High and Low Carbon Footprint Feedback Produces Similar Levels of Proenvironmental Intentions Over Time

Abstract

Although there is an increasing number of studies that have investigated the effects that negative emotions have on proenvironmental intentions and environmental attitudes, there is almost no research on how negative emotions affect proenvironmental attitudes and intentions over time. We examined how manipulated guilt over one's carbon footprint (CF) produced changes in concerns toward the environment and proenvironmental intentions over a 2-day period. Participants in the high CF feedback condition reported more guilt on day 1 than those in the low CF feedback condition. Although several between-subject differences in concerns for the environment and intentions to engage in proenvironmental behaviors were found on day 1, there were no differences between the conditions in the aforementioned variables on day 2. The results suggest that memory consolidation is significant for understanding environmental attitudes and behaviors. Specifically, we replicate findings from previous studies (i.e., people experiencing negative emotion report more concern toward the environment and intentions to engage in proenvironmental behavior after CF feedback) as well as report that over time, receiving feedback stating that one's CF is either high or low leads to similar levels in the aforementioned variables. Our findings imply that regardless of the positive or negative appeal, simply thinking about the extent to which one behaves sustainably is significant for combating climate change. Future directions are discussed.

Introduction

Climate change has emerged as one of the greatest existential threats to life in the 21st century (Friedman & Thrush, 2017). According to the National Centers for Environmental Information (2017), the average global temperature across land and ocean surface areas in the 21st century was 14.7°C (58.5°F) compared with the 20th century average of 13.9°C (57.0°F). The change in global temperature is having detrimental effects in regional and seasonal weather, sea level rise, ice mass loss, and plant blooming (Linsey & Dahlman, 2020). Researchers have reached consensus that these environmental changes are caused by human behavior (Friedman & Thrush, 2017). It then becomes incumbent on researchers to identify human intervention strategies to change human behavior to be more environmentally friendly. The research presented here tests how guilt over one's carbon footprint (CF) influences eco-related behaviors 2 days after the intervention. The goal is to identify persuasive appeals that extend beyond the single experimental session for relatively long-term effects of guilt inducing manipulations.

Time-Dependent Memory Consolidation

Memory formation is critical to our everyday lives and, as such, should inform the questions we ask in research. The memory consolidation research posits that new information is, at first, weakly stored into memory structures, and easily disrupted by behavioral and physiological factors (McGaugh, 2004). Over time, however, the new information becomes more readily accessible, unified as part of prevailing cognitive structures, and generalized across novel domains (Enge, Lupo, & Zárate, 2015). This cognitive process through which information “withstands the passage of time” is referred to as memory consolidation (Payne & Kensinger, 2010).

Sleep, when following learning, has been shown to facilitate memory consolidation and improve performance on declarative learning tasks (e.g., vocabulary lists; Gais, Lucas, & Born, 2006), procedural learning tasks (e.g., motor memory; Stickgold, 2005), and visual discrimination tasks (Stickgold, Whidbee, Schirmer, Patel, & Hobson, 2000). Consolidated memories are considered more lasting and integrated with existing memory structures. This suggests that researchers should consider the effects of their manipulations beyond the single experimental session. The research presented here tests the effects of a persuasion appeal across a 2-day span to initiate investigation of how memory influences eco-related behaviors.

Emotional Memory Consolidation

Emotion has a powerful impact on our memory structures (Payne & Kensinger, 2010). Not only can emotions change the manifestations of the information learned, but emotions can also facilitate what information is selectively retained and recalled (Payne, Stickgold, Swanberg, & Kensinger, 2008). For instance, one study examining the recall of scenes portraying negatively arousing or neutral objects found that negatively arousing objects were better remembered than neutral objects (Payne et al., 2008). Similar effects are also found when negatively arousing information is placed in a social context. Enge et al. (2015) found that for learned targets, participants responded consistent with an in-group bias, where racial out-group members became more automatically associated with negative traits over time. Lupo and Zárate (2019) further demonstrate that over time, people generalize negative group traits to other group members and to the self. Taken together, these studies suggest that negative emotional arousal and social contexts may interact to facilitate memory consolidation effects over time. In this study, we investigate how negative emotions and guilt, in particular, influence behavior over time within a persuasion appeal.

