Introduction
Waste sorting and recycling (WSAR) has grown increasingly important due to worldwide economic and environmental concerns related to the conservation of energy and environmental protection (Xu et al., Reference Xu, Elomri, Pokharel, Zhang, Ming and Liu2017; Huang et al., Reference Huang, Law, Geng, Niu and Kettunen2022). Many cities worldwide have implemented sorting policies and encouraging public participation in recycling. For example, as one of the world’s largest producers of municipal solid waste (MSW), China has enacted a variety of new initiatives in recent years, such as promoting a circular economy and implementing MSW classification. However, these efforts have yielded limited success and the public has even expressed negative attitudes toward the MSW sorting policies (Zhang et al., Reference Zhang, Zhang, Yu and Ren2016; Xiao et al., Reference Xiao, Dong, Geng and Brander2018; Wu et al., Reference Wu, Zhang, Chen and Wang2021). The COVID-19 pandemic has presented an extra challenge for Chinese and even global waste management, with all the talk of waste sorting seeming to fade away (Zhou et al., Reference Zhou, Yang, Ma, Liu and Zhao2021; Kahlert and Bening, Reference Kahlert and Bening2020). Thus, it is imperative to identify effective ways to promote public WSAR behavior.
Similar to other sustainable behaviors, recycling will lead to preferable environmental outcomes in the future (Wittmann and Sircova, Reference Wittmann and Sircova2018). In other words, sustainable development through recycling is a long-term goal and benefits future generations, which appears to be a form of intergenerational beneficence (Wade-Benzoni et al., Reference Wade-Benzoni, Hernandez, Medvec and Messick2008). For example, if we spend time on waste sorting today and in the future, people will benefit from improved (or just no longer worse) air and water quality. Such decisions pertaining to the allocation and consumption of resources are often considered to be intergenerational in nature, i.e., intergenerational decision making. These decisions present an intergenerational dilemma, where the priorities of current decision-makers may potentially clash with the needs of future generations (Wade-Benzoni, Reference Wade-Benzoni2008; Wade-Benzoni and Tost, Reference Wade-Benzoni and Tost2009; Tost and Wade‐Benzoni, Reference Tost and Wade‐Benzoni2012).
Unlike decisions under other social dilemmas, the implications of intergenerational decisions are both temporally distant and personally detached from the decision-maker (Wade-Benzoni and Tost, Reference Wade-Benzoni and Tost2009). The integration of intertemporal and interpersonal dimensions in the intergenerational context results in a compounded force that acts in opposition to intergenerational beneficence (Wade-Benzoni, Reference Wade-Benzoni2019). Consequently, the tradeoff in intergenerational decision making is neither based solely on intertemporal considerations nor exclusively on self-other comparisons. Instead, it resembles a tradeoff between the present self and future others. The psychological distance between the decision maker and future generations is identified as a critical barrier to intergenerational beneficence (Wade-Benzoni et al., Reference Wade-Benzoni, Tost, Hernandez and Larrick2012).
Wade-Benzoni (Reference Wade-Benzoni1999) first views environmental protection as an intergenerational dilemma, as present resource consumption and waste disposal may have future consequences. Overfishing is one such example where the current generation’s actions of catching as much fish as they can to maximize benefit in the present may lead to a shortfall of fish in the future, which would burden future generations. In this situation, reducing consumption means providing resource for future generations. Recycling, as defined by the United States Environmental Protection Agency, is ‘the process of collecting and processing materials that would otherwise be thrown away as trash and turning them into new products’. The immediate purpose of WSAR is to reduce waste and promote resource reuse. Although the relationship between recycling and maintaining the interests of future generations is not as direct as that between recycling and sustainable fishing, few studies have considered this issue from an intergenerational perspective. For example, Bagozzi and Dabholkar (1994) explored the consumer goals behind recycling, among which saving the environment and providing for future generations are central goals. Petts (Reference Petts2000) reviewed the intergenerational risk of MSW. The impacts of MSW can extend over hundreds of years, especially in the case of landfills. Taebi and Kloosterman (Reference Taebi and Kloosterman2008) considered the choice between open and closed nuclear fuel cycles to be intergenerational decision making since the latter involves less radiological risk to future generations but may also pose threats for contemporary people (Taebi et al., Reference Taebi, Roeser and van de Poel2012; Vilhunen et al., Reference Vilhunen, Kojo, Litmanen and Taebi2022). As a result, people may engage in WSAR out of concern for future generations. Strangely, however, few studies have considered promoting recycling behavior in terms of this intergenerational characteristics.
Indeed, a series of studies have regarded strengthening temporal proximity as a powerful strategy to fostering pro-environmental behavior (Pahl et al., Reference Pahl, Sheppard, Boomsma and Groves2014; McDonald et al., Reference McDonald, Chai and Newell2015). Reducing intertemporal distance does not consistently increase policy support and behavioral intentions related to environmental protection (Brügger et al., Reference Brügger, Dessai, Devine-Wright, Morton and Pidgeon2015). For example, Schuldt et al. (Reference Schuldt, Rickard and Yang2018) framed climate change as proximal but decreased psychological distance did not translate into a high level of policy support. Böhm and Pfister (Reference Böhm and Pfister2005) observed only a weak effect of subjective time perspective on environmental risk perception. Similarly, Bø and Wolff (Reference Bø and Wolff2020) found that episodic future thinking may not influence the risk perception of climate change or the willingness to donate to environmental organizations, although it facilitates the perception of climate change as closer to the present self. Carmi (Reference Carmi2013) also indicated that highly future-oriented respondents did not exhibit more pronounced pro-environmental attitudes. In fact, their willingness to make sacrifices for the environment was significantly lower.
