The Importance of Timing in Reciprocity: An Investigation of Reciprocity Norms Among Indians and Americans

Abstract

After receiving help, individuals tend to experience an immediate increase in obligation to be responsive to the helper. Cross-cultural research has shown that whereas this sense of obligation dissipates for Americans after reciprocation, it remains unchanged after reciprocation for Indians. Is this decrease in obligation felt by Americans temporary, or can it endure over years such that it provides immunity from responding to the helper? And is there a statute of limitations on the experience of obligation for Indians? If individuals do not reciprocate, can obligation expire? We addressed these questions in a vignette-based experimental investigation involving American and Indian adults. Study 1 (N = 153) demonstrated that Americans but not Indians felt less obligation to aid the helper after reciprocation than in situations in which they had not reciprocated years after receiving a benefit. Reciprocation thus provided Americans, but not Indians, with immunity from being responsive to the helper. Study 2 (N = 141) demonstrated that Americans but not Indians felt less obligation years after as compared with months after receiving a benefit. The passage of time thus expired obligation to be responsive to the helper for Americans but not for Indians. Study 3 (N = 129) provided ecological validity to our hypotheses by assessing real-life friendships, showing how both reciprocations and time passage affect obligation independently and in combination. The findings imply that prosocial behavior is affected by both time passage and reciprocation among Americans but not Indians.

Keywords

reciprocity culture time interpersonal responsiveness exchange motivation

Reciprocity has profound consequences for social solidarity and is a defining feature of social life (Groves, Cialdini, & Couper, 1992; Hartup & Stevens, 1997; Molm, 2010). The norm of reciprocity motivates cooperation by giving rise to a sense of obligation to be responsive to individuals from whom we receive benefits (Watkins, Scheer, Ovnicek, & Kolts, 2006). Generally, the obligation one feels to be responsive to another individual increases after receiving help from them and decreases after we have reciprocated (Burger, Sanchez, Imberi, & Grande, 2009; Gouldner, 1960; Greenberg, 1980).¹ However, a recent cross-cultural investigation documented cultural variation in the experience of obligation after reciprocation. Whereas Americans experienced a decrease in obligation to aid the giver of help immediately after reciprocation, Indians did not experience such a decrease in obligation (Miller et al., 2014).

Cross-cultural differences observed in the experience of obligation toward the helper have been explained in terms of reliance on contrasting reciprocity norms. In family and friend relationships, Americans are seen as relying on exchange norms as they tend to feel an obligation to reciprocate soon after receiving a benefit, expressing reciprocity in terms of repayments (Miller et al., 2014; Molm, 2010; Molm, Schaefer, & Collett, 2007; Molm, Whitham, & Melamed, 2012). In contrast, Indians are seen as relying on communal norms as they tend to feel an obligation to be responsive to the needs of the giver of help as and when these needs arise, expressing reciprocity in terms of mutuality (Miller et al., 2014).²

Questions remain, however, about the effects of time passage on the obligation one feels to reciprocate. For how long do Americans, as a population relying on exchange norms, experience a reduced sense of obligation after reciprocation? Is it temporary or can it endure over years? Similarly, as a population relying on communal norms, do Indians feel an obligation toward the giver even years after receipt of a benefit? And, is there a statute of limitations on the experience of obligation, such that it dissipates if reciprocation is not made soon after the receipt of a benefit?

We address these questions in a three-study cross-cultural investigation that introduces the constructs of immunity and of expiration to understand the temporal course of obligation when individuals apply communal versus exchange norms. We sampled an Indian population to contrast with a U.S. population because of documented differences in their levels of interdependence and independence, with Indians displaying a stronger felt obligation to ingroup members than found among Americans (Oyserman, Coon, & Kemmelmeier, 2002). Also, previous research has shown that Indians and Americans differ in their tendencies to rely on communal as compared with exchange norms (Miller et al., 2014) and that reliance on communal norms is not generalizable to all interdependent cultures (e.g., it does not occur among Japanese; see Miller, Akiyama, & Kapadia, 2017).

Immunity, Expiration, and Reciprocity Norms

Immunity and expiration are both ways by which individuals can experience a decrease in obligation toward the giver of help. Immunity is the consequence of fully repaying one’s debt within the expected time frame, thereby absolving the receiver of help from feeling obligated to help the giver. Expiration is the consequence of not reciprocating within the expected time frame, thereby reducing the obligation felt by the receiver of help toward the giver. Whereas immunity bears on timing effects on obligation after reciprocation, expiration bears on timing effects on obligation in the absence of reciprocation. Our contention is that Americans (and not Indians) experience a reduction in obligation through the experience of immunity and expiration, as we believe these constructs apply exclusively to exchange norms (and not communal norms) for two reasons.

One reason is that exchange and communal norms have qualitatively different accounting systems. Whereas exchange norms involve a turn-taking kind of reciprocity in which individuals monitor individual inputs and returns, communal norms involve a holistic kind of reciprocity in which individuals are responsive to each other’s needs as and when these needs arise. Thus, in the case of exchange norms, inputs and returns are balanced immediately; however, in the case of communal norms, individuals balance inputs and returns over extended periods of time (Clark & Aragón, 2013). Immunity, thus, should be applicable solely to exchange norms as the act of reciprocating immediately allows the receiver of help to feel that it is no longer his or her turn to help, thereby absolving the receiver of help of obligation toward the giver. This reduced sense of obligation should endure over time, as the receiver of help feels like he or she does not need to do anything more to clear debt. In contrast, immunity should not be applicable to communal norms as reciprocating does not absolve obligation immediately, since individuals are expected to be responsive to the needs of the giver in a more holistic way. This sense of obligation should, thus, endure over time as the receiver of help feels like he or she must be responsive to the needs of the giver to fully reciprocate.

Another reason that the effects of immunity and expiration will be observed only among Americans is that exchange and communal norms have qualitatively different temporal expectations for reciprocity. Whereas exchange norms involve a short-term repayment system, communal norms do not involve the expectation of short-term reciprocation. In the case of exchange norms, reciprocation is expected soon after receiving a benefit (Clark & Aragón, 2013). This establishes what can be considered a reciprocation window, that is, a period of time within which reciprocation is expected, usually within 3 months or less of receiving a benefit (Miller et al., 2014). Expiration, thus, should be applicable to exchange norms as reciprocating after the closing of the reciprocation window would not be counted as a reciprocation for help received, thereby reducing the obligation the receiver feels to reciprocate to the giver. This sense of reduced sense of obligation should further decrease over time with the movement further and further away from the reciprocation window. In contrast, expiration should not be applicable to communal norms, as communal norms do not involve time windows within which reciprocations are expected. Since there is a non-time-bound expectation of responsiveness to need, the sense of obligation should endure over time as responsiveness to need can still be counted as reciprocations for the initial help received even years into the future (Clark & Aragón, 2013; Miller et al., 2014).

No research to date has explored whether obligation toward the giver of help is affected by the extended passage of time. In lab-based research on communal norms (e.g., Clark, Lemay, Graham, Pataki, & Finkel, 2010; Clark, Mills, & Corcoran, 1989; Clark, Mills, & Powell, 1986; Clark, Ouellette, Powell, & Milberg, 1987; Clark & Waddell, 1985), as well as in research in the game theory tradition (e.g., Molm et al., 2007; Molm et al., 2012), attention has been given to reciprocity behaviors generally lasting no more than an hour. Likewise, in their experiment that directly pitted exchange against communal norms, Miller and colleagues (2014) portrayed the time period within which help was reciprocated as 1 week and did not explore the influence of extended periods of time on obligation. Existing research then provides no information about whether extended time passage influences obligation toward the giver of help.

Goal of the Present Investigation

The goal of the present investigation is to assess ways in which extended periods of time influence obligation felt by the receiver of help toward the helper through the experiences of immunity and expiration in friendships. We examined whether immunity and expiration have contrasting effects among Indians and Americans, as cultural populations that differ in their reliance on communal as compared with exchange norms, respectively (Miller et al., 2014).

