipulation: the researcher belongs both to a different university and to a different nationality. It was explained that participants would not receive additional course credit and that they could stop at any time. If participants chose to help, they were played a short voice clip (± 1 second) of either a human or chimpanzee vocalization, and were asked to identify the social situation in which this clip was most likely recorded (Sauter, Slocombe, Eisner, & Haun, 2017). The helping task was designed to be repetitive. After each clip, participants were provided the opportunity to stop the helping task or continue by listening to another clip. Helping was operationalized as the number of clips analyzed, as well as the total duration of helping in seconds.

Participant Exclusion

The data of 12 participants were excluded from analyses. Three criteria had been pre-specified as reasons for exclusion. Firstly, participants with a nationality other than Dutch were excluded (N = 3), because the in-/out-group recipient manipulation was based on Dutch recipients as in-group and German recipients as out-group members. The second criterion was that participants with a current feeling score of the target emotion that deviated more than 2

SD’s from the group mean were to be excluded. No participants were excluded based on this

criterion. Thirdly, if participants did not complete the emotion induction phase in good order (e.g., used gibberish or copied the same memory multiple times), these participants were also excluded from analysis (N = 2).

After inspecting the data, it became evident that two additional exclusion criteria were warranted. Firstly, five participants started the experiment multiple times before completing it. As the experimental conditions were randomly assigned at the start of the experiment, starting the experiment multiple times, may have exposed participant to multiple experimental conditions. Secondly, one participant took almost 18 hours to complete the experiment (average response time for other participants was 21 minutes, SD = 10.77). Finally, the data of one participant were not included due to an error in saving his/her response.

(7)

A total 79 participants completed the experiment in good order. Table 1 shows the distribution of participants across conditions, including the number of participants who helped in each condition. All scripts, analyses, and plots (e.g., assumption checks and Q-Q plots) are available in the online Appendix*.

Table 1

Distribution of participants across conditions. The numbers between parentheses represent prosocial helping behavior.

Beneficiary

Emotion In-Group Out-Group Total

N % N % N % Love 5 (2) 6.3 % 6 (0) 7.6 % 11 (2) 13.9 (2.5) % Awe 8 (1) 10.1 % 7 (1) 8.9 % 15 (2) 19 (2.5) % Pride 7 (1) 8.9 % 14 (2) 17.7 % 21 (3) 26.6 (3.8) % Neutral 12 (2) 15.2 % 20 (3) 25.3 % 32 (5) 40.5 (6.3) % Total 32 (6) 40.5 (7.6) % 47 (6) 59.5 (7.6) % 79 (12) 100 (15.2) % Note. Participants were not equally distributed across conditions. This was the result of a true – but unbalanced –

randomization function in the Qualtrics software.

Emotion Induction

Successful emotion induction was operationalized as participants in the target emotion condition (e.g., love) rating their current feelings of that emotion higher than participants in the other conditions (e.g., awe, pride, neutral; Piff et al., 2015). Figure 1 shows participants’ ratings of current feelings per emotion condition.

(8)

Figure 1. Raw data plot of the items measuring current feelings of love [left], awe [middle], and pride [right]. Group means

are represented by the bold horizontal lines. The grey boxes visualize the 95% CI of the mean and the grey dots represent individual data points. The data are outlined by a grey line to show the distribution of the data. All feelings were rated on a 7-point Likert scale from 1 ‘does not describe my feelings’ to 7 ‘clearly describes my feelings’.

Planned analyses. In order to assess whether ratings of current feelings differed