The inconclusive results may be attributed to an overemphasis on the intertemporal dimension and oversight of the interpersonal dimension. From the perspective of intergenerational choice, pro-environmental behavior, including WSAR, has a combination of future-oriented (futureness) and other-oriented (otherness) attributes (Wade-Benzoni, Reference Wade-Benzoni2019). Intergenerational discounting decreases significantly only when the psychological distance is low both intertemporally and interpersonally (Wade-Benzoni, Reference Wade-Benzoni2008; Pearson et al., Reference Pearson, Tsai and Clayton2021). Eliciting concerns for future others may be a critical factor to simultaneously overcoming shortsightedness and selfishness in the transformation toward sustainability.
Recently, incorporating interpersonal negotiation into environmental research has attracted increasing interest (Majer et al., Reference Majer, Barth, Zhang, van Treek and Trötschel2021, Reference Majer, Zhang, Zhang, Höhne and Trötschel2022). A small but growing number of researchers are attempting to promote pro-environmental behavior by eliciting concerns for future others. For example, future design is an emerging methodology in the field of sustainability science, in which problem solvers are asked to take the perspective of and act as members of future generations before making their decisions (Pandit et al., Reference Pandit, Nakagawa, Timilsina, Kotani and Saijo2021). Some qualitative case studies have demonstrated that people who experience the perspective of future generations tend to prefer sustainable policy options (Hara et al., Reference Hara, Yoshioka, Kuroda, Kurimoto and Saijo2019; Nakagawa et al., Reference Nakagawa, Arai, Kotani, Nagano and Saijo2019). Furthermore, empirical studies have demonstrated that certain manipulations, such as legacy motive priming (Zaval et al., Reference Zaval, Markowitz and Weber2015; Wickersham et al., Reference Wickersham, Zaval, Pachana and Smyer2020) and immersive virtual reality (Hershfield et al., Reference Hershfield, Goldstein, Sharpe, Fox, Yeykelis, Carstensen and Bailenson2011), can effectively prime concerns for future generations, thereby increasing engagement in environmentally conscious behaviors.
However, to our knowledge, limited research has explored the relationship between concerns for future others and recycling behavior. Can stimulating concerns for future others promote recycling behavior? Futhermore, certain pro-environmental interventions tend to be domain-specific and may not be generalizable to other green behaviors (Lange and Dewitte, Reference Lange and Dewitte2019; Yuriev et al., Reference Yuriev, Dahmen, Paillé, Boiral and Guillaumie2020). While future-others-oriented interventions are generally effective in contexts such as increasing pro-environmental intentions and shifting policy preferences, they may encounter challenges when applied to specific WSAR behaviors. Inspired by the perspective of intergenerational choice, the current study aimed to examine the potential of concerns for future others in a specific pro-environmental situation – WSAR. This research contributes to the literatire in two ways: First, we introduce the perspective of intergenerational decision-making into a specific pro-environmental behavior (i.e., WSAR). Through three studies, we will examine whether future-others-oriented interventions, such as imagining future people and interacting with virtual future people, could induce concern for those in the future and in turn encourage individuals to recycle more. Then, we will compare the difference in the effects of imagining the future and imagining future others. The latter has received limited attention in prior research (Wang et al., Reference Wang, Leviston, Hurlstone, Lawrence and Walker2018; Lau et al., Reference Lau, Song, Morrison, Fabinyi, Brown, Blythe and Adger2021), but it warrants consideration due to the intergenerational-oriented characteristics of recycling behavior.
In summary, we have proposed that interventions oriented toward future people would induce concern for those in the future and in turn encourage individuals to be greener. According to the intergenerational decision-making model, we hypothesize that individuals would exhibit greater recycling intention and more recycling behavior when their concerns for future others are elicited (H1). Second, merely reducing the temporal gap between the present and the future may not be adequate to motivate people to act in the interest of future generations. We also hypothesize that there is a difference between future other-imagining and future-imagining conditions, as individuals in the former condition are more likely to recycle (H2). We conducted two main studies, employing different techniques to elicit concern for future others and measure the outcomes. A pilot study provided an initial demonstration of the relationship between concerns for future others and pro-environmental behavioral intention. Study 1A tested the main effect of eliciting concern for future others on intentions to recycle and policy support. Study 1B sought to replicate the findings of Study 1A by employing a measure of actual behavior. Study 2A and Study 2B examined Hypothesis 2 by adding a future-imagining condition.
Pilot study
This study aimed to provide evidence regarding the effectiveness of our manipulation, which served as the basis for the follow-up study. Additionally, this study aimed to explore the relationship between concerns for future generations and general intentions to engage in pro-environmental behavior.
Methods
Participants
We recruited 318 adults from a Chinese survey platform (Wenjuanxing: https://www.wjx.cn). Those who did not pass the attention test were automatically rejected by the system. We dropped 27 participants from the sample because they failed to write meaningful sentences in the manipulation task (described below), leading to a final sample size of 291 (M age = 29.29, SD = 4.61, age range = 21–46, 48.11% female). Informed consent was granted by all participants. This study employed a single-factor between-participants experimental design. The independent variable was whether concerns for future others were elicited.
Procedure and measures
After the participants viewed the instructions, they were randomly assigned to one of two conditions: 1) a future-other condition: imagining and writing about the positive influence they could leave on future others in the next 20 years (final n = 141); and 2) a control condition: writing about their current environment (final n = 150). The content had to be no less than 50 words. This task was adapted from Zaval et al. (Reference Zaval, Markowitz and Weber2015) to make people psychologically close to future generations.
