How prompt concreteness shapes consumer food waste in restaurants: evidence from field and laboratory experiments

Abstract

Introduction:

The increasing problem of food waste has become a global issue. Displaying prompts to restaurant consumers is widely regarded as an important measure to reduce food waste, though its effectiveness remains debated. Based on this, this study explores the impact of a common design feature of prompts—prompt concreteness—on consumer food waste in restaurants.

Methods:

This study first conducted a field experiment in a Chinese-style casual dining restaurant by placing prompt labels on restaurant tables and weighing customers’ leftover food, providing initial evidence for the effect of prompt concreteness on consumer food waste. Subsequently, this study conducted two laboratory experiments to explore the psychological mechanism and boundary conditions of this effect.

Results:

The field experiment (N = 261) found that concrete prompts can reduce consumer food waste by an additional 17.2% compared with abstract prompts (p < 0.001). The first laboratory experiment (N = 193) confirmed that concrete prompts (vs. abstract prompts) were more effective in increasing consumers’ willingness to take leftovers (p = 0.003), with moral justification playing a mediating role. Another laboratory experiment (N = 357) showed that the average weight of food ordered by participants in the concrete prompt group was 10.54% lower than that of participants in the abstract prompt group (p = 0.008), indicating that concrete prompts can effectively reduce consumers’ over-ordering behavior. Furthermore, this effect disappeared when participants had higher levels of trait reactance.

Discussion:

These findings reveal how the concreteness of prompts influences consumer food waste in restaurants, as well as the underlying mechanisms and boundary conditions. The results provide a scientific explanation for the debate in previous research on prompt effectiveness, and offer valuable insights for marketers to scientifically design and apply prompts to reduce consumer food waste.

1 Introduction

Food waste has become a global issue of significant concern (Zhang et al., 2024). According to the United Nations Environment Programme (UNEP), approximately 1.05 billion tons of food are wasted globally each year. This accounts for nearly 30% of agricultural land and contributes to 8–10% of global greenhouse gas emissions (UNEP, 2024). These figures highlight the significant challenge food waste poses to global resource conservation and environmental protection (Scherhaufer et al., 2018; Liao et al., 2025). To address this issue, the United Nations has committed to halve global per capita food waste at the retail and consumer levels by 2030 (Sustainable Development Goal 12.3) (UNEP, 2021). Given that the food waste rate in consumer dining (11.7%) is much higher than that in households (4.8%) (China, 2023), reducing consumer food waste in restaurants has become a crucial issue to address in achieving the SDG (Kaman et al., 2024; Xiao et al., 2024).

Many governments and non-profit organizations around the world have made significant efforts to promote sustainability-oriented consumption and waste reduction (Agarwal, 2018). Specifically, in the context of anti-food waste campaigns in restaurants, providing prompts is seen as a common measure (Stöckli et al., 2018). It typically take the form of visual or auditory messages that are intended to remind people of the social expectations regarding their behavior (Schultz et al., 1995). Compared to interventions like rewards and punishments (Vizzoto et al., 2021) or changes in tableware design (Gao et al., 2024), prompts are easy to implement and cost-effective (Whitehair et al., 2013). As a result, many countries have adopted them as an important measure to reduce consumer food waste in restaurants. For example, in the UK, Waste and Resources Action Program (WRAP) recommends placing informational prompts on menus, posters, buffet areas, or table cards in restaurants as part of their “Resource Pack for Hospitality and Food Service Sector” to guide consumers in reducing plate waste (WRAP, 2020). In China, the “Anti-Food Waste Law” explicitly requires all restaurants to display prompts to remind consumers to reduce food waste (China, 2021).

From an academic perspective, there is still some debate about the effectiveness of prompts. While most scholars agree that prompts play a positive role in reducing food waste. For example, Whitehair et al. (2013) found that posting simple prompts in cafeterias reduced consumer food waste by 15%. Kallbekken and Sælen (2013) successfully reduced food waste by 20% by placing prompts in buffet restaurants. Stöckli et al. (2018) conducted a field experiment and found that prompts can effectively reduce consumer food waste in restaurants. However, a few studies argue that the impact of prompts on reducing food waste is limited. For instance, Yu et al. (2016) found in a survey of restaurants in Beijing that placing prompts had only a small effect on reducing consumer food waste. Similarly, Visschers et al. (2020) conducted a field experiment in two university cafeterias in Switzerland and found that prompts did not reduce plate food waste. These discrepancies have led scholars to recognize that the design features of prompts may influence the effectiveness (Jiang et al., 2024). However, further evaluation is still needed on how to design prompts to improve their effectiveness.