between conditions we planned to run a 4 (between subject – love, awe, pride, neutral) by 3 (within subject – love, pride, awe) MANOVA. The MANOVA showed significant differences in current feelings (love, awe, and pride) between the emotion conditions (love, awe, pride, neutral; Pillai’s Trace = .95, F(3, 75) = 2.76, p = .004). We conducted three post-hoc ANOVA’s to assess the effect of emotion condition (between subject – love, awe, pride, neutral) on current feelings of love, awe, or pride. After Bonferroni correction for multiple comparisons (N = 3), emotion conditions differed significantly on self-reported current feelings of awe (F(3,75) = 3.42, p = .022) and pride (F(3,75) = 3.72, p = .015). The emotion conditions showed a trend towards significant differences on current feelings of love (F(3,75) = 2.71, p = .051). Keeping in mind that the data were not normally distributed (see online Appendix* for details), we have followed up all three ANOVA’s with independent sample t-tests. T-tests were conducted to compare the two emotion conditions with the highest group means on ratings of current love, awe, or pride. Participants in the target emotion condition love (M = 5.65, SD = 1.12) showed a trend towards rating current feelings of love higher than participants in the pride (M = 4.90, SD = 1.02) condition (t(18.69) = 1.86, p = .080, d = .70). Participants in the target emotion condition (awe) did not rate their current feelings of awe (M = 3.5, SD = 1.71) higher than participants in the love condition (M = 3.2, SD = 1.95; t(19.91) = .41, p = .688, d = 0.16). Ratings of current feelings of pride did not differ significantly between the target emotion condition (pride; M = 4.34, SD = 1.52) and the love condition (M = 4.74, SD = 1.54; t(20.13) = -.69, p = .499, d = 0.26).

In conclusion, when adopting the criterion for successful emotion induction used by Piff et al. (2015), the current emotion induction was unsuccessful. Participants in each of the target emotion conditions did not rate current feelings of that emotion higher than participants in the other experimental conditions. For example, participants in the pride and love condition did not differ on their ratings of current feelings of pride and love. This criterion, however, is very stringent, and we thus explored the possibility that a less stringent criterion would conclude successful emotion induction.

Exploratory analyses. A less stringent criterion for successful emotion induction is

that participants in each emotion condition rate current feelings of the target emotion higher

*

(9)

than other feelings of interest. In other words, participants in the love condition should report feeling higher levels of love than awe and pride. This criterion is less stringent because participants merely have to experience the target emotion strongest, independent of how the participants in the other emotion conditions rated their current feelings. It is thus possible, in contrast to the first criterion, that participants across conditions report similar feelings of love, while still being able to conclude successful emotion induction. Differences between current feelings were tested by paired sample t-tests with Bonferroni correction for multiple comparisons (N = 6). Participants in the love condition rated their current feelings of love (M = 5.65, SD = 1.12) higher than current feelings of awe (M = 3.2, SD = 1.95) and pride (M = 4.74, SD = 1.54). However, where the difference between awe and love was significant (t (10) = 3.62, p = .028, d = 1.10), the difference between pride and love was not (t (10) = 1.80, p = .608, d = .54). Participants in the awe condition rated their current feelings of love (M = 4.78,

SD = 0.90) higher than current feelings of awe (M = 3.5, SD = 1.71; t (14) = 3.12, p = .047, d

= .80). Participants in the awe condition did not rate their feelings of awe and pride (M = 3.14,

SD = 1.31) significantly different (t (14) = .78, p = 1, d = .20). Participants in the pride

condition reported higher feelings of love (M = 4.90, SD = 1.02) than pride (M = 4.33, SD = 1.52), this difference was not significant (t (20) = 2.01, p = .346, d = .44). Participants in the pride condition however did rate their current feelings of pride higher than feelings of awe (M = 2.31, SD = 1.33; t (20) = 3.55, p = .012, d = .77).

Even after adopting the less stringent criterion that participants, on average, should rate current feelings of the target emotion higher than other emotional feelings, we were unable to conclude successful emotion induction. We therefore attempted to re-distribute participants across conditions based on their self-report ratings of feeling the experimental emotions. These attempts (included in the Online appendix*) did not result in viable groups for use in further analyses, due to greatly distorted sample sizes. We therefore re-distributed participants into two conditions; all participants from the love, awe, and pride conditions were grouped together in the “positive emotions condition” (N =47) and the neutral condition was left intact (N = 32). Further analyses will utilize this distinction between positive emotions conditions and the neutral condition.