Focusing on the concept of immunity in Study 1, we predicted that reciprocation would absolve Americans from feeling obligation to be responsive to the giver, thereby granting them a sense of immunity that would endure over years, due to their emphasis on exchange norms. However, reciprocation would not absolve Indians from feeling obligation to be responsive to the giver and thus not grant them immunity over years, due to their emphasis on communal norms. Focusing on the concept of expiration in Study 2, we predicted that the passage of time would decrease obligation to be responsive to the giver of help for Americans, eventually granting them expiration, due to their emphasis on exchange norms. However, the passage of time would not reduce obligation to be responsive to the needs of the giver for Indians, thus not granting them expiration, due to their emphasis on communal norms. Exploring the function of immunity and expiration as simultaneous experiences, in Study 3, we predicted that the experience of immunity would better predict reduction in obligation than would the experience of expiration for Americans, given their emphasis on exchange norms. Specifically, time passage would influence obligation only when Americans have not reciprocated. However, Indians would feel the same degree of obligation to be responsive to the needs of the giver regardless of the passage of time from when they received a benefit (expiration), or whether or not they have reciprocated (immunity).

Study 1

Study 1 utilized a vignette-based experiment involving a story about an individual who receives help and is then asked by the helper for a favor 5 years into the future. We manipulated the presence of a reciprocation by portraying the individual as having already reciprocated soon after receiving the help or as not having reciprocated. In terms of within-culture differences, we predicted that American participants would judge the receiver of help to feel less obligation toward the giver years after receiving help if they had reciprocated than if they had not reciprocated. Reciprocation would thus provide immunity over years for Americans. In addition, we predicted that Indian participants would judge the receiver of help to feel the same degree of obligation to respond to the needs of the helper years after receiving help, regardless of whether or not they had reciprocated. Reciprocation would, thus, not provide immunity over years for Indians. In terms of between culture differences, we expected Indian participants to judge the receiver to feel more obligation to respond to the needs of the helper than Americans in general, that is, years after receipt of a benefit, regardless of whether or not they have reciprocated. Felt obligation to be responsive to the needs of the giver, thus, would remain stable over years among Indians. We expected these cultural differences to be explained by emphasis on exchange as compared with communal norms.

Method

Participants

We recruited American participants (N = 80) from a university in New York City and Indian participants (N = 73) from a university in the city of Mysore (India). Overall, the sample consisted of 50.3% females with the Indian sample including 49.3% females and the American including 51.2% females. The participants did not differ in age (Indians: M = 25.7 years, Americans: M = 28.7 years) or educational level (mode = bachelor’s degree), with all participants having a minimum of a bachelor’s degree. The U.S. sample included native English speakers who were born in the United States and the Indian sample included native Kannada language speakers who were born in India. We provided a currency-equivalent monetary compensation to all participants.

Materials and procedures

Research assistants fluent in the local language of Kannada conducted interviews with the Indian participants. Native speakers translated the Indian versions of the protocol into Kannada using familiar words that would be readily comprehended by participants. In developing the Indian materials, we used translation and back translation processes to ensure that the meaning of the English questionnaire was preserved. We prepared two culturally specific versions of the vignette that differed only in minor details, such as the use of Indian proper nouns in the Kannada version.

The stimulus materials consisted of a short vignette that portrayed a couple (helper) in need of help from an individual named Anna (receiver) whom they had previously helped. The vignette was adapted from open-ended responses generated by Indian and American participants in previously published data about everyday helping experiences (Miller et al., 2014). Keeping the passage of time constant at 5 years,³ we randomly assigned each participant to reciprocation or no reciprocation conditions. Both conditions portrayed Anna (receiver) who receives a help from a couple (helper) in the form of rent-free housing for 5 months. In the reciprocation condition, Anna reciprocates soon after receiving the help by volunteering to spend three weekends uploading documents on the Internet for the couple from whom she had received help. We included a major act of reciprocation so that the reciprocation would be memorable in relation to the need that occurred several years later. In the no-reciprocation condition, Anna did not make any reciprocation. In both conditions, we portrayed the couple as requesting help from Anna 5 years later, and we portrayed Anna as politely declining this request for help. The gender of the protagonist was counterbalanced, that is, half the participants read a story about how a male protagonist receives help from a couple. However, for ease of explanation, we refer to the protagonist as Anna, henceforth. Below is the vignette, with the bolded text indicating the presence of the sentence included in the reciprocation condition only.

Anna is a graduate student who is collecting archaeological data for her dissertation in Colorado. She meets a local middle-aged couple who invite her to stay with them, as they live close to her archaeological site and think it would be interesting to have a scholar live with them. Anna stays with the couple for 5 months, rent free. As a thank you to the couple for letting her stay with them, she decides to help them set up their business online. She spends three consecutive weekends uploading all the data onto their computers. She puts a lot of time and effort into this job to show her appreciation.

Five years later, after Anna has moved to New York, the couple contacts her and asks her to spend several days in New Jersey receiving furniture shipments for them as they are planning to move to New Jersey. This would include supervising each unloading of the furniture shipments to make sure no goods are damaged. Anna works in the city as a college professor, and this would require her to spend a significant amount of time over a few days doing this work for them. Anna decides to decline to help the couple with their furniture shipment. She politely tells the couple that she cannot help them out with their furniture shipment.

Measures

To assess whether participants were using exchange norms or communal norms, we first asked participants to generate their spontaneous explanations of Anna’s decision to decline helping the couple. We asked all participants, “Please explain why you think Anna declined helping the couple.”⁴

Measures of obligation

To assess felt obligation that Anna (receiver) was perceived to feel toward the couple (helper), we asked participants to respond to the following question on a 0-to-6-point scale: “How much of an obligation, if any, does Anna have to be responsive to the couple’s request for help?” We assessed the perceived legitimacy of Anna’s decision to decline helping the couple by asking, “Do you think Anna is justified or not justified in deciding to decline helping the couple?” (Yes/No). We also assessed moral emotions associated with Anna declining to help the couple by asking, “Do you think Anna feels guilty or not guilty after deciding to decline helping the couple?” (Yes/No).

Qualitative coding

We devised a two-category mutually exclusive coding scheme to assess whether participants generated exchange-based reasons or communal reasons in explaining Anna’s decision to decline helping the couple. After attending a training session about the characteristics of exchange and communal norms for 1 hr, two coders (both blind to the hypotheses and experimental conditions in the study) independently coded the open-ended data. We coded for exchange when participants made reference to any of the characteristics of exchange, that is, (a) the length of time that had passed since receiving the benefit (as benefits are expected in the short term), for example, “Too much time has gone by since she received help from them” or (b) the presence of a reciprocation or the expectation of return, for example, “She already set up the business online,” or “Anna is not expected to reciprocate, it is Anna’s choice.” We coded for communal norms when participants made reference to any of the characteristics of communal norms, that is, (a) references to competing needs or role-related duties experienced by Anna, for example, “Anna has responsibilities to attend to as a professor,” and “She is neglecting her responsibility,” or (b) obligation to respond to the needs of the couple, for example, “The couple need her help and she should have helped them.” We obtained inter-rater reliability at 81.4% (k = .814, p < .006).

Results

In this and in the other studies in the investigation, we undertook factorial and mixed-design ANOVAs with Bonferroni corrections to assess the dependent variables and the chi-square test to analyze the open-ended data. We used the Preacher and Hayes (2004) bootstrapping method for moderation and mediation analysis. When the mediator was dichotomous, we employed the Mplus 7.4 software using the bootstrapping method to test mediation (Muthen & Asparauhov, 2015). We used gender as a covariate in all the main analyses to control for gender effects.

Statistical power analysis for sample size estimation for this and the other two studies in this investigation revealed that to obtain a medium/large effect size (Cohen, 1992) with an alpha = .05 and power = 0.80, the projected sample size needed (Faul, Erdfelder, Buchner, & Lang, 2009) was N ≥ 120 for the between-participant comparisons. Thus, our samples in Studies 1 (N = 153), 2 (N = 141) and 3 (N = 129) were adequate for the objectives of this study.

Reliance on communal versus reciprocal exchange norms

In generating reasons for why Anna declined to help the couple, Americans (71.42%) were more likely to mention exchange-based reasons, whereas Indians (75.34%) were more likely to mention communal reasons, χ² (2, N = 144) = 30.247, p < .001. As seen in the following sample response, Americans mentioned exchange-based reasons for declining help: “Because Anna compensated the couple for their generosity/“I think its ok that she declined, it’s been 5 years, that’s a long time, plus she has done her part.” In contrast, in explaining Anna’s actions, Indians focused on competing needs or role-related duties, as may be seen in the sample Indian response: “Anna declines because he probably has to take tuitions, make notes, and has other responsibilities toward her students”/”She has responsibilities toward her family, so she cannot leave.”