As an important feature of prompts, message concreteness refers to the extent to which information is clear and specific (Miller et al., 2007). High-concreteness information is typically described in a more vivid, perceptible, specific, detailed, or precise way, whereas low-concreteness information is more abstract and primarily captures intangible qualities, thoughts, or concepts (Packard and Berger, 2021). Existing research has shown that message concreteness can influence the persuasive impact (Miller et al., 2007; Packard and Berger, 2021). For example, in marketing, Schellekens et al. (2010) found that abstract word of mouth is more persuasive than concrete word of mouth. Meanwhile, Packard and Berger (2021) found that service staff using concrete language can increase consumer satisfaction. In the health domain, Grandpre et al. (2003) found that specific and concrete anti-smoking messages had limited effectiveness, while Miller et al. (2007) argued that presenting health information concretely can more effectively persuade individuals to adopt healthier lifestyles. In the field of green consumption, Baca-Motes et al. (2013) compared the impact of different levels of message concreteness on hotel consumers’ green behavior. They found that abstract commitments (e.g., “I will try to engage in environmentally friendly behaviors during my stay”) were more easily agreed upon, while concrete commitments (e.g., “I will try to reuse towels during my stay”) were more effective in changing behavior. However, they did not deeply explore the potential mechanisms behind this finding.

In summary, although scholars have gradually recognized that concreteness is an important factor influencing the persuasive impact, research on this topic remains limited (Packard and Berger, 2021), and conclusions about the direction of concreteness’ effects are not consistent. Drawing from Baca-Motes et al. (2013), the common prompts in restaurants can be categorized based on concreteness into concrete prompts and abstract prompts. Concrete prompts often involve specific behavioral reminders of social expectations, such as “Please take your leftovers to-go,” while abstract prompts are typically vague, general appeals, such as “Please cherish food.” This raises some important questions: Does the prompt concreteness influence consumer food waste in restaurants? If so, which type of prompt is more effective? We know little about this.

Building on the above discussion, this study conducts one field experiment and two laboratory experiments to systematically examine the role of prompt concreteness in reducing consumer food waste in restaurants. Specifically, this study seeks to address the following research questions: (1) Is there a difference in the impact of concrete prompts and abstract prompts on consumer food waste? (2) If so, what are the psychological mechanisms and the boundary condition of this effect? By answering these questions, this study theoretically reveals how message concreteness influences the effectiveness of anti-food waste prompts, providing a scientific explanation for the debate in previous research on prompt effectiveness. On a practical level, these findings provide scientific recommendations and empirical evidence for marketers to design more effective prompts to guide consumers in reducing food waste.

2 Theoretical background and hypotheses

2.1 Prompt concreteness and consumer food waste in restaurants

Consumer food waste in restaurant is primarily caused by inappropriate consumption behaviors (Gao et al., 2024), such as over-ordering (Li et al., 2022) or refusing to take leftovers home (Hamerman et al., 2018). Therefore, the key to reducing consumer food waste is to encourage people to change their behavior. Prompts in public spaces serve as reminders of socially expected behaviors (Schultz et al., 1995), and can help consumers build cognitive awareness of behavioral norms (Goldstein et al., 2008). The consistency principle of cognitive dissonance theory suggests that conflicts between cognition and behavior lead to cognitive dissonance, which is an uncomfortable negative emotional state (Roe et al., 2020). Therefore, individuals are strongly motivated to maintain harmony between their cognition and behavior to avoid dissonance (Festinger, 1957). Based on this, some scholars argue that consumers who receive prompts are more motivated to regulate their behavior to align with normative cognition (Kallbekken and Sælen, 2013). However, it is often overlooked that cognitive dissonance theory also points out that individuals have a second way of coping with dissonance: by justifying inappropriate behavior cognitively to make it acceptable (Mazar et al., 2008), and this method is typically characterized by low cost and quick occurrence (Vasquez et al., 2019). In contrast, sustainable consumption behaviors, such as reducing food waste, often require additional effort, time, or financial costs (White et al., 2019). Therefore, we speculate that after receiving the prompt, consumers may not immediately take action to save food. Instead, they might first try to rationalize their food waste behavior cognitively. Once successful, the motivation to change their behavior will be weakened (Festinger, 1957).