(10)

General State

Planned analyses. Participants in the positive emotions condition rated their current

state as more positive (M = 5.2, SD = 0.9) than did participants in the neutral condition (M = 4.7, SD = 1.2). This trend approached significance (t(54.37) = -1.93, p = .059, d = .45). Participants in the positive emotions condition (M = 2.3, SD = 1.10) rated their current state as slightly less negative than did participants in the neutral condition (M = 2.61, SD = 1.16), this difference was not significant (t (64.138) = 1.16, p = .252, d = .27). On average, participants in the positive emotions condition (M = 3.77, SD = 1.30) reported equal levels of arousal as participants in the neutral condition (M = 3.28, SD = 1.43; t (62.45) = 1.55, p = .126, d = .36). Participants in the positive emotions condition (M = 5.44, SD = 0.86) rated their memories as more intense than participants in the neutral condition (M = 2.54, SD = 1.02). This difference was significant (t(59.026) = 13.18, p < .01. d = 3.07)

Helping

Planned analyses. We hypothesized that participants in the positive emotion

conditions (love, awe, and pride) would help in-group beneficiaries more (number of clips analyzed) and longer (time spent helping) than would participants in the neutral condition.

Exploratory analyses. Due to the small number of prosocial helping instances (N = 12;

15.2 %), we analyzed whether the decision to help (yes/no) was influenced by experimental condition. A Chi-square test of independence between helping (yes/no) and experimental condition (positive emotions/neutral) indicated that participants in the neutral and positive emotions condition were equally as likely to help a beneficiary (χ2

(N = 79) = .01, p = .929; OR = 1.04 [0.46, 2.35]). In addition, a chi-square test of independence between helping (yes/no) and the beneficiary (in-/out-group) revealed that participants were equally as likely to help in- and out-group beneficiaries (χ2

(N = 79) = .53, p = .467, OR = 1.34 [0.59, 3.05]).

Content Analysis

It was not planned to run content analysis on the memories provided by the participants. However, after concluding unsuccessful emotion induction, we sought to establish whether this tendency was reflected in the recalled memories. We analyzed the memories of the participants by means of the Linguistic Inquiry and Word Count software (LIWC; Pennebaker, Booth, Boyd, & Francis, 2015) for use of positive and negative emotion words. Figure 2 shows the raw data plots of the mean percentage of positive and negative words used across three memories.

(11)

Figure 2. Raw data plots of the mean percentage of positive and negative emotion words used during the

emotion induction phase. The top two plots show grouping according to the original experimental conditions, the bottom two plots show the data from the grouped positive emotions condition compared to the neutral condition.

An ANOVA showed that the effect of experimental emotion condition was significant for the number of positive words used in the recalled memories (F (3,75) = 11.03, p < .01, η2 = .30). Post hoc Tukey’s Honest Significant Differences tests (N = 6) indicated that the number of positive words used was higher in the awe (p = .05), love (p < .01), and pride (p < .01) condition than in the neutral condition. The positive emotion conditions (love, awe, and pride) did not differ in the number of positive words used to recall their emotionally salient memories (love-awe p = .237; pride-awe p = .409; pride-love p = .929). A second ANOVA showed that experimental emotion condition was not related to the use of negative emotion words during memory recall (F (3,75) = 1.672, p = 0.18).

An independent samples t-test of the mean percentage of positive emotions words used across the three recalled memories showed that participants in the positive emotions condition (M = 5.06, SD = 2.66) used more positive emotion words to recall their memories than participants in the neutral condition (M = 2.30, SD = 1.52; t (75.134) = 5.86, p < .01).

(12)

SD = 0.84) used approximately the same number of negative emotions words (t (58.375) =

0.85, p = .399).

Discussion

The current study aimed to explore whether positive emotions (love, awe, and pride) increase prosocial helping in general, and whether incidental love and awe – compared to incidental pride and a neutral state – additionally increase prosocial helping of out-group recipients. Analyses indicated that our sample did not respond to the design as expected: the emotion induction was not successful, and only a small number of participants (15.2%) opted to help. We will discuss how differences in sample and study characteristics compared to earlier studies could have influenced the results.