Obligation index

Since the observed alpha level was high (Cronbach’s alpha = .87), we calculated a mean composite score of the three dependent variables (obligation to help, justification in declining, and guilt for declining) after transforming them into z scores. We refer to this composite score as the obligation index: High scores indicate greater obligation.

A 2 (Culture) × 2 (Condition) factorial ANOVA, including gender as a covariate, on the obligation index revealed significant effects of culture, F(1,148) = 62.995, p < .001, η² = .299; condition, F(1,148) = 14.069, p < .001, η² = .087; and gender, F(1,148) = 7.026, p = .009, η² = .045; and a significant Culture by Condition interaction, F(1,148) = 4.618, p = .033, η² = .030. The gender effect revealed that females (M = 0.152, SD = 0.833) scored higher on the obligation index than did males (M = −0.152, SD = 0.911). As predicted, Americans reported greater obligation in the no-reciprocation condition as compared with the reciprocation condition, F(1,148) = 18.113, p < .001, η² = .109, while Indians showed no condition differences in perceived obligation, F(1,148) = 1.232, p = .269, η² = .008. As predicted, a main effect of culture occurred indicating that Indians felt greater obligation in both the reciprocation, F(1,148) = 53.060, p < .001, η² = .264, and no-reciprocation conditions, F (1,148) = 16.056, p < .001, η² = .098 (see Table 1 for means).

Table 1.

Descriptive Statistics for the Dependent Variables in Studies 1 and 2.

Study 1: Immunity	The United States		India
Study 1: Immunity	ReciprocationM [95% CI]	No ReciprocationM [95% CI]	ReciprocationM [95% CI]	No ReciprocationM [95% CI]
Obligation to help	2.01 [1.52, 2.50]	3.13 [2.60, 3.67]	4.58 [4.04, 5.13]	4.92 [4.38, 5.45]
Justified to decline	0.91 [0.76, 0.97]	0.50 [0.32, 0.68]	0.13 [0.05, 0.32]	0.00 [0.00, 1.00]
Guilt after declining	0.40 [0.26, 0.55]	0.73 [0.55, 0.86]	0.75 [.59, 0.87]	1.00 [0.0, 1.00]
Study 2: Expiration	YearsM [ 95% CI]	MonthsM [95% CI]	YearsM [ 95% CI]	MonthsM [ 95% CI]
Obligation to help	2.66 [2.27, 3.04]	3.48 [3.13, 3.83]	4.32 [3.95, 4.70]	4.19 [3.83, 4.55]
Justified to decline	2.93 [2.50, 3.36]	1.84 [1.45, 2.23]	1.69 [1.27, 2.12]	2.32 [1.92, 2.72]
Likely to help	0.67 [0.54, 0.81]	0.87 [0.75, 0.99]	0.90 [0.77, 1.03]	0.79 [0.66, 0.92]
Importance of timing	2.24 [1.71, 2.76]	0.82 [0.35, 1.30]	2.00 [1.49, 2.52]	1.72 [1.23, 2.21]
Busyness of Schedule	4.53 [3.98, 5.07]	3.30 [2.80, 3.80]	3.14 [2.60, 3.69]	3.43 [2.92, 3.94]

Note. All means presented in this table are adjusted means for the covariate gender. CI = confidence interval.

We performed a mediational analysis to test whether the cultural differences observed on the obligation index were explained by reliance on communal (scored 1) versus exchange (scored 0) norms. In line with our hypotheses, we found that the indirect effect of culture on obligation via the mediatior (exchange vs. communal norms) was significant, b = .252, SE = .111, p = .024, 95% CI [0.086, 0.587]. With the mediator present in the model, the direct effect of culture remained significant, indicating that the reliance on communal versus exchange norms partially mediated the relationship between culture and the measure of obligation, b = .692, SE = .182, p < .001, 95% CI [0.388, 1.156]. Americans were lower on the obligation index in both conditions due to their emphasis on exchange as compared with communal norms (see Figure 1).

Figure 1.

Study 1: Emphasis on communal versus exchange norms partially mediates the relationship between culture and the obligation index.

Dicussion

The findings support the study hypothesis regarding the role of reciprocation in providing immunity from being responsive to the helper’s needs even years after receipt of a benefit among Americans. Equally, the findings support the hypothesis regarding the inconsequential nature of such reciprocations on future responsiveness toward needs of the helper among Indians. This was observed in the tendencies of Americans, but not Indians, to report less obligation to aid the helper years after receiving a benefit after reciprocation as compared with a situation in which the initial help had not been reciprocated. The study also showed that the decrease in overall obligation in both conditions felt by Americans was explained by reliance on exchange as compared with communal norms. Americans in the reciprocation condition felt less obligation to be responsive to the helper’s needs than did Indians since they felt they had already reciprocated. Americans also felt less obligation than did Indians in the no-reciprocation condition and reciprocation condition due to the passage of time that had passed since the receiver had received a benefit. Such findings underscore the importance of turn-taking for those relying on exchange but not for those relying on communal norms.

We designed Study 2 to address questions about expiration. Study 2 also addressed certain methodological limitations of Study 1. Study 1 did not take into account whether cultural differences occurred in perceived cost of the initial benefit. To address this methodological issue, Study 2 assessed perceived cost of the initial benefit received. Furthermore, we held constant the degree of contact between the helper and the receiver, and recruited (via Mechanical Turk) a more diverse sample to make the results generalizable to larger populations.

Study 2

Study 2 involved a vignette-based experiment in which we manipulated the passage of time (months vs. years) after which the helper asks the receiver for a favor in return. We predicted that American participants would judge the receiver to feel less obligated to respond to a request for help years after receiving a benefit as compared with months after receiving a benefit as requests made within months would be within the reciprocation window, and requests made years after would be considered outside of the recirpocation window. Time passage would thus provide expiration of obligation for Americans. We also predicted that Indian participants would judge the receiver to feel equally obligated to respond to a request for help made by the helper months or years after receiving a benefit, as responsiveness to need made at any time would still be considered as reciprocations. Time passage would thus not expire obligation for Indians. In terms of cultural differences within each condition, we predicted that Indians and Americans would feel equally obligated to respond to a request made by the helper within the reciprocation window, that is, months after reciving a benefit, but that Americans would feel less obligated to respond to a request for help than Indians outside of the reciprocation window, that is, years after receiving a benefit. We expected these cultural differences to be explained by reliance on exchange as compared with communal norms.

Method

Participants

We collected data via Mechanical Turk. Balanced by gender, the sample included 71 European American and 70 Indian participants, approximately half of whom were randomly assigned to the months and half to the years condition. The participants did not differ in age (Indians: M = 33.04 years, Americans: M = 33.58) or educational level (Mode: bachelor’s degree), with all participants having a minimum of a bachelor’s education. Overall the sample consisted of 50.3% females with the Indian sample consisting of 50.0% females and the American sample consisting of 50.7% females. The U.S. sample included native English speakers whereas the Indian sample included individuals (95.6%) who reported a high level of English proficiency. Research assistants checked English level proficiency via the open-ended responses from grammar and sentence structure to ensure English diction and comprehension. No participants were dropped from the sample for lack of English proficiency. We provided a currency equivalent monetary compensation to all participants.

Materials and procedures

The stimulus materials consisted of a vignette similar to the one used in Study 1 except that we controlled for closeness of the characters by adding relevant details regarding the extent of contact over time between the helper and receiver, and equalized the age of the helper and receiver. Keeping the degree of contact between the characters (closeness) constant, we randomly assigned each participant to a months or years condition. Each condition portrayed a student Anna (receiver) who receives help from a couple (helpers) in the form of rent-free housing for a semester. In the months condition, the couple asks Anna for help after 3 months, and in the long-term condition, the couple asks Anna for help after 3 years. We prepared two culture-specific versions of the vignette that differed only in minor details, such as in using Indian proper names for people and places. In addition, the gender of the protagonist was counterbalanced, that is, half the participants read a story about how a male protagonist receives help from a couple. However, for ease of explanation, we refer to the protagonist as Anna henceforth. Below is the vignette used in the study (participants read only one of the bolded phrases below):

Initial Situation

Christina and George are a young married couple who lived in a small town in Colorado. One day at a local eatery, they met Anna—a graduate student who was collecting archaeological data for her dissertation. Anna needed a place to live as all the motels and hotels were very far from her archeological site; Christina and George invited Anna to stay with them as they lived close to her archaeological site and had a spare room. They also felt that it would be interesting to have a scholar staying with them.