Participants were then asked to assess their general concern about others and themselves in the future on a 7-point Likert scale (1 = not at all, 7 = very much). Subsequently, participants completed measures of their pro-environmental behavioral intentions. They rated their willingness to engage in environmental actions over the next three months (e.g., ‘Reusing plastic bags’) on a scale ranging from 1 (never) to 5 (always) (adapted from Gong, Reference Gong2008). A total of ten items were averaged to create a score for each participant’s general pro-environmental behavioral intention (α = 0.68).
In addition to the primary measures of interest, we asked all participants to report their pro-environmental orientations using the 15-item new ecological paradigm (NEP), which served as a control variable in the analysis. NEP is considered to represent a belief and concern for the environment and has been widely used to measure pro-environmental orientation (Dunlap et al., Reference Dunlap, Van Liere, Mertig and Jones2000; Dunlap, Reference Dunlap2008). Ratings were reported on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree), with a Cronbach’s α of 0.69. Finally, the participants provided their demographic information.
Results and discussion
With respect to the manipulation check, the rating of concerns for future others was higher under the future-other condition (M future-other = 4.25, SD = 1.34) than under the control condition (M control = 3.36, SD = 1.48), t(289) = 5.34, p < 0.001, Cohen’s d = 0.63, 95% CI [0.40, 0.86]. In addition, ANCOVA on participants’ willingness to engage in pro-environmental behavior (with pro-environmental orientations and future self-concern as covariates) revealed a significant main effect of priming for future other-concern, F(1, 287) = 11.02, p = 0.001, partial η 2 = 0.04, 95% CI [0.01, 0.08]. Compared with participants in the control condition, participants in the future-other condition manifested greater pro-environmental intentions (M future-other = 3.67, SD = 0.48, M control = 3.50, SD = 0.51). In addition, we analyzed two items related to the willingness to recycle in the measure (i.e., ‘Place used batteries in the battery recycling bin’ and ‘Separate household waste for recycling’). Similarly, ANCOVA showed a significant difference between the two conditions, F(1, 287) = 5.64, p = 0.018, partial η 2 = 0.02, 95% CI [0.002, 0.05], and people under the condition in which their concerns for future others were elicited reported greater willingness to recycle (M future-other = 3.66, SD = 0.90) than those under the control condition (M control = 3.44, SD = 1.01).
This pilot study provided preliminary evidence that elicitation of future other-concern promotes intentions to engage in general pro-environmental behavior, including intentions to recycle. However, such a general pro-environmental intention may not represent specific intentions to recycle. Study 1 aimed to examine the effect of concerns for future others on recycling intentions through more specific outcome measures.
Study 1A
Study 1A tested our primary hypothesis that eliciting concerns for future others increases the intention to recycle. In line with the pilot study, Study 1A utilized a between-participants design with a single factor (the presence or absence of an intervention to stimulate future other-concern). Furthermore, the envisioned persons were specified as future strangers to more clearly define the term ‘others’.
Methods
Participants
To achieve 80% power and to detect a small to medium effect (partial η 2 = 0.02 in the pilot study), a minimum sample size of 387 participants was predetermined using G*Power (Faul et al., Reference Faul, Erdfelder, Lang and Buchner2007). We recruited participants online using the Wenjuanxing website with the following inclusion criteria: i) participants passed the instructional attention check; ii) participants wrote meaningful sentences in the manipulation task; and iii) participants did not participate in the pilot study. The final sample included four hundred participants (M age = 31.29, SD = 8.17, age range = 18–68, 51.75% female). For more details on the demographic characteristics of all the participants, please see Table S1 in supplemental materials. Informed consent was granted by all participants.
Procedure and measures
Participants were told that they would participate in a recycling behavior survey. At the start of the survey, participants were randomly assigned to either a future-other condition or a control condition. In the future-other condition (final n = 207), participants were instructed to imagine future strangers. On the next page, they were asked to write down the positive influence they could have on future strangers in the next 20 years, whereas participants in the control condition (final n = 193) were only encouraged to relax and wait for 15 seconds. On the next page, one question was used as an attention check for participants in the future-other condition: According to the instructions, how many years in the future did you envision strangers?
After the manipulation, participants assessed their recycling intention in the next month. The measure with three items was adapted from previous studies (Wan et., Reference Wan, Shen and Choi2017; Wan et al., Reference Wan, Shen and Choi2021): 1) I intend to recycle my recyclables in the next four weeks; 2) I will recycle my recyclables every time I have them; and 3) I am willing to participate in recycling schemes in the future. Responses were reported on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). The Cronbach’s α was 0.65. The participants were then given a hypothetical environmental scenario that informed them of a local policy about WSAR:
In recent years, China has achieved some success in waste sorting, but there is still a need to standardize the way to sort and recycle waste. If your community plans to conduct courses on waste sorting and recycling, you will need to attend the course in your community and be assessed at the end of the course, the failure of which will result in retraining.
First, we asked participants about their willingness to participate in waste sorting training (1 = strongly disagree; 7 = strongly agree). Second, we asked them for the maximum acceptable course duration. Responses were made on six options from 0 (indicating no willingness to take the course) to 60 minutes or longer.
In addition to the primary measures of interest, we also asked participants to report their general attitudes toward recycling (1 = not useful, 7 = useful) and past recycling experiences (1 = not at all, 7 = always), which served as control variables in the analyses.