Emotion Induction

Directive writing and memory recall have been utilized in many research paradigms to induce emotions (e.g., Dunn & Schweitzer, 2003; DeSteno, Dasgupta, Bartlett, & Cajdric, 2004; Cavanaugh et al., 2015; Piff et al., 2015). We adopted a similar induction method to Piff and colleagues and thus operationalized successful emotion induction according to their criterion: participants in a target emotion condition (e.g., love) should rate their current feelings of that emotion higher than participants in the other conditions (e.g., awe, pride, neutral). Piff and colleagues reported successful emotion induction as, for example, participants in the awe condition reported higher feelings of awe (M = 6.24) than participants in the pride (M = 3.75) and neutral (M = 2.77) conditions (Table 3 on p. 888). In contrast, participants in the current study reported current feelings of the target emotions on similar levels in at least one other condition than the target condition. For example, participants in the awe condition did not rate their current feelings of awe higher than participants in the love condition. Participants in the love condition did show a trend towards rating their current feelings of love higher than participants in the awe, pride, and neutral conditions, however this trend was insignificant (p = .08) with a large effect size (cohen’s d = .70) and power (.95). From these finding one would conclude that emotion induction was unsuccessful. Therefore, in addition to the criterion adopted by Piff and colleagues, we explored a less stringent criterion: whether participants reported feeling higher levels of the target emotion (e.g., love) than other positive emotions (e.g., awe and pride). In the current study, this was not the case. For example, participants in the awe condition report higher feelings of love than

(13)

awe. Consequently, based on the two different criterion, we conclude that the current sample was not susceptible to the emotion induction paradigm†. To allow for further analyses we re-grouped participants into two condition: the positive emotions condition (prior love, awe, and pride) and the neutral condition. Manipulation checks indicated that – as expected – the positive emotions condition reports higher feelings of general positive affect, but similar levels of general negative affect, arousal, and memory intensity compared to the neutral condition. Therefore, we were able to utilize the distinction between the positive emotions condition and the neutral condition for further analyses concerning helping behavior.

The current paradigm followed the Piff et al. (2015) design and instructions, and thus relied on the understanding of emotion labels (e.g., “Please take a few minutes to think about a particular time, fairly recently, when you experienced companionate love…”). Although Piff et al. successfully induced positive emotions, their series of studies differ from the current study on one major component: our participants were Dutch but completed the experiment in a non-native language (English). Research on bilingualism shows that people experience less emotional resonance (i.e., lower intensity) when operating in their second language (e.g., Caldwell-Harris & Ayçiçeği-Dinn, 2009; Ivaz, Costa, & Duñabeitia, 2016). A decrease in emotional resonance may explain why participants, after emotion induction, rated current feelings of awe with a 3.5, where participants from Piff et al. (2015, study 2) rated feelings of awe with a 6.24 after emotion induction (on 7-point Likert scales). However, inductions of incidental love appeared to be less affected by the non-native operationalization. Self-reports of current feelings of love were negatively skewed (i.e., predominantly higher ratings; Figure 1), with an average rating of 5.65 (on a 7-point Likert scale). These findings contribute towards the field of psychological research seeking to differentiate between positive emotions based on their behavioral consequences (e.g., Fredrickson, 2004), by suggesting that distinct positive emotions may be differently affected by non-native operationalization. Pavlenko (2008) demonstrated that bilingual speakers have different emotion concepts than do monolingual speakers. Bilinguals also perceive themselves as less proficient in expressing salient emotions (e.g., ‘I love you’) in their second language (Dewaele, 2008). Our participants may thus have lacked the vocabulary needed to describe and differentiate between the positive emotions, possibly so that the emotion conditions became more similar

The study was underpowered with regards to studying differences in prosocial behavior between positive emotion conditions (“Using G-power’s a-priori ANOVA fixed effects calculation this resulted in a required

(14)

to each other than would be the case in a native speaker sample. In conclusion, asking Dutch participants to complete the experiment in a non-native language may have influenced the emotional induction paradigm so that emotions were induced less intensely and possibly more similar to each other.