Emotions, Affective Arousal, and Prosocial Behavior

Guilt is a negative emotional state that results from a violation of personal or social values. Guilt is also a prosocial motivator. One study (Mallett, 2012) confronted participants with instances in which they failed to meet standards for environmental protection and found that guilt predicted proenvironmental intentions. Using a similar manipulation, a second set of studies found that guilt partially mediated the relationship between one's CF feedback and support for a proenvironmental group (Mallett, Melchiori, & Strickroth, 2013). Other studies have found that guilt is more likely to motivate environmental mitigation behavior (Ferguson & Branscombe, 2010).

Guilt appears to produce consistent but relatively small effects in persuasion appeals to produce proenvironmental intentions. One explanation for the small effect sizes may be that these effects are being tested through single-session designs. In this study, we test the hypothesis that the effects of guilt should be larger after a period containing sleep, when memory consolidation processes have occurred.

The Present Study

This study manipulated guilt of one's CF to test for memory effects for a 2-day period. As a test of our manipulations, we hypothesized that feedback indicating that one's CF is high (compared with feedback indicating that the participant's CF is low) would produce greater concerns for the environment and greater proenvironmental intent. Furthermore, we predicted that time-dependent memory consolidation would produce greater environmental concerns and proenvironmental intent at day 2 than at day 1. Lastly, we predicted that there would be a feedback by day interaction such that those with high CFs would produce the greatest concerns for the environment and proenvironmental intent after a long delay in time (compared with those with low CFs).

Methods

Participants

Undergraduate students at the University of Texas at El Paso (N = 135) were recruited to participate in this study. Data from 18 participants were removed from our analyses due to not completing any of the day 2 questionnaires. Thus, our final sample consisted of 117 participants. A majority of participants identified as Latinx (83%) and female (77%) with an average age of 20.19 years (SD = 2.90). Participants reported sleeping between 1 and 10 h during the first night (M = 6.84, SD = 1.56) and between 1 and 10 h during the second night (M = 6.91, SD = 1.44).

Design

Participants were recruited to complete a two-part study on environmental attitudes. The study employed a 2 (Feedback: low CF vs. high CF) × 2 (Time delay: sleep vs. no sleep) between-participants mixed-model design. Feedback served as the between-participants variable in this study. Time delay was treated as a within-participant variable. Experimental conditions will be simply labeled low CF and high CF conditions, to represent low carbon footprint and high carbon footprint, respectively.

Materials

CF calculator

Guilt was manipulated through an ostensible CF calculator. The calculator asked questions regarding consumption behaviors, including transportation (car, air travel, and public transportation), housing (household size), energy usage, and spending habits (food and personal goods).

Rather than receiving a true CF score, participants were randomly assigned to one of two feedback conditions (high CF or low CF). Participants in the high CF condition learned that their CF was larger than that of the average U.S. citizen. Participants in the low CF condition learned that their CF was smaller than that of the average U.S. citizen. See our open science framework account* or our laboratory website^† for our experimental stimuli.

Measures

Emotions and intentions to conduct ecofriendly behaviors were administered to the participants. Each item was measured on a six-point scale, unless otherwise noted.

Manipulation checks

After the false feedback, participants were asked to confirm the size of their CF based on the feedback from the calculator. Participants also rated the extent to which the calculator was an accurate reflection of his/her consumption habits.

Emotions

Five items measured environmental guilt [day 1: α = 0.88; day 2: α = 0.92; adapted from Ferguson & Branscombe (2010)]. We also had other emotion-related questions asking about anger and shame for exploratory purposes. An example item is: “I feel guilty [angry, ashamed] about consuming non-renewable natural resources.”