Results and discussion
An ANCOVA of participants’ willingness to recycle revealed a significant main effect of future other-concern elicitation, with attitude toward recycling and past recycling experience as covariates, F(1, 396) = 4.20, p = 0.041, partial η 2 = 0.01, 95% CI [0.00, 0.03]. That is, participants who mentally envisioned future strangers reported greater willingness to recycle (M future-other = 6.10, SD = 0.77) than those in the control condition (M control = 5.90, SD = 0.83). Figure 1 illustrates the results.
Recycling intention and acceptance of recycling course by conditions in Study 1A. Note. Error bars indicate standard error of the mean. *p < .05. **p < .01. ***p < .001.
Analysis of the acceptance of the recycling course yielded similar results. The main effect of future other-concern elicitation was statistically significant when the attitudes toward recycling and past recycling experience were controlled for, F(1, 396) = 3.98, p = 0.047, partial η 2 = 0.01, 95% CI [0.00, 0.03]. Participants in the future-other condition (M future-other = 5.85, SD = 1.27) were more likely to participate in waste sorting and recycling training compared to those in the control condition (M control = 5.54, SD = 1.29). In terms of the maximum acceptable course duration, we ran a logistic regression on a dummy variable for participants’ choices (1 = 60 minutes or longer vs 0 = shorter than 60 minutes), with attitudes toward recycling and past recycling experience as covariates. In the condition that evoked future other-concern, there were slightly more people willing to spend longer time on waste sorting and recycling (38%) than in the control condition (31%). However, this difference was statistically insignificant (b = 0.27, SE = 0.21, p = 0.201, 95% CI [−0.15, 0.70]).
Study 1A provided initial evidence that eliciting concerns for future others increases recycling intentions, including the intention to engage in specific recycling behaviors and the intention to be trained. We did not observe a statistically significant difference in the acceptable length of the training program. This may be because the option given involved a lengthy temporal interval (30 min), which limited our ability to detect differences. In Study 1B, we further examined the effect of future other-concern on recycling in the context involving personal loss.
Study 1B
Like Study 1A, Study 1B employed a between-participants design with a single factor (the presence or absence of elicitation of future other-concern). We sought to replicate Study 1A’s findings by (1) changing the manipulation of future other-concern from the static writing task to a dynamic virtual manipulation. Previous research has shown that one reason for discounted future outcomes is that they are perceived less vividly (Rosch et al., Reference Rosch, Stramaccia and Benoit2021; Lee et al., Reference Lee, Parthasarathi, Cooper, Zauberman, Lerman and Kable2022). The aging progression manipulation used in Study 1B not only vividly brings decision-makers closer to future others but also provides a direct representation of strangers in the future, thereby mitigating the potential uncontrollability of imaging. (2) Modifying the outcome measure to include willingness to pay for recyclable issues as well as actual recycling behavior, resulting in greater ecological validity than in Study 1A.
Methods
Participants
We recruited subjects from a Chinese university subject pool. A total of 192 first-year graduate students completed the lab experiment in exchange for ¥20 (equal to $2.9). However, we found that 16 students learned about the content of our experiment from previous participants. Thus, our final sample consisted of 176 students (M age = 22.50, SD = 0.97; age range = 21–28, 67.05% female).
Procedure and materials
After the participants arrived at the laboratory and provided informed consent, they were randomly assigned to either the future-other condition or control condition (n = 88 in each condition). The participants were instructed to complete a series of tasks, one of which involved watching the screen of an iPad Pro (11 inches) in front of them. In the future-other condition, they viewed an old version of the experimenter 20 years from now (the experimenter sat opposite to the participants), while those in the control condition were asked to look at the lab environment only through the iPad camera. The people were virtually aged using an application named Oldify (http://www.oldify.net/), which can take photos acquired through the camera and simulate realistic face animations. Student participants in the future-other condition viewed the experimenter with wrinkles and grey hair (See Figure S1 in the supplemental materials). The aging progression manipulation was revised based on the study of Hershfield et al. (Reference Hershfield, Goldstein, Sharpe, Fox, Yeykelis, Carstensen and Bailenson2011), which aimed to increase connectedness with people in the future. Then, participants in the future-other condition were asked to rate their level of concern for the elderly people presented on the screen. In contrast, those in the control condition rated the clarity of the camera. Additionally, all participants completed a single-item measure of future self-concern as in the pilot study, i.e., how much they concerned about their future selves 20 years from now. The ratings ranged from 1 (not at all) to 7 (very much).
While the subject viewed the screen, the main experimenter read an environmental policy related to express package charging and asked for the subject’s approval of the policy (0 = completely disapprove, 100 = completely approve), as higher approval indicated a greater willingness to pay for the packaging box. The policy context was as follows:
The waste and pollution caused by excessive express packages are becoming increasingly worse, making environmental governance more challenging. To this end, the municipal government plans to charge an additional small fee for the recyclable cardboard, which will be used to govern environmental pollution. The provisional charging standard is as follows: the minimum fee is 1 yuan, and there is an additional fee of 1 yuan per additional 2.5 kg. Packaging boxes with low weight but high volume will be billed by converted weight (kg). Converted weight = length (cm) × height (cm) × width (cm)/6000. The municipal government is currently consulting the public on this policy. Please rate how strongly you support or oppose this policy.
Then, all participants were instructed to draw a simple painting of what was presented on the iPad screen, such that those in the future-other condition drew the presented aged face and those in the control condition drew the environment of the lab. The final task was to cut their drawings from the paper and glue them to the questionnaire, thus generating paper waste. When the participant had finished, the experimenter reminded each participant to take their paper waste away. As shown in Figure 2, there was a trash bin close to the door and a recycling bin farther away from the door. The outcome measure of recycling behavior was whether the participants walked a few extra feet to recycle their scrap paper (adapted from Whillans and Dunn, Reference Whillans and Dunn2015). Notably, the experimenter always had his back to the trash bins, waiting for the participants to leave before checking the bins and cleaning them.