Anna ended up living in the couple’s house for 5 months as their guest, where they provided her with free room and board. Anna completed her dissertation and moved back to New York.

Three months later/Three years later

Three Months/Three Years went by and Anna got married and secured a job as an assistant professor at a state university in New York City. George got a job in the computer industry, and he and his wife had a baby; over this period, Anna and the couple had no particular contact with each other; however, they had exchanged email addresses before Anna moved back to New York.

Near the end of the 3 months (3 years), George unexpectedly lost his job after a downturn in the economy in Colorado; the only place that he was able to find new work was on the East Coast. The family arranged for a moving van to transport their furniture to their new home in New Jersey. However, the moving van was scheduled to arrive before they would be able to reach New Jersey by car. Christina and George remembered that Anna lived in New York and decided to call her and ask if she would be willing to spend 2 days in New Jersey receiving the furniture shipment for them to make sure that no goods were damaged. Anna works in the city as a college professor, and this would require her to spend a few days away from her office to do this.

Measures

To assess whether participants were using exchange or communal norms, we first asked participants to generate open-ended spontaneous responses about the factors they felt Anna would consider in making her decision: “In your opinion, what are some of the factors Anna will consider in making a decision?”⁴

Measures of obligation

We assessed the likelihood of helping by asking participants to respond to the following question on a dichotomous (Yes/No) scale: “Do you think Anna will help the couple?” We then presented participants with the following prompt:

As it turns out Anna declines to help the couple with their furniture shipment. She politely tells the couple that she cannot help them out with their furniture shipment.”

To assess the degree of obligation that participants felt Anna (the receiver of help) had toward the couple (helpers), we asked on a 0-to-4-point Likert-type scale: “How much of an obligation, if any, does Anna have to help the couple even if she does not want to?” To assess participants’ evaluation of Anna’s decision to decline helping the couple, we asked on a 0-to-4 Likert-type scale: “Is Anna justified in deciding not to help the couple?”

Impressions of Anna’s personality

To assess participants’ evaluations of Anna’s nature and personality, we asked participants to generate open-ended responses to the following question: “Briefly describe your impressions of Anna’s nature and personality. What does she seem like as a person? Please be as detailed as possible.”

Reasons for declining help

We assessed participants’ explanations of Anna’s actions by asking participants on a 0-to-5-point Likert-type scale, “Please rate the importance of the following considerations in explaining why Anna did not agree to help the couple.” We used two filler items: Busyness of Anna’s schedule: “Anna’s schedule is too busy” and “Inappropriateness of the request: The request made by the couple is inappropriate” that we included along with our dependent variable Importance of timing: “Too much time has passed by since the couple helped her.”

Control questions

To assess perceived cost of the initial benefit, we asked participants on a 0-to-5-point Likert-type scale: “Please rate the cost/effort involved on part of the couple in letting Anna live with them for 5 months rent free.”

Qualitative coding

We devised a two-category mutually exclusive coding scheme to assess whether participants generated exchange-based factors or communal factors in deciding whether or not to help. We used the same coding scheme and procedure as in Study 1. Two research assistants (both blind to the hypotheses and experimental conditions in the study) coded the open-ended responses independently. We obtained inter-rater reliability at 94.6 % (k = .946, p < .001).

We also devised a three category mutually exclusive coding scheme to assess participants’ evaluations of Anna’s nature and personality in terms of positive, negative, and neutral impressions. We coded for positive impressions when participants made positive remarks about Anna, for example, “ Anna seems like a well to do, good person/She seems caring and trusting.” We coded for negative impressions when participants made negative remarks about Anna, for example, “I think Anna is selfish and self-cenetered/She is ungrateful and unkind.” We coded for neutral impressions when participants generated neither positive nor negative impressions of Anna, for example, “She is an ordinary person; she reacted to the situation rather than the people who helped her in need.” We obtained inter-rater reliability at 81.8 % (k = .818, p < .001).

Results

Control questions

A 2 (Culture) × 2 (Condition) ANOVA on ratings of the cost of the help received revealed nonsignificant effects of culture, F(1,139) = 1.393, p = .240, η² = .010; condition, F(1,39) = 1.311, p = .254, η² = .009; and of the Culture × Condition interaction, F(1,139) = .206, p = .651, η² = .001. Indians and Americans showed no condition differences or cultural differences in ratings of cost of the initial benefit received.

Reliance on communal versus exchange norms

In generating factors taken into consideration in the decision about whether to help the couple, Americans (68.25%) were more likely to mention exchange-based factors, whereas Indians (94.04%) were more likely to mention communal factors, χ² (2, N = 130) = 54.667, p < .001. As seen in the following sample response, Americans mentioned exchange-based factors: “She probably won’t help them; it’s been a long time since she received help, and it would be inconvenient” or “She should help them because it would be a good way to pay them back.” In contrast, Indians focused on the needs of the couple, as may be seen in the sample Indian response below: “As far as I am concerned, she will respond positively because they helped her when she was in need, and now they need help” or “She probably won’t accept the help of the couple because she is settled in her career now and has other responsibilities.”

Impressions of Anna’s nature and personality

In generating impressions of Anna’s personality, Americans in the short-term condition (72.5%) were more likely to express negative personality impressions of Anna than were those in the long-term condition (25.8%), χ² (2, N = 71) = 17.052, p < .001. In contrast, Indians showed no condition differences in their impressions of Anna’s personality, χ² (2, N = 68) = 1.648, p = .307.

Reasons for declining to aid the helper

A 2 (Culture) × 2 (Condition) MANOVA, with gender as a covariate, on ratings of the reasons for declining to help the giver of help (importance of timing, busyness of schedule, and inappropriateness) revealed significant effects of culture, F(3,134) = 3.706, p = .013, η² = .077; condition, F(3, 134) = 4.591, p = .004, η² = .093; and Culture × Condition interaction, F(3,134) = 4.610, p = .004, η² = .094. Importance of timing, F(1,134) = 11.041, p = .029, η² = .035, and busyness of schedule, F(1,134) = 20.105, p = .005, η² = .056, yielded a significant Culture × Condition interaction. Americans rated busyness of schedule, F(1,36) = 10.643, p = .001, η² = .073 as a more important reason for declining to help in the months condition, and timing, F(1,36) = 15.524, p < .000, η² = .102, as a more important reason in declining to help in the years condition. In contrast, Indians showed no condition differences on busyness of schedule, F(1,36) = 1.456, p = .446, η² = .004, or importance of timing, F(1,36) = 1.445, p = .425, η² = .005, as reasons to decline helping the giver (see Table 1 for means).

Obligation index

Since the observed alpha level was high (Cronbach’s alpha = .522), we calculated a mean composite score of the three dependent variables (obligation to help, justification in declining, likelihood of helping) after transforming them into z scores. We refer to this composite score as the obligation index: High scores indicate greater readiness to help.

A 2 (Culture) × 2 (Condition) factorial ANOVA, including gender as a covariate, on the obligation index revealed significant effects of culture, F(1,136) = 19.799, p < .001, η² = .120, and a significant Culture by Condition interaction, F(1,136) = 19.799, p <.001, η² = .127. As predicted, Americans reported greater obligation in the months condition as compared with the years condition, F(1,136) = 18.779, p < .001, η² = .121, while Indians showed no condition differences in obligation, F(1,136) = 3.896, p = .050, η² = .028. In terms of differences within each condition, as predicted, Americans and Indians did not differ in the degree of obligation they felt toward the helper in the months condition, F(1,136) = .013, p = .909, η² = .000, but Americans reported lower obligation than did Indians in the years condition, F(1,136) = 35.583, p < .001, η² = .207 (see Table 1 for means).

Using the Preacher and Hayes (2004) method (PROCESS MODEL 14), we undertook a moderated mediation analysis to examine whether importance of timing as a reason for declining to help mediated the effect of condition (months versus years) on the obligation index for Americans but not for Indians. Based on this procedure, conditional indirect effects were calculated on two levels of the moderator: culture. In line with our hypotheses, results yielded a significant indirect effect of condition (months vs. years) on the obligation index via importance of timing for Americans, b = −.0882, SE = .0577, 95% CI [−.2355, −0.0056] but not for Indians, b = .0140, SE = .0459, 95% CI [−0.0585, 0.1306] indicating that importance of timing was a significant mediator of the relationship between condition and the obligation index for Americans but not for Indians (see Figure 2).