In an attempt to explain why the current sample was not susceptible to the emotion induction, we analyzed the content of the recalled memories. Specifically, we analyzed the mean percentage of positive and negative words used in the separate conditions. Participants in the love condition reported feelings of love, awe, and pride, that were higher than or comparable to feelings reported by participants in the respective emotion conditions. We expected these high levels of self-reported feelings of positive emotions to reflect a higher percentage of positive words used during memory recall. However, this expectation was not reflect by the content analysis: specific emotion conditions did not differ from each other on the percentage of positive words used during memory recall. We propose that in addition to looking at the emotional valence of the memories, further content analyses explore whether the conditions differ in thematic content‡. Griskevicius, Shiota, and Neufeld (2010) described the need for researchers to ask participants to describe events as defined by prototypical elicitors (e.g., ‘Please try to recall an event in your life when another person really took care of you and made you feel better’; p. 194) rather than by emotion labels (e.g., ‘Please recall an event in your life when you felt love’). They argue for this method because 1) it relies less heavily on the participants’ understanding of the emotion label, and 2) emotions are elicited by events, and thus describing the event rather than the emotion itself is more likely to elicit subsequent emotions. Based on this argumentation we would predict additional content analysis to reveal that the recalled memories describe similar elicitors (e.g., a bike ride), thereby possibly inducing similar emotions across the emotion conditions. If analyses would provide evidence for this prediction, future research could explore whether unsuccessful emotion induction such as in the current study, can indeed be prevented by asking participants to recall prototypical elicitors rather than emotionally labeled events.

The combination of a research paradigm that relies on the understanding of emotion labels and a non-native operationalization of the induction paradigm may provide a possible explanation as to why the induction of emotions was unsuccessful in our sample. Future research should explore the possibility that (emotion) induction methods that rely heavily on (emotion) vocabulary are less effective when participants preform the experiments in a

(15)

native language. These findings are especially relevant for research that seeks to compare participants from different nationalities or groups with different native languages whilst using a single research paradigm. In the meantime, when attempting to construct an internationally applicable design, researchers could consider using less verbally biased induction methods such as music (Zentner, Grandjean, & Scherer, 2008) or videos (Hubert & de Jong-Meyer, 1991) to induce incidental positive emotions.

Prosocial Helping

The prosocial helping paradigm was created to measure differences in the extent of helping (i.e., number of clips analyzed and duration of helping). However, offering participants a choice between helping or going straight to the end of the experiment resulted in only 15% (12 out of 79 participants) of the participants deciding to help at all. In combination with the unsuccessful emotion induction, we were thus unable to compare positive emotions on these measures of prosocial helping. As an alternative we explored whether the decision to help (yes/no) was influenced by the target recipient (in-/out-group) or condition (positive emotions condition/neutral). Analysis indicated that participants were equally as likely to help in-group as out-group participants, and participants in the positive emotions condition (love, awe, and pride grouped together) were as likely to help a recipient as participants in the neutral condition.

The Central Bureau for Statistics (CBS) reported that in 2016 over 60% of the higher educated Dutch citizens (i.e., college and/or university level education) voluntarily helped at local (sports) clubs, schools, and religious institutions. In addition, Movisie reported that Dutch citizens on average volunteer between one and four hours a week (2015). These numbers suggest that in general the participants (Dutch students) should not have been unlikely to display prosocial helping behavior. We will therefore discuss how specific design characteristics may have negatively influenced known mediators of prosocial (helping) behavior.

Charness and Gneezy (2008) demonstrated that an anonymous recipient received smaller monetary donations in a prosocial sharing task than recipients who’s family name was known. Levine et al. (2005) showed that participants were less likely to help a falling confederate when he/she was not identifiable as belonging to either the in- or out-group. These studies suggest that anonymity of the recipient decreases the likelihood of prosocial

(16)

possible that participants would have helped more often if we had provided additional personal information about the target recipient. The extent to which the presence and/or content of personal information mediates prosocial sharing and helping behavior could be explored in future studies.

Feelings of empathy have also been found to mediate prosocial behaviors (e.g., Eisenberg, 1991; Eisenberg & Miller, 1987; Prot et al., 2014). Batson and colleagues (1989) showed that spontaneous and induced feelings of empathy increase prosocial helping behavior. Lee and Chang (2007) showed that empathy is a significant predictor for both volunteering and monetary donations. In contrast to real-life helping paradigms such as picking up dropped pens (Piff et al., 2015) or helping someone that fell down (Levine et al., 2005), the current paradigm measures prosocial helping in the absence of a physically present recipient. Studies on bullying suggest that in the absence of a physically present victim (i.e., cyberbullying), feelings of empathy and remorse are less likely to occur (Slonje & Smith, 2008; Freis & Gurung, 2013). It may thus be possible that feelings of empathy are also less likely to occur in the absence of a physically present recipient. One way through which empathy has been induced in the absence of a physically present recipient is by manipulating feelings of guilt (Basil, Ridgway, & Basil, 2008). This method is often used in large scale donation campaigns in newspapers and magazines – acts of prosocial sharing with physically absent (Huhmann & Brotherton, 1997). In the present study, however, we relied on naturally occurring feelings of empathy. Although we did not measure feelings of empathy, the finding that only 15% of the participants decided to help, may indicate that feelings of empathy are less likely to develop naturally in the absence of a physically present recipient. In addition, future research could explore whether empathy demands differ for prosocial helping and sharing behaviors.