Intentions to engage in proenvironmental behavior

Nine items measured intent to engage in proenvironmental behavior. The reliability of the scales was acceptable across days [day 1: α = 0.89; day 2: α = 0.93; adapted from Urien & Kilbourne (2011)]. An example item reads: “I intend to buy environmentally friendly products in the future.”

Concern for the environment

Concern for the environment was assessed through five items (Kellstedt, Zahran, & Vedlitz, 2008): An example item is: “Climate change will have a noticeably negative impact on my health in the next 25 years.” One item (“I believe climate change is human-induced”) was removed from this scale because it weakened the reliability of the scale. Scale reliability was consistent across days 1 (α = 0.73) and 2 (α = 0.79).

Demographic questionnaire

In the demographic section of the study, participants indicated their gender, age, race/ethnicity, and political ideology (1, extremely liberal; 7, extremely conservative).

Procedures

The study was completed in two experimental sessions. During the first session, participants answered questions about their ecological behavior and received false feedback about their consumption behaviors through a CF calculator. Participants were randomly assigned to learn that their CF was larger or smaller than that of the average U.S. citizen. After the manipulations, participants rated their feelings of guilt, concerns toward the environment, and intent to engage in proenvironmental behavior. Participants were then instructed on how to sign up for the Remind application.

Participants received a link (through the Remind application) for the second portion of the survey 48 h after the initial testing session. The second survey included the same dependent measures from session 1 and a sleep questionnaire regarding the participants' quantity and quality of sleep for each night during the 48-h delay.

Results

We had three predictions. First, participants in the high CF condition would experience significantly more personal guilt on day 1 in comparison with those in the low CF feedback condition. Second, those in the high CF feedback condition would report higher concerns toward the environment after a long delay in time (between the conditions as well as within the condition). Finally, participants in the high CF feedback condition would report greater intentions to engage in proenvironmental behavior after a long delay in time (between the conditions as well as within the conditions). Each dependent variable was analyzed through separate General Linear Models.

Manipulation checks

Manipulation checks (e.g., is your CF larger than the average U.S.'s citizen?) were carried out before assessing whether there were differences between our conditions in the aforementioned dependent variables. The results confirmed that those in the high CF condition (M = 5.87, SD = 1.21) believed that their CF was significantly larger than those in the low CF condition (M = 2.25, SD = 0.84), F(1, 115) = 331.43, p < 0.0001, η² = 0.74. In addition, those in the high CF (M = 4.64, SD = 1.41) and low CF (M = 4.89, SD = 1.32) conditions reported the information they received as highly accurate, F(1, 115) = 1.00, p = 0.32. Thus, the manipulations were effective in producing the desired effects.

Environmental guilt

The findings analyzed using a repeated measures analysis on guilt by day across experimental conditions. Analyses revealed that there was a significant difference in environmental guilt on day 1, F(1, 115) = 9.93, p = 0.002, η² = 0.08, such that those in the high CF condition reported feeling more environmental guilt (M = 5.23, SD = 1.29) than those in the low CF condition (M = 4.40, SD = 1.56). There were no differences, however, in environmental guilt between the high CF (M = 4.99, SD = 1.37) and low CF (M = 5.17, SD = 1.43) conditions on day 2, F(1,115) = 0.47, p = 0.49.** Once we test for the effects of time, we find a significant day by feedback interaction, F(1, 115) = 23.28, p < 0.0001. Specifically, those in the high CF feedback condition reported lower environmental guilt on day 2 (M = 4.99, SD = 1.37) in comparison with their day 1 scores (M = 5.23, SD = 1.29), F(1, 60) = 7.78, p = 0.007. In addition, those in the low CF condition reported higher environmental guilt scores on day 2 (M = 5.17, SD = 1.43) in comparison with their day 1 scores (M = 4.40, SD = 1.56), F(1, 55) = 15.19, p < 0.001.