Previous research has found that when behavioral norms are ambiguous or vague, people gain more autonomy to interpret the situation, which makes it easier for them to rationalize their behavior (Shalvi et al., 2015). For example, posting the reminder “DO YOUR BEST” in the workplace does not improve employee performance because it does not provide specific behavioral requirements, allowing employees to interpret any effort as “doing their best” (Locke et al., 1989). A field experiment in hotels also found that simply making a commitment to “try to engage in environmentally friendly behaviors” does not effectively encourage consumers to take action (Baca-Motes et al., 2013). Similarly, concrete prompts clearly inform consumers about the behaviors that are encouraged, making it easier for them to assess whether their behaviors align with the norm, and harder to find excuses for behaviors that contradict the prompt. In contrast, abstract prompts are often more general and broad, lacking clear behavioral requirements, which gives consumers more room to justify their behaviors. This may lead consumers to more easily interpret inappropriate consumption behaviors as aligning with the norm, ultimately weakening the intervention effect of the prompt. Consequently, we propose:

H1: Compared with abstract prompts, concrete prompts are more effective in reducing consumer food waste in restaurants.

2.2 The mediating effect of moral justification

Individuals can avoid cognitive dissonance by consciously interpreting inappropriate behaviors as justified, Bandura et al. (1996) referred to this cognitive adjustment strategy as moral justification. Research shows that the occurrence of moral justification depends on two key factors. One is the ambiguity and malleability of the behavior in people’s common sense (Shalvi et al., 2015). For example, stealing money is generally considered unquestionably immoral, so its malleability is low and it is difficult to justify it as acceptable. In contrast, taking a friend’s pen is more malleable, making it easier to find rationalizations for such behavior (Hsee et al., 2003). In other words, behaviors that fall into the gray area are more likely to be subject to moral justification (Shalvi et al., 2015). The second factor is whether clear behavioral norms are provided in the external context (Mazar et al., 2008). When people receive clear reminders about norms, they are less likely to flexibly redefine their behavior (Vasquez et al., 2019). For example, asking participants to recall the 10 Commandments can inhibit moral justification and make them more honest in experiments (Mazar et al., 2008). In the context of consumer food waste, while wasting food is generally seen as a violation of moral and traditional values, some behaviors that lead to waste are often given positive connotations. For example, ordering too much food might be viewed as a symbol of generosity (Li et al., 2022), and refusing to take food leftovers may be seen as a way to avoid awkwardness (Hamerman et al., 2018). Therefore, the food waste behaviors can be considered highly malleable. In this case, whether clear behavioral guidelines are provided externally becomes particularly important. Prompts in restaurants are an important source for understanding socially expected behaviors (Goldstein et al., 2008). When prompts provide reminders related to concrete consumer behaviors, it becomes easier for consumers to assess whether they meet the normative requirements. This undoubtedly makes it more difficult for them to justify their actions morally, encouraging them to reduce food waste. On the other hand, if the prompts are abstract and vague, without making specific behavioral demands, consumers are given more flexibility in defining the situation, making it easier for them to use moral justification to enhance the perceived legitimacy of their food waste behavior. Thus:

H2: Moral justification mediates the effect of prompt concreteness (concrete vs. abstract) on consumer food waste in restaurants.

2.3 The moderating effect of trait reactance

Psychological reactance refers to a negative motivational state that arises when individuals’ freedom are threatened (Reynolds-Tylus, 2019). In clinical psychology, it is often considered a stable personality trait, known as trait reactance (Karno and Longabaugh, 2005). Individuals with high trait reactance are more sensitive to external control, which leads to stronger tendencies to resist (Beutler et al., 2011). This is typically reflected in an increased attraction to behaviors that are forbidden or discouraged, and a decreased attraction to behaviors that are advocated or recommended. As a result, they express opposing views or engage in contrary actions (Buboltz Jr et al., 2002). For example, research in the health field has found that explicit anti-smoking messages may have the opposite effect on people with high trait reactance, as they might perceive these messages as a threat to their freedom and thus intentionally increase smoking behavior (Grandpre et al., 2003). Additionally, Baca-Motes et al. (2013) found that hotel customers were less willing to accept specific commitments, likely because they felt a stronger sense of restriction. Regarding anti-food waste prompts, for consumers with low trait reactance, due to their higher tolerance for external control, reminders focused on specific behaviors do not make them feel uncomfortable. As discussed earlier, concrete prompts (vs. abstract prompts) are more effective in reducing food waste for these consumers. However, for consumers with high trait reactance, although concrete prompts limit their ability to justify food waste behavior morally, they may perceive these prompts as an infringement on their autonomy, which could trigger their resistance. As a result, the superiority of concrete prompts (vs. abstract prompts) in reducing food waste may not be as significant for these consumers. Formally, we hypothesize the following:

H3: Trait reactance moderates the effect of prompt concreteness (concrete vs. abstract) on consumer food waste in restaurants. Specifically, for consumers with low trait reactance, concrete prompts are more effective in reducing food waste compared with abstract prompts. However, for consumers with high trait reactance, this effect disappears.

2.4 Overview of studies

We conducted a field experiment and two laboratory experiments to ensure that our findings meet both internal and external validity. Study 1 was a field experiment conducted in a real restaurant to examine the effect of the prompt concreteness on consumer food waste. Study 2 was a laboratory experiment that replicated the findings of Study 1 using new stimuli. It also tested the mediating role of moral justification and ruled out potential alternative explanations. Study 3 further replicate the results of prior studies by changing the stimuli and the Method of measuring food waste (using a simulated ordering task). It also examined the moderating effect of trait reactance on the main effect.

3 Study 1: field evidence from a restaurant

Study 1 is a field experiment. We placed prompt labels on restaurant tables and weighed customers’ leftover food. The aim is to preliminarily test the main effect of the study (H1).

3.1 Pre-test

Before conducting the formal experiment, a pre-test was carried out to assess whether the manipulative materials met the experimental requirements. A total of 70 participants (62.8% female, M_age = 28.50, SD = 7.63) were recruited through the Credamo platform (one of the leading online survey platforms in China). Participants were randomly assigned to either the concrete prompt group or the abstract prompt group. Each group was presented with a corresponding prompt image, with the content adapted from the anti-food waste campaign by China Central Television (CCTV). As shown in Figure 1, the concrete prompt content was: “Order moderately; Packing leftovers makes sense.” The abstract prompt content was: “Cherish food through effort; Reducing waste makes sense.” Both prompts were of the same word count and shared a similar rhyming structure in Chinese.

Figure 1

Participants were then asked to evaluate the prompt’s text content using the following scales: “How concrete do you think the prompt you saw was (1 = not concrete at all, 7 = very concrete),” “How visually appealing do you think the prompt’s design is? (1 = very unattractive, 7 = very attractive),” and “How understandable do you think the prompt’s is? (1 = very difficult to understand, 7 = very easy to understand).”

One-way analysis of variance (ANOVA) results showed that participants in the concrete prompt group rated the concreteness of the prompt significantly higher [M_concrete = 5.49, SD = 1.27; M_abstract = 2.66, SD = 1.16; F(1, 68) = 94.62, p < 0.001]. There were no significant differences in the ratings for the visual appeal (p = 0.206) and comprehensibility (p = 0.482) of the labels. These results indicate that the experimental materials met the requirements and can be used in the formal experiment (Table 1).

3.2 Method

This study used a 3 (prompt concreteness: concrete vs. abstract vs. no prompt) between-subjects design. We conducted a 30-day experiment in a Chinese casual dining restaurant (From February 16 to March 20, 2025. Due to staff holidays after the Chinese New Year, data collection was not conducted for 3 days during this period). A total of 261 Tables (2,429 people) were recorded for the weight of leftover food.

The experiment was conducted in the restaurant lobby. We selected 15 easily observable tables as the experimental area. Ten prompt labels, based on pre-test results (five with concrete prompts and five with abstract prompts), were placed in the center of the tables near the utensils to ensure that customers could easily see the labels while picking up their utensils The remaining five tables, which did not have any labels, served as the control group (see Figure 2). Prior to the experiment, we trained the headwaiter and service staff on how to record and weigh the leftover food. To avoid the Hawthorne effect, the headwaiter was instructed to record the number of customers, their gender, and the dining time at each table in the experimental area without disturbing their meals. After the customers left, following the procedure of Gao et al. (2024), the cleaning staff poured the leftover food (excluding liquids) into a container on the scale and recorded the weight. They were then required to write their names next to each record, which helped researchers track the quality of the records and distribute corresponding rewards. Additionally, we required that each weighing be done by two staff members together to improve efficiency and ensure mutual supervision. Considering that food waste might differ between lunch and dinner (Wang et al., 2017), the experiment was conducted during both lunch and dinner hours. The researchers periodically checked the experimental records and distributed rewards. Additionally, the position of the labels was rotated weekly to minimize the potential impact of table location on the results.