Top view of the laboratory environment.
Results and discussion
Concerning the level of concern for the virtually aged people, the mean score for participants in the future-other condition was 5.14 (SD = 1.25), which was significantly higher than the midpoint of the scale (mean value = 4, p< 0.001). This suggests that the dynamic virtual manipulation of future other-concern was successful. A linear regression analysis was conducted to examine the effect of manipulation (1 = future-other, 0 = control) on participants’ approval of the charging policy while controlling for average consumption. Supporting our Hypothesis 1, the results indicated that future other-concern elicitation significantly affected policy approval, with future self-concern and monthly expenses as the covariate (M future-other = 70.93, SD = 19.85; M control = 64.60, SD = 22.23), b = 0.32, SE = 0.15, p = 0.033, 95% CI [0.03, 0.62].
We categorized the recycling choices into three categories: (1) used the recycle bin, (2) used the general waste bin and (3) did not use the bin (some individuals just took the paper away when leaving the classroom). As shown in Table 1, the distribution of recycling behavior (vs nonrecycling behavior) significantly differed between our two conditions. Compared with participants in the control condition, participants who viewed the aged version of another person were more likely to use the recycle bin (
${{{\chi }}^2}$ = 5.83, p = 0.016, OR = 2.09) and less likely to use the general waste bin (
${{{\chi }}^2}$ = 12.19, p < 0.001, OR = 2.96). Study 1B replicated the results of Study 1A in a lab experiment in which eliciting future other-concern was effective in promoting recycling behavior. Additionally, a logistic regression was conducted on participants’ recycling choices with the dummy variable of condition (1 = recycling bin vs 0 = general waste or take away) as the dependent variable with future self-concern as a control variable. The results were similar to those of the tests on distribution differences. The mental simulation of recycling had a significant effect on recycling choice (b = 0.72, SE = 0.31, p = 0.021, 95% CI [0.12, 1.34]) such that participants were more likely to recycle the paper case in the future-other condition (61.36%) than in the control condition (43.18%).
Results of nonparametric tests on recycling behavior
Policies acceptance (without and with monetary penalty) by conditions in Study 2B.
Study 1B utilized an aging simulation process to increase the individual’s level of concern for future others. The results of Study 1B replicated those of Study 1A. Compared to participants in the control condition, those who viewed the aged version of other people reported a greater willingness to pay for the packaging box and were more likely to throw the paper in the recycling bin while taking into account concern for future self. The behavior measure used in this study was more closely aligned with real-life recycling behavior than other questionnaire measures. This is because the recyclable bin was positioned farther away than the nonrecyclable bin and required more effort to access, while recycling in real life also involves personal cost. However, ‘aging’ the experimenter in Study 1B might lead participants to consider a future involving themselves and others, even with the inclusion of future self-concern as a control variable. This approach might deviate from the concept of future others or future generations as defined by intergenerational decision-making studies. Wade-Benzoni (Reference Wade-Benzoni2008) highlighted that a critical feature distinguishing intergenerational dilemmas from traditional social dilemmas is the removal of the decision-makers from the collective after their decision. Therefore, Studies 2A and 2B aimed to further test our hypothesis by employing manipulations specifically oriented toward future generations.
Study 2A
Study 1 was useful for establishing an initial link between future other-concern and recycling behavior. However, the term ‘future others’ encompasses not only older future generations relative to the present, but also younger cohorts such as infants or those yet to be born. In Study 2A, we presented an image of a newborn baby to prime participants’ concerns for future generations and included a control condition that only imagined the future. In addition, the manipulation tasks for the control conditions in Study 1 were directed to the present, and we considered the role of imagining the future in Study 2A. Specifically, we compared the effect of imagining the future vs imagining future others.
On the other hand, we developed hypothetical environmental policies in Study 1. In Study 2A, we examined the impact of future other-concern on the acceptance of a real policy called the ‘Certain Time Certain Place (CTCP)’ program. The CTCP policy stipulates that only designated times and locations are allowed for waste disposal, and waste should be collected in four bins (i.e., food waste bins, residual waste bins, hazardous waste bins and recyclables bins). This policy has been implemented in Shanghai since 2019 and is being adopted by other cities in China (SLCAAB, 2019).
Methods
Participants
Due to the difference in waste sorting polices implemented by each province, we collected data only from Guangzhou, Guangdong Province, China. Guangzhou is 1 of the 46 pilot cities in China that required to implement mandatory waste classification policy according to the 2017 nationwide plan. In response, Guangzhou Municipality has introduced a series of interventions, including the implementation of the CTCP waste sorting policy in some communities. We recruited adult residents of Guangzhou via the Credemo platform, a Chinese data collection platform (https://www.credamo.com/). The inclusion criteria closely align with those employed in Study 1A, with the addition of the following two stipulations: (i) participants resided in Guangzhou and (ii) the community in which they resided did not implement the CTCP policy. We dropped five participants because they reported that their current residence was not Guangzhou, leading to a final sample size of 467 (M age = 28.40, SD = 8.21, age range = 18–63, 62.31% female). The demographic characteristics of the participants are displayed in supplemental Table S2. All participants received a small reward via the platform.