Concerns toward the environment

As for concerns toward the environment, there was a difference between the feedback conditions in concerns toward the environment on day 1, F(1, 115) = 5.89, p = 0.017, η² = 0.05. Specifically, those in the high CF feedback condition (M = 5.42, SD = 0.86) reported more concerns toward the environment on day 1 in comparison with those in the low CF condition (M = 4.89, SD = 1.44). There were no differences in concerns toward the environment on day 2 between the high CF (M = 5.38, SD = 0.86) and low CF conditions (M = 5.49, SD = 1.12), F(1, 115) = 0.38, p = 0.54. Lastly, there was a significant day by feedback interaction, F(1, 115) = 10.75, p = 0.001, such that those in the low CF condition reported higher concerns for the environment on day 2 (M = 5.49, SD = 1.12) in comparison with their scores on day 1 (M = 4.89, SD = 1.44), F(1, 55) = 11.05, p = 0.002. There were no differences in the high CF condition between participant's concerns toward the environment on day 1 (M = 5.41, SD = 0.86) and on day 2 (M = 5.38, SD = 0.86), F(1, 60) = 0.23, p = 0.63.

Intentions to engage in proenvironmental behavior

Lastly, there were several significant between-subjects and within-subject differences in intentions to engage in proenvironmental behavior. On day 1, those in the high CF condition (M = 5.35, SD = 0.92) reported higher intentions to engage in proenvironmental behaviors in comparison with those in the low CF condition (M = 4.62, SD = 1.39), F(1, 115) = 11.19, p = 0.001, η² = 0.09. On day 2, there were no differences in intentions to engage in proenvironmental behavior between the high CF (M = 5.31, SD = 0.94) and low CF (M = 5.31, SD = 1.20) conditions, F(1, 115) = 0.00, p = 0.98. Importantly, there was a significant day by feedback interaction, F(1, 115) = 13.40, p < 0.001, such that those in the low CF feedback condition reported significantly higher intentions to engage in proenvironmental behaviors on day 2 (M = 5.31, SD = 1.20) in comparison with their day 1 scores (M = 4.62, SD = 1.39), F(1, 55) = 13.85, p < 0.001. In contrast, there were no differences in the high CF condition between intentions to engage in proenvironmental behavior on day 1 (M = 5.35, SD = 0.92) and day 2 (M = 5.31, SD = 0.94), F(1, 60) = 0.15, p = 0.70.

Post hoc analyses

As a further test of our hypotheses, we made a composite of the day 1 negative emotion data we collected (guilt, shame, and anger; α = 0.89). We then regressed intentions to engage in proenvironmental behavior on day 2 on the aforementioned negative emotions composite. The findings indicated that negative emotions on day 1 positively predicted intentions to engage in proenvironmental behavior on day 2, b = 0.36, t(115) = 5.58, p < 0.0001, such that for every 1 standard unit increase in negative emotions felt on day 1, there was a 0.36 increase in intentions to engage in proenvironmental behavior on day 2. Negative emotions on day 1 also explained a significant proportion of variance in intentions to engage in proenvironmental behavior on day 2, R²_adjusted = 0.21. The slopes of the aforementioned regression did not differ by experimental condition, t(113) = −0.55, p = 0.58.

Discussion

The purpose of this study was to experimentally test the effects of guilt on concerns toward the environment and intentions to engage in ecofriendly behavior over time. The findings revealed that over time, participants in the low CF condition appeared to become more proactive regarding their concerns toward the environment and proenvironmental intentions. In contrast, those in the high CF condition appeared to have greater intentions to engage in proenvironmental behaviors on day 1, but that effect dissipated over time. The results also revealed that negative emotions felt on day 1 significantly predicted proenvironmental intentions on day 2. These findings build off the findings found by Mallett (2012), wherein the researchers tested the effects of guilt on the aforementioned dependent variables through a single-session design. Like Mallet's findings (2012), immediately after being confronted with evidence that participants were not behaving sustainably (e.g., high CF feedback), participants reported feeling more guilt and reported greater concerns toward the environment and proenvironmental intentions than when they were told that they were behaving sustainably (e.g., low CF feedback).