Figure 2.

Study 2: (a) The importance of timing fully mediates the relationship between condition and the obligation index on one level of the moderator culture, that is, Americans, but not for Indians.

Finally, a mediation analysis was perfomed to test whether the cultural differences observed on the obligation index in the years condition were explained by reliance on communal (coded 1) versus exchange (coded 0) norms. In line with our hypotheses, we found that the indirect effect of culture on the obligation index via the mediatior (exchange vs. communal norms) was significant, b = .441, SE = .241, 95% CI [0.178, 0.1.672]. With the mediator present in the model, the direct effect of culture remained significant, indicating that the emphasis on communal versus exchange norms partially mediated the relationship between culture and the obligation index, b = .505, SE = .234, p = .031 95% CI [0.141, 0.969]. Americans scored lower than did Indians on the obligation index years after receiving a benefit due to their emphasis on exchange as compared with communal norms.

Discussion

The findings support the study hypotheses regarding the existence of an expiration date on felt obligation toward the helper among Americans, and for the nonexistence of an expiration date on felt obligation among Indians. This was observed in the tendencies of Americans, but not Indians, to report less obligation toward the helper after the closing of the reciprocation window, that is, years after, as compared with within the reciprocation window, that is, months after receiving the initial help. The study also showed that this decrease in obligation to help felt by Americans after years was indeed due to the importance of timing in responding to help. Whereas consideration of timing as an important reason to decline helping decreased obligation to help years after receipt of the initial benefit, it had no impact on obligation to help months after receipt of the benefit. In addition, we found evidence to support our prediction that timing would not affect obligation to help for Indians: Timing did not predict obligation toward the helper in the months or years conditions for Indians. Finally, the study showed that differences in overall obligation felt by Americans and Indians years after receiving a benefit was explained by reliance on exchange as compared with communal norms, respectively.

Taken together, the results demonstrate the time-sensitive nature of exchange norms as discussed by both Molm (2010) and Miller et al. (2014). Whereas these authors stress the importance of reciprocation within a short-term time span for individuals relying on exchange norms, our results provide two new ways of looking at the impact of reciprocations over time. First, the receiver of help must reciprocate benefits given in a short-term time span, or else obligation toward the helper will expire. Second, the helper must redeem this reciprocation in a short-term time span or else the reciprocation window will close, and the helper will have missed his or her chance to redeem any benefits he or she gave to the receiver.

Although Studies 1 and 2 provide evidence of expiration and immunity in relation to one type of hypothetical situation, questions remain about the generalizability of the results to real-life relationships and different types of help received. In addition, Studies 1 and 2 demonstrate how immunity and expiration work independently of each other, but do not explore situations in which the experiences of immunity and expiration occur simultaneously. We undertook Study 3 to test the research hypotheses in the context of social support interactions among real-life friends and to assess the relationship between the constructs of immunity and expiration. How does reciprocating (leading to immunity) and time passage (leading to expiration) together reduce obligation toward the helper? To address this question, our design crossed both time passage (months vs. years) and reciprocation (reciprocation vs. no reciprocation). In addition, we included measures of cost of the help received from the real-life friend, as well as of the cost that would be involved in later helping this friend.

Study 3

Study 3 involved an interview-based design in which we crossed both time passage (months vs. years) and reciprocation (reciprocation vs. no reciprocation). We hypothesized that reciprocation would moderate the relationship between time passage and obligation, since time passage can only influence obligation in the absence of reciprocation. Thus, we hypothesize that reciprocation will decrease obligation toward the helper regardless of time passage for Americans but not for Indians. Specifically, we predict that, when Americans have not reciprocated, they will be less ready to aid the helper years after than months after receiving a benefit, as the reciprocation window will still be open months after receiving a benefit but not years after receiving a benefit. However, when Americans have reciprocated, they will be equally likely to aid the helper years or months after receiving a benefit, as reciprocation will give Americans immunity. In contrast, we expect Indians to be equally likely to feel obligation to respond to the helper’s needs, whether or not they have reciprocated and whether the helper makes this request months or years after receiving a benefit.

Method

Participants

We collected data among university student populations, with American participants recruited from the greater New York City area and Indian participants recruited from the city of Mysore (India). The sample included 64 European American and 65 Indian participants (N = 129). Overall, the sample consisted of 53.1% females with the Indian sample including 50.7% females and the U.S. sample including 56.2% females. Participants did not differ in age (India: M = 25.62, the United States: M = 24.48) or educational level (Mode: bachelor’s degree), with all participants having a minimum of a bachelor’s degree. The U.S. sample included native English speakers whose parents were born in the United States, and the Indian sample included native Kannada language speakers whose parents were born in India. We provided a currency-equivalent monetary compensation to all participants.

Procedure and materials

We used the same translation procedures as in the earlier studies. We conducted individual interviews with participants in which we asked respondents to narrate a real-life situation in which they had received a major tangible help from a friend. The interview was conducted in a semistructured format in which the interviewer would ask the question to ensure that participants were indeed thinking about tangible helps and then participants would write down their answers on the sheet of paper provided to them. We sampled tangible help because greater cost is involved in situations that involve tangible help as compared with merely emotional support.⁵

Time-passage manipulation

We randomly assigned participants to a months or years condition. In the months condition, participants narrated a real-life situation in which they received help within the last 12 months, while in the years condition, participants narrated a real-life situation in which they received help more than 3 years ago. We prepared two culture-specific versions of the interview that differed only in minor details, such as in the use of local proper names and city names for the Indian sample. We used the probe listed below in asking respondents for the example (participants read only one of the bolded phrases below):

Think of a time when you received tangible help from a friend. This should be a major help you have received more than 3 years ago/within the last 12 months. It should be a situation that involved significant cost/effort on your friend’s part. It should be a situation in which your friend undertook a service for you or gave you something tangible, and not merely a situation involving emotional support. It should not be a situation in which your friend merely provided you with some information or something you consider trivial. Please describe this behavior.

Reciprocation categorization

We assessed whether participants reciprocated by asking the following two questions: “Did you do anything for your friend after you received this help? (Yes/No)” and “If yes, please describe in detail what you did for your friend” (open-ended question). It was expected that some of the participants would not have reciprocated the help received from their friend (see “Results” of Study 3). To avoid making participants uncomfortable by pressing them on why they did not reciprocate help received, we did not ask participants to explain their motivation for reciprocating or not reciprocating. Thus, emphasis on communal versus exchange norms was not empirically investigated in this study.

Next, we presented participants with hypothetical scenarios that asked them to respond to requests for help from the real-life friend who had helped them. These scenarios were adapted from open-ended responses generated by Indian and American participants in previously published data about everyday helping experiences (Miller et al., 2014). In this questioning, we asked participants to respond to hypothetical rather than real-life instances of help to hold constant the types of situations under consideration. To reduce social desirability effects that might lead participants to agree to help, we asked participants to first respond to a scenario that involved providing low cost help, followed by a scenario that involved providing high cost help. We treated the first scenario as a filler item whose sole purpose was to reduce social desirability effects,⁶ and did not expect to find any cultural differences on this scenario. We used the high cost scenario that was presented second to assess readiness to help the real-life helper. Below is the text we employed in the low cost and high cost scenarios:

Low Cost Scenario: Imagine that you have an extremely busy week at work and this friend whom you just described, the one who helped you, was, moving from New York to New Jersey. (Name of friend) was travelling for a few weeks before the move and needed help moving. Imagine that he or she asked you if you could stop by his or her apartment one morning and be available to accept and sign boxes he or she was getting delivered. This would mean that you would have to take a morning off work to do this for him or her and hence would only be able to get into work a few hours late (getting to work late would not result in loss of pay).

High Cost Scenario: Now, imagine that you have an extremely busy week at work, and this friend whom you just described, the one who helped you, was moving from Boston to Washington, D.C. (Name of friend) was travelling for a few weeks before the move and needed help with driving packed boxes to Washington. Name of friend does not have a driver’s license and asks you to help him or her drive down with him or her. This would mean that you have to spend days helping your friend instead of having free time for yourself, and you only have three vacation days left in the calendar year.

Measures of obligation

We assessed likelihood of helping on a 0-to-5-point rating scale by asking, “How likely is it that you would help your friend in this situation?” We assessed guilt associated with declining helping on a 0-to-5-point rating scale by asking, “How guilty would you feel if you declined helping your friend in this situation?”