Procedure and materials
Participants were informed that they would be taking part in a public opinion poll regarding the ‘CTCP’ waste sorting policy. We introduced the policy, which involves depositing waste in a specific location at a specific time and sorting waste into four categories and then checked whether participants understood the policy and whether their place of residence had implemented it. Those who correctly understood the policy and were not required to comply with it proceeded to the next page. Participants were randomly assigned to one of three conditions, a future-other condition, a present-other condition or a future-imaging condition. In the future-other condition (n = 155), participants were provided with a picture of a newborn (See Figure S2 in the Supplementary Materials ). The picture stimulus was selected through a pilot survey, with a result of neutral valence, neutral affinity and gender estimation between boy and girl. They were instructed to imagine and write about how the newborn would be 20 years later. In contrast, participants in the present-other condition (n = 158) were provided with the same picture but were instructed to imagine and write about the newborn’s life in the present. In the future-imaging condition (n = 154), we instructed participants to imagine and write about how the world will be in 20 years. As a manipulation check, then, participants were asked to what extent they were concerned about and felt close to future individuals (1 = not at all to 7 = very much), r = 0.57.
After the manipulation, we asked participants to estimate their willingness to recycle in the next month, as in Study 1A (Cronbach’s α = 0.80). Then, participants read the following instructions: Suppose your community intends to implement the ‘Certain Time Certain Place’ waste sorting policy. The policy entails the removal of existing waste cans and their replacement with a designated four-sorting waste station. The station will be open at a fixed time each day and closed at other times, and residents will be required to separate their waste. The community will assign someone to monitor the waste disposal. We measured their approval of the CTCP waste sorting policy (1 = completely disapprove, 7 = completely approve). Participants also rated their waste sorting experience (1 = not at all, 7 = always) and reported the types of waste bins in their residential area. The options included food waste bins, residual waste bins, hazardous waste bins, recyclables bins, nonrecyclable bins and unsorted bins. In terms of demographic variables, we additionally measured whether participants have children.
Results and discussion
The results of the manipulation check confirmed that participants in the future-other condition exhibited greater concern and closeness toward future others (M future-other = 4.89, SD = 1.21) than participants in both the present-other condition (M present-other = 4.46, SD = 1.28), t(311) = 3.02, p = 0.003, d = 0.34, 95% CI [0.12, 0.56] and the future-imagining condition (M future = 4.61, SD = 1.35), t(307) = 1.88, p = 0.061, d = 0.21, 95% CI [–0.01, 0.44] (marginally significant with a small to medium effect size).
We conducted a linear regression analysis to examine the effect of future other-concern elicitation (future-other condition as the reference) on participants’ recycling intentions and approval of the CTCP policy. Three variables were introduced as control variables: (i) waste sorting experience; (ii) the total number of types of sorting bins in the community where the participant resides (0 = no bins with sorting marks, 5 = five types of sorted waste bins); and (iii) whether the participant has a child (0 = no, 1 = yes). As predicted, participants in the future-imagining condition reported a lower willingness to recycle than those in the future-other condition (M future = 5.77, SD = 1.03 vs M future-other = 5.95, SD = 0.89, b = −0.23, SE = 0.09, p = 0.014, 95% CI [–0.42, −0.05]), whereas there was no significant difference between the present-other and future-other conditions in terms of willingness to recycle (M present-other = 5.82, SD = 0.99, b = −0.12, SE = 0.09, p = 0.188, 95% CI [–0.31, 0.06]). For the acceptance of the CTCP policy, participants who only imagined future world were less likely to accept the policy than those who imagined future life of the newborn (M future = 5.44, SD = 1.69 vs M future-other = 5.76, SD = 1.43, b = −0.23, SE = 0.11, p = 0.030, 95% CI [–0.44, −0.02]). The difference in acceptance between those who imagined the future and present lives of the newborn was not statistically significant (M present-other = 5.89, SD = 1.39, b = 0.08, SE = 0.11, p = 0.438, 95% CI [–0.13, 0.29]).
Study 2A employed a more direct manipulation to prime concerns for future others and investigated its impact on the acceptance of the CTCP policy. As the intergenerational decision model predicted, imagining the future is not as effective as imagining future people in promoting recycling intention and policy approval. However, it should be noted that the same picture of a baby was used in both the present-other and future-other conditions, likely creating an effect comparable to identifiability effect. Extensive research indicates that identifiable victims elicit greater emotional responses and a higher propensity for helping compared to statistical victims who remain unidentified (Jenni and Loewenstein, Reference Jenni and Loewenstein1997; Small and Loewenstein, Reference Small and Loewenstein2003; Small et al., Reference Small, Loewenstein and Slovic2007). The difference in writing instructions might be overwhelmed by the salience of the visual stimulus, resulting in similar results in the present-other and future-other conditions. Thus, Study 2B was conducted without the presentation of pictures to avoid this issue.
Study 2B
In light of the potential confounding from vivid image stimuli, Study 2B was designed to repeatedly assess the effectiveness of a future-other-oriented strategy, employing the legacy writing task used in Study 1A. Additionally, all previous studies specified a 20-year future temporal frame, which may inadvertently prompt participants to think of their near-future selves. To address this, Study 2B did not impose a specific temporal frame.
Method
Participants
Participants were recruited nationwide from locations that did not implement the CTCP policy. The final sample size was N = 600 (M age = 29.52, SD = 8.64, age range = 18–61, 65.5% female). The demographic characteristics of the participants are displayed in supplemental Table S4. All participants received a small reward via the platform.