A third mediator of prosocial behavior is the expectation of reciprocity. Reciprocity expectations differentiate prosocial behaviors from purely altruistic behaviors in that altruism is not characterized by external benefits (Batson & Powell, 2003, p. 463). Reciprocity can occur through the beneficiary returning the favor, from the society at large, but also through internal benefits such as increased positive affect (Snippe, Jeronimus, Bos, de Jonge, & Wichers, 2017). Simpson and Willer (2008) demonstrated that predominantly ‘egoistic’ people will not display prosocial behavior in the absence of reputational benefits. Lacetera and Macis (2010) showed that positive image concerns (i.e., reciprocity from the community) was a significant predictor for prosocial blood donations in an Italian village. Studies like these illustrate the importance of the potential for reciprocity when promoting prosocial

(17)

behaviors. Reciprocity from the beneficiary is less likely from out-group recipients, which might partially explain why people are generally more inclined to help in-group rather than out-group members (e.g., Cavanaugh et al., 2015; Levine et al., 2005). In the present study, participants were aware that responses were anonymous and helping would thus also be anonymous; therefore decreasing chances of reciprocity by the recipient or reputational benefits from the society.

In conclusion, we proposed that three mediators may have affected participants so that they were unlikely to display prosocial helping behavior in the current paradigm.

Conclusion

The current study aimed to build on findings by Cavanaugh et al. (2015) and Piff et al. (2015) that specific positive emotions increase prosocial behaviors. We hypothesized that incidental love, awe, and pride – compared to a neutral state – would increase prosocial helping of in-group recipients, and that love and awe would additionally increase prosocial helping of out-group recipients. The current sample did not appear to be susceptible to the induction of emotions through recall of emotionally salient memories. In addition, participants were reluctant to help target beneficiaries, regardless of in- or out-group membership.

(18)

References

Aknin, L. B., Dunn, E. W., & Norton, M. I. (2012). Happiness runs in a circular motion:

Evidence for a positive feedback loop between prosocial spending and happiness. Journal of Happiness Studies, 13(2), 347-355.

American Red Cross. (2016). Annual Report. Retrieved from

http://www.redcross.org/images/MEDIA_CustomProductCatalog/m64340273_Annual -Report-2016.pdf

Bartlett, M. Y., & DeSteno, D. (2006). Gratitude and prosocial behavior: Helping when it costs

you. Psychological science, 17(4), 319-325.

Basil, D. Z., Ridgway, N. M., & Basil, M. D. (2008). Guilt and giving: A process model of empathy and efficacy. Psychology & Marketing, 25(1), 1-23.

Batara, J. B. L., Franco, P. S., Quiachon, M. A. M., & Sembrero, D. R. M. (2016). Effects of Religious Priming Concepts on Prosocial Behavior Towards Ingroup and Outgroup. Europe's Journal of Psychology, 12(4), 635.

Batson, C. D., & Powell, A. A. (2003). Altruism and prosocial behavior. Handbook of

psychology. Retrieved from:

https://is.muni.cz/el/1441/podzim2012/SZk1025a/um/Handbook_Of_Psychology_-_Personality_And_Social_Psychology__Malestrom_.pdf#page=486

Caldwell-Harris, C. L., & Ayçiçeği-Dinn, A. (2009). Emotion and lying in a non-native language. International Journal of Psychophysiology, 71(3), 193-204.

Campos, B., Shiota, M. N., Keltner, D., Gonzaga, G. C., & Goetz, J. L. (2013). What is shared,

what is different? Core relational themes and expressive displays of eight positive emotions. Cognition & emotion, 27(1), 37-52.