Control questions

To assess timing of the receipt of the benefit, we asked participants, “How long ago did you receive this help you just described from your friend?” To assess perceptions of cost, we asked participants at the end of the interview to respond, on a 0-to-5-point rating scale, to the following questions: (a) “Please rate the cost that is involved on part of your friend (the help you initially described) in doing this for you,” (b) “Please rate the cost that would be involved in the scenario we gave you about taking a morning of work to receive and sign boxes on behalf of a friend,” and (c) “Please rate the cost that would be involved in the scenario about driving from Boston to Washington D.C.”

Qualitative coding

We devised a two-category mutually exclusive coding scheme to assess the types of reciprocations made by participants. Two coders (both blind to the hypotheses and the experimental conditions of the study) independently coded the open-ended data. We coded reciprocations as need based when participants reciprocated by responding to the needs of their friend, for example, “She needed to learn a few dance steps for a wedding, so taught her how to dance.” We coded reciprocations as non-need based when participants reciprocated by showing appreciation without responding to a specific need, for example, “I took her out for dinner to say thank you.” We obtained inter-rater reliability at 89.81% (k = .898, p < .001).

Results

Control analyses

We undertook a 2 (Culture) × 2 (Time Passage: short vs. long) × 2 (Benefit) mixed-design ANOVA on ratings of cost, with the within-factor variable being cost of benefits given and received: cost involved for the real-life helper vs the cost involved in fulfilling the hypothetical request for help made by the helper in the vignette. Results indicated nonsignificant effects of culture, F(1,26) = 1.645, p = .202, η² = .013; Culture × Time passage, F(1,126) = 3.536, p = .062, η² = .027; Culture × Benefit, F(1,126) = 0.681, p = .441, η² = .005; and Culture × Cost × Time passage, F(1,126) = 0.269, p = .605, η² = .002. Culture did not affect either the perceived cost of the real-life benefit received or of the help later requested by the helper in the vignette.

We ran a 2 × 2 factorial ANOVA to test whether culture influenced the timing of the real-life help received. No significant cultural differences occurred in how long ago the situations occurred in either condition, F(1,112) = 0.584, p = .446, η² = .000. We also ran a 2 × 2 chi-square analysis to test whether culture influenced whether individuals had reciprocated the help received. No cultural differences occurred on this measure, χ² = 0.493, p = .482 (India: 55.38%; the United States: 49.23%).

We ran a 2 × 2 chi-square analysis to assess the qualitatively different types of reciprocations (need vs. non-need based) made by Americans and Indians. We found that whereas Indians (47.22%) tended to make need-based reciprocations, Americans (94.11%) made non-need-based reciprocations, χ² (2, N = 70) = 15.11, p = < .001.

Obligation index

Since the observed alpha level was high (Cronbach’s alpha = .846), we calculated a mean composite score of the two dependent variables (likelihood of helping and guilt for declining) after transforming them into z scores. We refer to this composite score as the obligation index: High scores indicate greater obligation. A 2 (Culture) × 2 (Time Passage) × 2 (Reciprocation) factorial ANOVA, with gender as a covariate, on the obligation index, revealed significant effects of gender, F(1,120) = 9.957, p = .002, η² = .077; reciprocation, F(1,120) = 4.457, p = .037; η² = .036; culture, F(1,120) = 23.898, p <.001, η² = .166; Time passage × Reciprocation, F(1,120) = 8.679, p = .004, η² = .067; and Time passage × Reciprocation × Culture, F(1,120) = 9.698, p = . 002, η² = .075. The gender effect indicated that females (M = 0.282, SD = 0.911 ) scored higher on the obligation index than did males( M = −0.217, SD = 0.865).

In looking at time-passage differences among Americans, as predicted, we found that whereas Americans who had reciprocated showed no time-passage differences (months vs. years) on the obligation index, F(1,120) = 3.626, p = .059, η² = .029, Americans who had not reciprocated showed time-passage differences on the obligation index. Americans were more likely to help in the months rather than years condition after receiving a benefit, F(1,120) = 17.833, p <.001, η² = .129.

In looking at reciprocation differences within each time-passage condition, contrary to our hypotheses, Americans in the years condition did not differ on the obligation index whether or not they had reciprocated, F(1,120) = 3.556, p = .062, η² = .029. However, as predicted, Americans in the months condition differed in obligation depending on whether or not they had reciprocated: Americans who had not reciprocated were higher on the obligation index than those who had reciprocated, F(1,120) = 19.514, p <.001, η² = .140.

In looking at time-passage and reciprocation differences among Indians, as predicted, Indians showed no time-passage differences on the obligation index when they had reciprocated, F(1,120) = 0.377, p = .540, η² = .003, or when they had not reciprocated, F(1,120) = 0.152, p = .698, η² = .001, and no reciprocation differences in the years condition, F(1,120) = 1.711, p = .193, η² = .014, or in the months condition, F(1,120) = 0.1.390, p = .241, η² = .011.

In looking at cultural differences and similarities within each time-passage condition and reciprocation condition, Indians were higher than Americans on the obligation index in the months condition if they had reciprocated, F(1,120) = 15.610, p <.001, η² = .115, and in the years condition if they had not reciprocated, F(1,120) = 15.618, p < .001, η² = .115. However, Indians and Americans felt the same amount of obligation in the months condition when they had not reciprocated, F(1,120) = 0.502, p = .480, η² = .004, and in the years condition when they had reciprocated, F(1,120) = 1.152, p = .285, η² = .010 (see Table 2 for means).

Table 2.

Descriptive Statistics for the Obligation Index in Study 3.

	Americans				Indians
	Months		Years		Months		Years
	RM [95% CI]	NRM [95% CI]	RM [95% CI]	NRM [95% CI]	RM [95% CI]	NRM [95% CI]	RM [95% CI]	NRM [95% CI]
Likely to help	2.26 [1.67, 2.48]	4.04 [3.48, 4.61]	3.03 [2.39, 3.67]	2.09 [1.25, 2.94]	4.01 [3.47, 4.55]	4.46 [3.75, 5.18]	3.91 [3.23, 4.59]	4.33 [3.47, 4.55]
Guilt after declining help	2.43 [1.83, 3.03]	3.99 [3.41, 4.57]	3.20 [2.55, 3.86]	2.31 [1.44, 3.17]	3.64 [3.08,4.19]	4.18 [3.44, 4.91]	3.26 [2.56, 3.95]	3.97 [3.32, 4.63]

Note. All means presented in this table are the adjusted means with the covariate: gender. CI = confidence interval.

Using the Preacher and Hayes (2004) moderated moderation procedure (PROCESS MODEL 3), we investigated whether reciprocation moderated the relationship between time-passage and the obligation index for Americans but not for Indians. Adding the three-way interaction term (Time passage × Culture × Reciprocation) improved the model significantly R^2change = .048, F(1, 122) = 13.04, p = .004. In line with our hypotheses, reciprocation moderated the relationship between time passage and the obligation index for Americans, b = 1.646 SE = .412, t(122) = 3.995, p < .001, 95% CI [0.8304, 2.461]. Reciprocation did not moderate the relationship between time passage and obligation for Indians, b = −.0267, SE = .4051, t(120) = −.0658, p = .947, 95% CI [−0.828, 0.775] (see Figure 3).

Figure 3.

Study 3: (a) Reciprocation moderates the relationship between time passage and likelihood of helping for Americans and (b) Reciprocation does not moderate the relationship between time passage and likelihood of helping for Indians.

Discussion

The findings provide evidence for the ecological validity of our hypotheses regarding the existence of an expiration date for reciprocation and for the power of reciprocations in providing immunity from obligation to aid the helper among Americans. Equally, the findings provide evidence for the ecological validity of our hypotheses regarding the inconsequential nature of the passage of time and reciprocation on obligation to aid the helper among Indians. Americans, but not Indians, were less likely to aid the real-life helper when they had reciprocated and after a long passage of time.