Procedure and materials
Participants were randomly assigned to one of three conditions: future-other condition (n = 201), present-other condition (n = 201) and future-imaging condition (n = 198). In the future-other condition, participants were instructed to write how they could influence the circumstances for future generations, similar to the approach used in Study 1A (adapted from Zaval et al., Reference Zaval, Markowitz and Weber2015; Bang et al., Reference Bang, Zhou Koval and Wade-Benzoni2017; Paek et al., Reference Paek, Goya-Tocchetto and Wade-Benzoni2024). In the present-other condition, participants imagined and wrote about a stranger’s current life. The task in the future-imaging condition mirrored that of Study 2A, without a temporal framing reminder. After completing the manipulation check, participants were presented with an introduction to the CTCP policy that restricts when, where and how residents could drop off their household waste. Policy support was evaluated in a manner similar to Study 2A, with a slight modification: participants additionally indicated their support for the policy that included penalties using a seven-point scale (1 = completely disapprove, 7 = completely approve) (e.g., ‘According to the CTCP program, fines will be imposed for failure to sort waste’).
Results and discussion
The results of the manipulation check confirmed that participants in the future-other condition exhibited greater concern and affinity toward future others (M future-other = 5.30, SD = 1.18) than participants in both the present-other condition (M present-other = 4.97, SD = 1.23), t(400) = 2.74, p = 0.006, d = 0.27, 95% CI [0.08, 0.47] and the future-imagining condition (M future = 4.86, SD = 1.32), t(397) = 3.54, p < 0.001, d = 0.35, 95% CI [0.16, 0.55].
A linear regression was conducted to examine the effect of future other-concern elicitation on participants’ acceptance of the CTCP policy and monetary penalty. Variables controlled for the analyses include experience of waste sorting, and whether the place of residence had four-category waste bins (0 = no, 1 = yes). As Figure 4 illustrates, participants in the future-imagining condition reported lower support for CTCP policy than those in the future-other condition (M future = 5.69, SD = 1.53 vs M future-other = 6.11, SD = 1.09), b = −0.27, SE = 0.09, p = 0.005, 95% CI [−0.45, −0.08]. There was still no significant difference in policy support between the present-other and future-other conditions (M present-other = 5.87, SD = 1.36), b = −0.10, SE = 0.09, p = 0.312, 95% CI [−0.28, 0.09]. For the acceptance of the CTCP policy that includes potential penalties, we additionally controlled income level. The results showed that those in the future-imagining condition were also less likely to accept fines than those in the future-other condition (M future = 4.92, SD = 1.61 vs M future-other = 5.29, SD = 1.34), b = −0.21, SE = 0.09, p = 0.026, 95% CI [−0.39, −0.03]. Notably, the difference between the present-other and future-other conditions was also significant. Participants who envisioned the present circumstances of a stranger exhibited lower support for the CTCP policy that involves monetary punishment (M present-other = 4.90, SD = 1.67), b = −0.19, SE = 0.09, p = 0.043, 95% CI [−0.38, −0.01]).
Policies acceptance (without and with monetary penalty) by conditions in Study 2B. Note. Error bars indicate standard error of the mean. *p < .05. **p < .01. ***p < .001.
Study 2B further demonstrated that priming concern for future others via the legacy task could enhance support for the CTCP policy more effectively than merely imagining the future. Specifically, participants in the future-other condition showed a greater willingness to accept fines for CTCP waste sorting violations, not only in comparison to those in the future-imaging condition but also relative to those in the present-other condition.
General discussion
In the present research, we examined whether eliciting concerns for future others results in recycling intentions and real-life recycling behavior. Across two main studies, the results showed that people behaved more sustainably when future other-concern was elicited. This effect manifests in an increased intention to engage in recycling behaviors and a greater willingness to participate in recycling training programs when individuals consider leaving a legacy for future strangers (Study 1A). This is also reflected in a greater willingness to pay for recyclable express packaging and increased actual recycling behavior when people view the aged face of another person (Study 1B). In Study 2, there was higher approval of the CTCP policy when participants imagined the future life of a newborn than when they imagined the future world.
Theoretical implications
The present research contributes to the literature in the following ways. First, the theoretical model of intergenerational decision-making speculates that a combination of intertemporal distance and interpersonal distance diminishes preferences and behaviors related to intergenerational sustainability (Wade-Benzoni and Tost, Reference Wade-Benzoni and Tost2009). Pro-environmental behavior, including recycling, also involves such a psychological barrier because people need to make a trade-off between the benefits to their present selves and benefits to future others. Our findings demonstrate that this essential feature of the intergenerational decisions can be leveraged to encourage recycling. Specifically, when eliciting concerns for future others before making a present-self vs future-other tradeoff, behavior shifts toward favoring the benefits of future others. This finding aligns with previous research on intergenerational affinity (Wade-Benzoni, Reference Wade-Benzoni2002, Reference Wade-Benzoni2003), perspective-taking of future generations (Ahn et al., Reference Ahn, Bostick, Ogle, Nowak, McGillicuddy and Bailenson2016; Scultz, Reference Scultz2002; Pahl and Bauer, Reference Pahl and Bauer2013) and legacy motivation (Zaval et al., Reference Zaval, Markowitz and Weber2015; Hurlstone et al., Reference Hurlstone, Price, Wang, Leviston and Walker2020).