Cavanaugh, L. A., Bettman, J. R., & Luce, M. F. (2015). Feeling love and doing more for distant

others: Specific positive emotions differentially affect prosocial consumption. Journal

of Marketing Research, 52(5), 657-673.

Centraal Bureau voor Statistieken (10/03/2016). Bijna één op de twee Nederlanders doet vrijwilligerswerk [Report]. Retrieved from

https://www.cbs.nl/nl-nl/nieuws/2016/10/bijna-een-op-de-twee-nederlanders-doet-vrijwilligerswerk

(19)

Charness, G., & Gneezy, U. (2008). What's in a name? Anonymity and social distance in dictator and ultimatum games. Journal of Economic Behavior & Organization, 68(1), 29-35.

Cuddy, A. J., Rock, M. S., & Norton, M. I. (2007). Aid in the aftermath of Hurricane Katrina: Inferences of secondary emotions and intergroup helping. Group Processes &

Intergroup Relations, 10(1), 107-118.

Dasgupta, N., DeSteno, D., Williams, L. A., & Hunsinger, M. (2009). Fanning the flames of prejudice: the influence of specific incidental emotions on implicit prejudice. Emotion, 9(4), 585.

DeSteno, D., Dasgupta, N., Bartlett, M. Y., & Cajdric, A. (2004). Prejudice from thin air: The effect of emotion on automatic intergroup attitudes. Psychological Science, 15(5), 319-324.

Dewaele, J. M. (2008). The emotional weight of I love you in multilinguals’ languages. Journal

of Pragmatics, 40(10), 1753-1780.

Dunn, J. R., & Schweitzer, M. E. (2003). Feeling and believing: the influence of emotion on trust. In Academy of Management Proceedings (Vol. 2003, No. 1, pp. B1-B6). Academy of Management.

Dunfield, K. A., & Kuhlmeier, V. A. (2013). Classifying prosocial behavior: Children's responses to instrumental need, emotional distress, and material desire. Child

Development, 84(5), 1766-1776.

Eisenberg, N. (1991). Meta-analytic contributions to the literature on prosocial behavior. Personality and Social Psychology Bulletin, 17(3), 273-282.

Eisenberg, N., Eggum, N. D., & Di Giunta, L. (2010). Empathy‐related responding: associations

with prosocial behavior, aggression, and intergroup relations. Social issues and policy

review, 4(1), 143-180.

Eisenberg, N., & Miller, P. A. (1987). The relation of empathy to prosocial and related behaviors. Psychological bulletin, 101(1), 91.

Fredrickson, B. L. (2004). The broaden-and-build theory of positive emotions. Philosophical

(20)

Fokkinga, M., Sauter, D., Fischer, A.H., & Desmet, P. (2017). Manuscript in preparation. Griskevicius, V., Shiota, M. N., & Neufeld, S. L. (2010). Influence of different positive

emotions on persuasion processing: a functional evolutionary approach. Emotion, 10(2), 190.

Hubert, W., & de Jong-Meyer, R. (1991). Autonomic, neuroendocrine, and subjective responses

to emotion-inducing film stimuli. International Journal of Psychophysiology, 11(2), 131-140.

Huhmann, B. A., & Brotherton, T. P. (1997). A content analysis of guilt appeals in popular magazine advertisements. Journal of Advertising, 26(2), 35-45.

Imuta, K., Henry, J. D., Slaughter, V., Selcuk, B., & Ruffman, T. (2016). Theory of mind and prosocial behavior in childhood: A meta-analytic review.

Ivaz, L., Costa, A., & Duñabeitia, J. A. (2016). The emotional impact of being myself: Emotions

and foreign-language processing. Journal of Experimental Psychology: Learning,

Memory, and Cognition, 42(3), 489.

Lacetera, N., & Macis, M. (2010). Social image concerns and prosocial behavior: Field

evidence from a nonlinear incentive scheme. Journal of Economic Behavior &

Organization, 76(2), 225-237.

Lee, Y. K., & Chang, C. T. (2007). Who gives what to charity? Characteristics affecting donation behavior. Social Behavior and Personality: an international journal, 35(9), 1173-1180.