Relative to those in the high CF condition, participants in the low CF condition reported more guilt on day 2 than on day 1. In addition, participants in the low CF feedback condition reported greater concerns for the environment and greater proenvironmental intentions over time. This may have occurred because the low CF feedback affirmed one's self-efficacy to engage in sustainable behaviors. Previous research suggests that self-efficacy—a person's belief in his/her capacity to perform a task—can influence the magnitude, strength, and generality of a behavior (e.g., Bandura, 1977). In this study, self-efficacy may have affirmed one's ability to perform sustainable behaviors (given the low CF feedback), which, in turn, may have led participants to want to engage in more sustainable behaviors. Thus, future research should explore the role of self-efficacy in promoting sustainable behavior over time. Time is particularly important as this effect became evident only on day 2.

The changes in guilt over time in the low CF condition but not in the high CF condition speak of two possible mechanisms. First, the negative valence in guilt may have stabilized over time and may have kept environmental concerns salient in one's mind (leading to similar results in the high CF condition at days 1 and 2). This mechanism supports the idea that guilt (e.g., negative information) influences memory consolidation such that information becomes stable over time and/or leads to less forgetting of the encoded information.

It could also be the case that the positive feedback, in this instance stating that one has a lower CF than do others, promotes long-term behavior. This suggests that providing some history of success against acid rain (University of Maine, 2003), and protecting the ozone layer (Blakemore, 2016), for instance, might produce better long-term effects through self-efficacy type effects. Research needs to further investigate this issue, both within this context and with other forms of environmentally friendly behaviors. Presumably, those studies will need to include measures of self-efficacy, but also motivation and other hypothesized mediators.

Research outside of the sustainability literature suggests that positive appeals are more effective than negative appeals at promoting behavioral change (e.g., in health), and negative appeals are more effective than no appeals (Robberson & Rogers, 1988). Future research should address this limitation and compare proenvironmental intentions across low CF feedback, high CF feedback, and no feedback conditions over short and long delays in time.

It is worth noting that negative emotions felt on day 1 positively predicted proenvironmental intentions on day 2. In other words, the more that participants felt negatively regarding their environmental behavior on day 1, the higher their intentions to engage in proenvironmental behavior on day 2. These findings provide further support that negative emotional arousal can interact with social contexts and produce memory consolidation effects over time as well as serve as a prosocial motivator, even within the context of becoming more environmentally friendly.

Implications

Given the urgency of climate change (Friedman & Thrush, 2017), more research is needed to understand the mechanisms that influence individuals to engage in environmentally sustainable ways. In this study, we present evidence that making individuals feel guilty (or positive) about their CF leads to different patterns of behavioral intentions. Guilt might produce more noticeable short-term behaviors, but here, the positive message produced effects better on day 2 than on day 1. One obvious implication of these data is that if researchers study the aforementioned issues through single-session designs, their studies may be incomplete or in some instances wrong. For example, research has shown that making people feel negatively (vs. positively) about their CF leads to higher intentions to engage in ecofriendly behavior on day 1 (Mallet, 2012). One interpretation of these findings could be that we must make people feel negatively about their behavior to change their behavior. In contrast, our research suggests that low CF feedback might produce better long-term levels of environmental concerns and proenvironmental intentions but only if tested after a delay in time. Because long-term change is the goal, research needs to further test the effects of their manipulations after their first experimental session.

Conclusion

Although our results did not fully support our hypotheses, they extend the traditional social experimental design. To our knowledge, very few studies have investigated (within the context of environmental behaviors) how representations at the individual level become integrated over time and how this memory consolidation process influences behaviors. Thus, our study extends previous research by allowing us to examine how emotional experiences impact environmental attitudes over long delays in time. In our study, low CF and high CF feedback differentially impacted proenvironmental intentions on day 1. Nevertheless, both feedback conditions led participants to engage in proenvironmental behavior on day 2. These results suggest that encouraging students to think about sustainable behavior (regardless of the positive or negative appeal used in such thinking) is an important step toward combating climate change.