According to the value-belief-norm theory, environmental value orientations can be classified into three categories: self-interest orientation, social-altruistic orientation and biospheric orientation (Schultz, Reference Schultz2001; Alibeli and White, Reference Alibeli and White2011; Obeng and Aguilar, Reference Obeng and Aguilar2018). As the intertemporal distance between the present and future decreases, individuals may become more likely to exhibit a self-interest orientation with respect to environmental concerns. This manifests as a focus on the personal implications of environmental issues and drives pro-environmental behavior for self-protection (Schultz et al., Reference Schultz, Gouveia, Cameron, Tankha, Schmuck and Franěk2005). For example, individuals with a self-interest orientation may exhibit concern for the environment due to apprehensions regarding the detrimental effects of air pollution on their personal health. In certain circumstances, however, a self-interest orientation may drive people to make environmentally unfriendly decisions. For example, people are more likely to order takeout food when PM2.5 levels rise, which in turn increases plastic waste (Chu et al., Reference Chu, Liu and Salvo2021). According to Ellis (Reference Ellis2013), more than 80% of respondents said that they continued to use private vehicles during smoggy days or even increased their use. Empirical evidence also suggests that self-interest orientation does not consistently predict engagement in pro-environmental behavior (Joireman et al., Reference Joireman, Lasane, Bennett, Richards and Solaimani2001; Tamar et al., Reference Tamar, Wirawan, Arfah and Putri2020). We speculated that an exclusive focus on a singular intertemporal dimension may engender concerns for future self while neglecting future others. Therefore, strengthening temporal proximity may encourage self-interest orientation, leading to self-protective behaviors rather than a consistent display of pro-environmental behavior. However, future other-concern elicitation may promote environmental engagement through a social-altruistic orientation, which warrants further empirical investigation.
Furthermore, the current work contributes to the literature on psychological strategies for encouraging recycling behavior (Varotto and Spagnolli, Reference Varotto and Spagnolli2017; Grilli and Curtis, Reference Grilli and Curtis2021) by introducing future other-concern elicitation as a theory-driven and effective strategy. While recycling is considered an action that benefits future generations, there has been limited research on how to apply the framework of intergenerational choice to promote recycling behaviors. Pandit et al. (Reference Pandit, Nakagawa, Timilsina, Kotani and Saijo2021) found that experiencing the perspective of future generations affected the preferences of individuals regarding solid waste management policy options. However, their findings are based on participants’ preferences or behavioral intentions rather than actual recycling behaviors, which thus limits the external validity of the case study. To our knowledge, this empirical research is the first to reveal that the elicitation of concerns for future others contributes to recycling intentions and behaviors.
Implications, limitations and future research directions
The current work provides a relatively low-cost and non-imperative method to promote and communicate about recycling. We found that individuals’ recycling intentions and behaviors can be modified in a sustainable manner by interventions that make future others salient and elicit concerns for future others. These interventions may play a crucial role in current situations where there is general public shortsightedness. Additionally, interpersonal distance along with long time horizons should be given more consideration in future practices. Numerous future-oriented slogans and campaigns have been developed to encourage recycling and waste sorting. Examples include By recycling today, you can change tomorrow and Recycle for a brighter future. From an intergenerational choice perspective, more precise future-others-oriented slogans, such as Think more about future generations before you act, should be created.
This paper is not without its limitations, which highlight potential directions for future research. First, we primarily examined near-future imagining (20 years) by adapting future self-continuity priming approaches used in previous studies (van Gelder et al., Reference van Gelder, Hershfield and Nordgren2013; Rutchick et al., Reference Rutchick, Slepian, Reyes, Pleskus and Hershfield2018). Particularly in Study 1B, the stimulus presented to participants depicted a near-future version of a contemporary individual. Participants may inevitably associate with their future selves, despite our attempts to control for the variable of future self-concern or remove temporal limits in Study 2B. It would be interesting to manipulate temporal distance in interpersonal negotiations and explore more elaborate approaches to promote environmental engagement. Second, we attempted to prime concerns for future others through various interventions. These interventions included instructing individuals to imagine what they could leave for people in the future (Study 1A and Study 2B), showing them a virtually aged face of a present-day person (Study 1B), and displaying a picture of a newborn and instructing them to imagine his or her future life (Study 2A). Those who live in the future (i.e., future others) were specified as an aged version of the present-day experimenter or future generation, but both were strangers to participants. However, future individuals can be further distinguished based on the interpersonal distance, such as their children, relatives and friends. Future work should disentangle how interpersonal distance affects intergenerational choice. Finally, although a manipulation check was passed in three studies to some extent, the influence of other confounding variables cannot be ruled out totally. For example, our future-others-oriented interventions could also be a reminder of death, even though the intertemporal dimension was specified as 20 years from now. Previous researchers have established that reminders of mortality can positively influence prosocial behavior through death reflection (Cozzolino et al., Reference Cozzolino, Staples, Meyers and Samboceti2004; Cozzolino, Reference Cozzolino2006; Blackie and Cozzolino, Reference Blackie and Cozzolino2011), while the path from death priming and intergenerational beneficence through feelings of connection with future others has also been shown (Wade-Benzoni et al., Reference Wade-Benzoni, Tost, Hernandez and Larrick2012; Hurlstone et al., Reference Hurlstone, Price, Wang, Leviston and Walker2020). If our manipulations reminded participants of death, is it the death reflection or concerns for future others that play a key role, or is there a chain relationship between the two? Future studies might clarify this point by using more typical death-priming manipulations and measuring death reflection.
Conclusion
Individuals frequently encounter the dilemma of whether to expend effort in recycling materials to conserve resources for future generations. A combination of intertemporal distance and interpersonal distance constitutes a barrier that hinders people’s present efforts to engage in recycling. Our findings support the hypothesis that this fundamental feature can be leveraged to promote recycling by eliciting concerns for future others. We anticipate that these results will offer intriguing guidance for researchers and practitioners working to advance public engagement with the environment.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S2398063X24000459.
Funding statement
This research was supported by the National Natural Science Foundation of China (Grant No. 71771209) and the Major Program of the National Social Science Foundation of China (Grant No. 19ZDA358).
Open access