Our findings highlight the interrelationship of the effects of reciprocation (leading to experience of immunity) and time passage (leading to experience of expiration) in affecting obligation to aid the helper for Americans only. Expiration only affected obligation toward the helper when Americans had not reciprocated and not when they had reciprocated. In addition, Americans showed greater obligation toward the helper when they had not reciprocated as compared with when they had reciprocated. However, these effects occurred in the short-term condition only. This unexpected result points to the importance of the cost of reciprocation. It is possible that only high cost reciprocations, such as spending three weekends uploading data onto a computer as in Study 1, are able to provide immunity to the receiver of help over extended periods of time. Study 3 did not assess the cost of the real-life reciprocations made by the participants, and thus, it is unclear whether individuals made high cost reciprocations or low cost reciprocations. However, it would be expected that reciprocations that are low in cost and non-personalized do not clear debt completely, but only reduce debt. The trends observed in Study 3 might then have reflected the impact of low cost reciprocations providing less immunity over time than high cost reciprocations.

These results provide insight into the roles of immunity and of expiration alone and in combination in affecting readiness to respond to the needs of the helper as well as point to other factors such as the cost of reciprocation that must be considered in the context of exchange and communal norms. Overall, the investigation showed how our theories of expiration and immunity work independently as well as simultaneously. Obligation to aid the helper is affected both by reciprocation and time passage; however, reciprocation holds more weight than does time passage. Reciprocations clear debt, whereas time passage reduces debt significantly. The study also showed how our theory of expiration and immunity does not apply when individuals apply communal norms. Giving of future benefits to helpers is resilient to reciprocation and time passage and is predicted most strongly by the helpers’ needs among individuals relying on communal norms.

General Discussion

We obtained support for the study hypotheses regarding the importance of time in influencing obligation to aid the helper among Americans, but not among Indians. Whereas Indians maintained a stable obligation toward the helper over time, Americans felt a heightened sense of obligation months after receipt of a benefit, which fluctuated depending on the passage of time and reciprocation. In testing our theory about the power of reciprocations made within the reciprocation time window in providing immunity over time, Study 1 (using a vignette-based experiment) and Study 3 (using a real-life interview) showed that American participants felt less obligated toward the helper after reciprocation as compared with cases in which they had not reciprocated, even years after receiving a benefit. Indian participants felt equally ready to meet the needs of the helper whether or not they had reciprocated even years after receiving an initial benefit. Reciprocations provided immunity from responding to the needs of the helper among Americans, but not among Indians. In testing our theory about obligation toward the helper expiring over time due to the closing of the reciprocation time window, Studies 2 (using a vignette-based experiment) and 3 (using a real-life interview) showed that American participants, presumably acting on the basis of exchange, felt less obligation to aid the helper years after receiving a benefit as compared with months after receiving a benefit. In contrast, Indians, presumably acting on the basis of communal norms, were equally responsive to the needs of the helper months or years after receiving a benefit. Obligation to respond to the needs of the helper expired over time among Americans but not among Indians.

Our findings also provided evidence that the observed cross-cultural differences did not arise merely from the more urbanized background of the U.S. as compared with Indian samples. The same types of cultural differences occurred both when utilizing geographically diverse online samples and samples from Mysore and the New York City area.

Reciprocity and Time

Our results point to the importance of time in influencing obligation toward a helper. A central feature of exchange norms is their time-bound nature, in which the giving of benefits creates obligation on the part of the receiver to make a return within a short-term time frame. This was observed in the tendencies of Americans to report higher likelihood of helping in the short term (months) than in the long term (years)—demonstrating that Americans felt more indebted immediately after receiving a benefit than after an extended passage of time.

Our results also point to the importance of the passage of time in highlighting commonalities between Indians and Americans in their readiness to aid the helper. In the short-term condition, Indians and Americans reported similar levels of obligation to aid the helper as well as reported similar levels of guilt about declining to aid the helper, and considered declining to help as equally unjustified. These findings demonstrate that when Americans acting on the basis of exchange norms feel a sense of indebtedness, they are as likely to aid the helper as are Indians acting on the basis of communal norms.

Taking into account time as a factor that affects obligation to aid the helper, thus, not only provides insight into the rules governing exchange norms in informal relationships but also highlights situations in which individuals acting on the basis of either exchange or communal norms may be equally ready to meet the needs of others. Thus, it may be concluded that individuals acting on the basis of exchange are not generally less ready to respond to the needs of the helper, but rather that obligation to attend to the needs of helpers is affected by factors such as the timing of the request and timely reciprocations.

Reciprocity and Social Solidarity

Social support theorists note that the giving of benefits entails uncertainty and risk as individuals help others without knowing if, when, or to what extent they will receive benefits in the future from the person whom they helped. In the case of communal norms, the act of being responsive to the needs of the giver reduces this uncertainty, thus building trust and social solidarity (Clark & Mills, 2012). The present investigation provides insight into the social solidarity that is created not only when individuals rely on communal norms but also when they rely on exchange norms. In assessing personality impressions of an individual who declines to help, we found that Americans were likely to form negative impressions of the protagonist when she declined to help months after, as compared with years after, receipt of a benefit. When the protagonist declines helping within the “reciprocation window”, that is, months after, Americans were more likely to view the protagonist as “selfish” and “self-centered” than if the protagonist had declined years after receipt of a benefit. These results highlight the importance of time-appropriate reciprocations in affecting the quality of relationships in the case of individuals relying on exchange norms.

The findings are also congruent with claims about the power of gestures of appreciation and gratitude in creating a sense of social solidarity in close relationships, in part, through reducing the levels of uncertainty that exist in such relationships (Algoe, 2012; Frederickson, 2004; Molm & Schaefer, 2007; Molm et al., 2012). Although social support theorists point to such a sense of gratitude as being associated exclusively with communal norms, the present findings suggest that gratitude may be evident in cases in which individuals are relying on exchange norms. In Study 3, Americans (presumably relying on exchange norms) qualitatively described reciprocations and returns in terms of expressive gestures or ways of showing appreciation. Our findings also support work by Molm and colleagues (Molm, Schaefer & Collett, 2007; Molm et al., 2012) who point to a form of exchange they label reciprocal exchange that is used in everyday informal relationships.⁷ We believe that it is this form of exchange that is adapted in informal relationships (i.e., nonmarket-based relationships) among Americans and that formal relationships (i.e., market-based relationships) involve negotiated exchange norms that are more rigid and explicit in nature. Since no prior agreement or understanding exists about the terms of the return when individuals engage in reciprocal exchange, the giving of benefit is able to create uncertainty and risk. The giver of benefits does not know when, how, or if the benefit that he or she gave will be reciprocated. The act of reciprocation eases this uncertainty, enhancing trust and social solidarity, which is a key component for the smooth functioning of informal social relationships.

Reciprocity and the Interdependent/Independent Self

Research has documented cultural variation in reliance on exchange versus communal norms involving interdependent cultures other than India. Shen, Fang, and Wyer (2011) found that East Asians were concerned with immediate indebtedness when exchanging gifts, implying that East Asians were applying exchange norms. Furthermore, Miller et al. (2017) found that Japanese and American populations tended to rely on exchange norms more frequently than did Indians, who tended to rely predominately on communal norms. The present findings then extend work by both Shen et al. (2011) and Miller et al. (2017) in adding to a body of evidence that supports variation in outlooks within interdependent cultures. Even though East Asian populations and Indian populations have collectivistic outlooks, they differ from each other in their reliance on exchange.

Future Directions

Our studies demonstrated that among Americans, readiness to reciprocate expires over time, and time-appropriate reciprocations provide immunity from needing to reciprocate. However, in understanding how long one has to reciprocate or for how long reciprocations provide immunity, one would have to consider both the cost of the initial help given by the helper as well as the cost involved in reciprocation. If the cost of help received is significantly greater than the reciprocation made, would receivers of help feel lingering indebtedness? Likewise, if the cost of the reciprocation is low, does one have more or less time before the need to reciprocate expires? To answer these questions and better understand the mechanisms of providing help, future research should investigate readiness to reciprocate to a helper’s needs as a product of all three constructs: time, reciprocation, and cost.