Footnotes

Acknowledgment

Special thanks to Robyn K. Mallet in the Department of Psychology at Loyola University of Chicago for allowing us to use her materials in this study.

Disclaimer

The authors do not have any interests that might be interpreted as influencing the content of this article, and APA ethical standards were followed in the preparation of this article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this study.

References

Bandura

(1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215.

Blakemore

(2016, January 13) The ozone hole was super scary, so what happened to it? Smithsonian Magazine. Retrieved from https://www.smithsonianmag.com/science-nature/ozone-hole-was-super-scary-what-happened-it-180957775/

Enge

L. R.

, Lupo

A. K.

, & Zárate

M. A.

(2015). Neurocognitive mechanisms of prejudice formation: The role of time-dependent memory consolidation. Psychological Science, 26, 964–971.

Ferguson

M. A.

, & Branscombe

N. R.

(2010). Collective guilt mediates the effect of beliefs about global warming on willingness to engage in mitigation behavior. Journal of Environmental Psychology, 30, 135–142.

Friedman

, & Thrush

(2017, November 3) U.S. report says humans cause climate change, contradicting top Trump officials. New York Times. Retrieved from https://www.nytimes.com/2017/11/03/climate/us-climate-report.html

Gais

, Lucas

, & Born

(2006). Sleep after learning aids memory recall. Learning and Memory, 13, 1335–1339.

Kellstedt

P. M.

, Zahran

, & Vedlitz

(2008). Personal efficacy, the information environment, and attitudes toward global warming and climate change in the United States. Risk Analysis, 28, 113–126.

Linsey

, & Dahlman

(2020, January 16) Climate change: Global temperature. National Aeronautics and Space Administration. Retrieved from https://www.climate.gov/news-features/understanding-climate/climate-change-global-temperature

Lupo

, & Zárate

M. A.

(2019). Guilty by association: Time-dependent memory consolidation facilitates the generalization of negative—But not positive—Person memories to group and self-judgments. Journal of Experimental Social Psychology, 83, 78–87.

10.

Mallett

R. K.

(2012). Eco-guilt motivates eco-friendly behavior. Ecopsychology, 4, 1–9.

11.

Mallett

R. K.

, Melchiori

K. J.

, & Strickroth

(2013). Self-confrontation via a carbon footprint calculator increases guilt and support for a proenvironmental group. Ecopsychology, 5, 1–8.

12.

McGaugh

J. L.

(2004). The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annual Review of Neuroscience, 27, 1–28.

13.

National Centers for Environmental Information. (2017, December). Global climate report—Annual 2017. National Centers for Environmental Information. Retrieved from https://www.ncdc.noaa.gov/sotc/global/201713

14.

Payne

J. D.

, & Kensinger

E. A.

(2010). Sleep's role in the consolidation of emotional episodic memories. Current Directions in Psychological Science, 19, 290–295.

15.

Payne

J. D.

, Stickgold

, Swanberg

, & Kensinger

E. A.

(2008). Sleep preferentially enhances memory for emotional components of scenes. Psychological Science, 19, 781–788.

16.

Robberson

M. R.

, & Rogers

R. W.

(1988). Beyond fear appeals: Negative and positive persuasive appeals to health and self-esteem. Journal of Applied Social Psychology, 18, 277–287.

17.

Stickgold

(2005). Sleep-dependent memory consolidation. Nature, 437, 1272–1278.

18.

Stickgold

, Whidbee

, Schirmer

, Patel

, & Hobson

J. A.

(2000). Visual discrimination task improvement: A multi-step process occurring during sleep. Journal of Cognitive Neuroscience, 12, 246–254.

19.

University of Maine. (2003, January 29) Clean air act reduces acid rain: UMaine research shows recovery in New England elusive. EurekAlert! Retrieved from https://www.eurekalert.org/pub_releases/2003-01/uom-caa012903.php

20.

Urien

, & Kilbourne

(2011). Generativity and self-enhancement values in eco-friendly behavioral intentions and environmentally responsible consumption behavior. Psychology & Marketing, 28, 69–90.