Lerner, J. S., & Keltner, D. (2000). Beyond valence: Toward a model of emotion-specific influences on judgement and choice. Cognition & Emotion, 14(4), 473-493.

Levine, M., Prosser, A., Evans, D., & Reicher, S. (2005). Identity and emergency intervention:

How social group membership and inclusiveness of group boundaries shape helping behavior. Personality and Social Psychology Bulletin, 31(4), 443-453.

Movisie (08/05/2015). Feiten en cijfers Vrijwillige inzet. Retrieved from

https://www.movisie.nl/feiten-en-cijfers/feiten-cijfers-vrijwillige-inzet

Pavlenko, A. (2008). Emotion and emotion-laden words in the bilingual lexicon. Bilingualism:

(21)

Language and cognition, 11(02), 147-164.

Piff, P. K., Dietze, P., Feinberg, M., Stancato, D. M., & Keltner, D. (2015). Awe, the small self,

and prosocial behavior. Journal of personality and social psychology, 108(6), 883. Prot, S., Gentile, D. A., Anderson, C. A., Suzuki, K., Swing, E., Lim, K. M., ... & Liau, A. K.

(2014). Long-term relations among prosocial-media use, empathy, and prosocial behavior. Psychological science, 25(2), 358-368.

Reddish, P., Tong, E. M., Jong, J., Lanman, J. A., & Whitehouse, H. (2016). Collective

synchrony increases prosociality towards non‐performers and outgroup members. British Journal of Social Psychology, 55(4), 722-738.

Sauter, D. (2010). More than happy the need for disentangling positive emotions. Current

Directions in Psychological Science, 19(1), 36-40.

Sauter, D., Slocombe, K. E., Eisner, F., & Haun, D. B. M. (2017). Manuscript in preparation. Simpson, B., & Willer, R. (2008). Altruism and indirect reciprocity: The interaction of person

and situation in prosocial behavior. Social Psychology Quarterly, 71(1), 37-52. Slonje, R., & Smith, P. K. (2008). Cyberbullying: Another main type of

bullying?. Scandinavian journal of psychology, 49(2), 147-154.

Snippe, E., Jeronimus, B. F., Bos, E. H., de Jonge, P., & Wichers, M. (2017). The Reciprocity of Prosocial Behavior and Positive Affect in Daily Life. Journal of personality.

Telzer, E. H., Ichien, N., & Qu, Y. (2015). The ties that bind: Group membership shapes the neural correlates of in-group favoritism. NeuroImage, 115, 42-51.

Warneken, F., & Tomasello, M. (2009). The roots of human altruism. British Journal of

Psychology, 100(3), 455-471.

Weinstein, N., & Ryan, R. M. (2010). When helping helps: autonomous motivation for

prosocial behavior and its influence on well-being for the helper and recipient. Journal

of personality and social psychology, 98(2), 222.

Zentner, M., Grandjean, D., & Scherer, K. R. (2008). Emotions evoked by the sound of music:

characterization, classification, and measurement. Emotion, 8(4), 494.

(22)

Appendix

All instructions, analyses, and plots are included in the online appendix:

Referenties

GERELATEERDE DOCUMENTEN

Introduction - Conceptualizing prosocial behaviors in 9 individual and group settings from the perspective of attachment theory. Attachment, caregiving and volunteering: Placing

Experimental studies based on attachment theory demonstrate that dispositional and manipulated attachment security facilitate cognitive openness and empathy,

Beyond the already reported main effects of attachment anxiety and avoidance, the regressions revealed a significant unique contribution of the VFI Values score to loneliness

The purpose of our study is to attempt to replicate these findings in a new sample while dealing with two unaddressed issues: (a) the unique explanatory power of attachment patterns

We conducted hierarchical regression analyses examining the contribution of Attachment Anxiety, Attachment Avoidance and Moral judgment to Voluntarism, Altruistic Reasons

(2) As compared to an average group respect condition, inductions of low group respect would lead to higher levels of group-related worries and lower group commitment but more

We showed that feelings of commitment as well as engagement in group serving efforts and donation to the group following signals of group respect and disrespect are highly

This thesis is dedicated to Medad, my soul mate, and true love, and to our amazing kids; Omry, Amitay, Romy and Alon. To Idit, Galit and Keren for always being there