Conclusion

Our findings highlight the importance of reciprocity in influencing helping behavior. The reciprocity norms relied on by both Americans and Indians involved an element of risk and operated on the principle of trust. However, Americans and Indians were found to have different approaches to uncertainty reduction and trust building. For Americans, trust depends to a large extent on abiding by exchange norms, that is, by reciprocating to others in time-sensitive ways. In contrast, for Indians, trust depends on abiding by communal norms, that is, by responding to the needs of others whenever these needs may arise. These results are a testament to the truly adaptive nature of human behavior in showing how social networks function based on principles that reflect the structure of different societies. The need for autonomy and personal choice that is salient for Americans (Miller, Bersoff, & Harwood, 1990; Miller, Das, & Chakravarthy, 2011) involves the reliance on exchange norms that promote trust and social solidarity on an individual level. On the other hand, in societies such as India in which individuals tend to live in close knit communities involving strong ingroup ties, individuals tend to acquire a diffuse sense of interpersonal responsibility to family and friends (Appadurai, 1985; Miller & Bersoff, 1992; Miller et al., 1990; Savani, Morris, Naidu, Kumar, & Berlia, 2011) that is supported by communal norms and promotes trust and social solidarity among networks of individuals.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by National Science Foundation Grant SBE-0642333 to Joan G. Miller.

Notes

References

Algoe

S. B.

(2012). Find, remind, and bind: The functions of gratitude in everyday relationships. Social and Personality Psychology Compass, 6, 455-469. doi:10.1111/j.1751-9004.2012.00439.x

Appadurai

(1985). Gratitude as a social mode in South India. Ethos, 13, 236-245. doi:10.1525/eth.1985.13.3.02a00020

Batson

(1993). Communal and exchange relationships: What is the difference? Personality and Social Psychology Bulletin, 19, 677-683.

Burger

J. M.

Sanchez

Imberi

J. E.

Grande

L. R.

(2009). The norm of reciprocity as an internalized social norm: Returning favors even when no one finds out. Social Influence, 4, 11-17. doi:10.1080/15534510802131004

Clark

M. S.

Aragón

O. R.

(2013). Communal (and other) relationships: History, theory development, recent findings, and future directions. In Simpson

J. A.

Campbell

(Eds.), The Oxford handbook of close relationships (pp. 255-280). Oxford, UK: Oxford University Press. doi:10.1093/oxfordhb/9780195398694.013.0012

Clark

M. S.

Lemay

E. P.

Graham

S. M.

Pataki

S. P.

Finkel

E. J.

(2010). Ways of giving benefits in marriage: Norm use, relationship satisfaction, and attachment-related variability. Psychological Science, 21, 944-951. doi:10.1177/0956797610373882

Clark

M. S.

Mills

J. R.

(1979). Interpersonal attraction in exchange and communal relationships. Journal of Personality and Social Psychology, 51, 333-338.

Clark

M. S.

Mills

J. R.

(2012). A theory of communal (and exchange) relationships. In Van Lange

P. A. M.

Kruglanski

A. W.

Higgins

E. T.

(Eds.), Handbook of theories of social psychology (Vol. 2, pp. 232-250). Thousand Oaks, CA: SAGE. doi:10.4135/9781446249222.n38

Clark

M. S.

Mills

J. R.

Corcoran

D. M.

(1989). Keeping track of needs and inputs of friends and strangers. Personality & Social Psychology Bulletin, 15, 533-542. doi:10.1177/0146167289154007

10.

Clark

M. S.

Mills

J. R.

Powell

M. C.

(1986). Keeping track of needs in communal and exchange relationships. Journal of Personality and Social Psychology, 51, 333-338. doi:10.1037/0022-3514.51.2.333

11.

Clark

M. S.

Ouellette

Powell

M. C.

Milberg

(1987). Recipient’s mood, relationship type, and helping. Journal of Personality and Social Psychology, 53, 94-103. doi:10.1037/0022-3514.53.1.94

12.

Clark

M. S.

Waddell

(1985). Perceptions of exploitation in communal and exchange relationships. Journal of Social and Personal Relationships, 2, 403-418. doi:10.1177/0265407585024002

13.

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155-159. doi:10.1037//0033-2909.112.1.155

14.

Faul

Erdfelder

Buchner

Lang

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149-1160. doi:10.3758/brm.41.4.1149

15.

Frederickson

(2004). The broaden-and-build theory of positive emotions. Philosophical Transactions of the Royal Society B, 359, 1367-1377. doi:10.1098/rstb.2004.1512

16.

Gouldner

A. W.

(1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161-178.

17.

Greenberg

M. S.

(1980). A theory of Indebtedness. In Gergen

K. J.

Greenberg

M. S.

Willis

R. H.

(Eds.), Social exchange: Advances in theory and research (pp. 3-26). New York, NY: Plenum Press.

18.

Groves

R. M.

Cialdini

R. B.

Couper

M. P.

(1992). Understanding the decision to participate in a survey. Public Opinion Quarterly, 56, 475-495. doi:10.1086/269338

19.

Hartup

W. W.

Stevens

(1997). Friendships and adaptation in the life course. Psychological Bulletin, 121, 355-370. doi:10.1037//0033-2909.121.3.355

20.

McCullough

M. E.

Kimeldorf

M. B.

Cohen

(2008). An adaptation for altruism: The social causes, social effects and social evolution of gratitude. Current Directions in Psychological Science, 17, 281-285.

21.

Miller

J. G.

Akiyama

Kapadia

(2017). Cultural variation in communal vs exchange and their implications for comfort in social support. Journal of Personality and Social Psychology, 113, 81-94. doi: 10.1037/pspi0000091.

22.

Miller

J. G.

Bersoff

D. M.

(1992). Culture and moral judgment: How are conflicts between justice and interpersonal responsibilities resolved? Journal of Personality & Social Psychology, 62, 541-554. doi:10.1037/0022-3514.62.4.541

23.

Miller

J. G.

Bersoff

D. M.

Harwood

R. L.

(1990). Perceptions of social responsibilities in India and in the United States: Moral imperatives or personal decisions? Journal of Personality and Social Psychology, 58, 33-47. doi:10.1037/0022-3514.58.1.33

24.

Miller

J. G.

Bland

Källberg-Shroff

Tseng

Montes-George

Ryan

. . . Chakravarthy

(2014). Culture and the role of exchange vs. communal norms in friendship. Journal of Experimental Social Psychology, 53, 79-93. doi:10.1016/j.jesp.2014.02.006

25.

Miller

J. G.

Das

Chakravarthy

(2011). Culture and the role of choice in agency. Journal of Personality and Social Psychology, 101, 46-61. doi:10.1037/a0023330

26.

Mills

Clark

M. S.

(1982). Communal and exchange relationships. Review of Personality and Social Psychology, 3, 121-144.

27.

Mills

Clark

M. S.

Ford

T. E.

Johnson

(2004). Measurement of communal strength. Personal Relationships, 11, 213-230. doi:10.1111/j.1475-6811.2004.00079.x

28.

Molm

L. D.

(2010). The structure of reciprocity. Social Psychology Quarterly, 73, 119-131. doi:10.1177/0190272510369079

29.

Molm

L. D.

Schaefer

D. A.

Collett

J. L.

(2007). The value of reciprocity. Social Psychology Quarterly, 70, 199-217. doi:10.1177/019027250707000208

30.

Molm

L. D.

Whitham

M. M.

Melamed

(2012). Forms of exchange and integrative bonds: Effects of history and embeddedness. American Sociological Review, 77, 141-165. doi:1177/0003122411434610

31.

Muthen

Asparauhov

(2015). Causal effects in mediation modeling: An introduction with applications to latent variables. Structural Equation Modeling: A Multidisciplinary Journal, 22, 12-23. doi:10.1080/10705511.2014.935843

32.

Oyserman

Coon

H. M.

Kemmelmeier

(2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-72. doi:10.1037/0033-2909.128.1.3

33.

Preacher

K. J.

Hayes

A. F.

(2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717-731. doi:10.3758/BF03206553

34.

Savani

Morris

Naidu

N. V. R.

Kumar

Berlia

N. V.

(2011). Cultural conditioning: Understanding interpersonal accommodation in India and the United States in terms of the modal characteristics of interpersonal influence situations. Journal of Personality and Social Psychology, 100, 84-102. doi:10.1037/a0021083

35.

Shen

Fang

Wyer

R. S.

(2011). Cross cultural differences in the refusal to accept a small gift: The differential influence of reciprocity norms on Asians and North Americans. Journal of Personality and Social Psychology: Interpersonal Relations and Group Processes, 100, 271-281. doi:10.1037/a0021201

36.

Trivers

R. L.

(1971). The evolution of reciprocal altruism. The Quarterly Review of Biology, 46, 35-57.

37.

Watkins

Scheer

Ovnicek

Kolts

(2006). The debt of gratitude: Dissociating gratitude and indebtedness. Cognition & Emotion, 20, 217-241. doi:10.1080/02699930500172291