Figure 2

3.3 Results

We conducted a one-way ANOVA on per capita food waste at the table with prompt concreteness (concrete vs. abstract vs. no prompt) as the independent variable. Levene’s test showed significant results (p < 0.001), indicating unequal variances across groups. Therefore, we performed a Welch-adjusted ANOVA, The results are shown in Figure 3. The results indicated that there were significant differences in the effects of prompt concreteness on per capita food waste at the table [M_concrete = 164.85, SD_concrete = 74.82; M_abstract = 199.30, SD_abstract = 101.78; M_control = 229.11, SD_control = 51.31; F_Welch (2, 168.68) = 21.72, p < 0.001, η_p² = 0.137]. When the number of diners, dining time, number of male diners, and number of female diners were included as covariates, the group differences remained significant (p < 0.001, η_p² = 0.097). Then we performed post hoc comparisons using the Dunnette’ T3 method. The results showed that the per capita food waste at the table in the concrete prompt group was significantly lower than that in the abstract prompt group (p = 0.028) and the control group (p < 0.001). Additionally, the per capita food waste at the table in the abstract prompt group was also significantly lower than that in the control group (p = 0.044). This provides preliminary support for H1.

Figure 3

3.4 Discussion

Study 1 confirmed that although abstract prompts can reduce consumer food waste in restaurants compared with no prompts, concrete prompts showed better results in this regard. This supports the existence of the main effect (H1). However, in the field experiment, we were only able to observe the relationship between prompt concreteness and consumer food waste in restaurants. It was difficult to explore the underlying mechanisms and boundary conditions in detail. Therefore, we need to conduct laboratory experiments to further test the research hypotheses.

4 Study 2: causal and mediating evidence

Study 2 is a laboratory experiment aimed at re-examining the impact of prompt concreteness on consumer food waste in restaurants and the mediating role of moral justification. We created a fictional scenario where participants dined at a restaurant with different types of prompts placed on their tables. Following the approach of Gao et al. (2024), we assessed consumers’ food waste intention by asking if they were willing to take leftovers home, which is a typical method in food waste research (Ellison and Lusk, 2018). Additionally, considering that consumers might interpret prompts as goals set for their consumption behavior, and previous research suggests that goal concreteness may influence an individual’s sense of self-efficacy (Locke et al., 1989), we measured self-efficacy as a potential alternative explanation. Finally, given that factors such as gender (Vizzoto et al., 2021), age (Filimonau et al., 2022), education level (Reynolds et al., 2019), income (Xu et al., 2020), and frequency of dining out (Wang et al., 2017) may affect consumer food waste, we also measured these variables in the experiment to control for their potential impact.

4.1 Method

The study followed a 2 (prompt concreteness: concrete vs. abstract) between-subjects design. We conducted a power analysis using G*Power 3.1, setting α = 0.05 and power = 0.95. The results indicated that at least 84 participants were required for a one-way ANOVA with two levels. Then we recruited 200 participants through the Credamo platform, and 193 participants passed attention check (66.3% female; M_age = 27.93, SD = 6.92).

Participants were randomly assigned to one of the two groups. First, both groups were asked to imagine entering a restaurant for lunch. To help participants immerse themselves in the dining context, we showed them the same interior photo of a Chinese casual dining restaurant. Next, participants were told that upon sitting down, they would see a prompt label placed on the table. Each group was then shown the corresponding image of the label. Given that the prompts in Study 1 rhymed in Chinese, which could influence participants’ impressions and attitudes toward the prompts, we deliberately made the content of the stimuli in Study 2 more direct and no longer rhyming. Specifically, the concrete prompt group saw the label that read “Pack up the leftover food to take away,” while the abstract prompt group saw the label that read “Cherish food and reduce waste.” The system required participants to view the label for at least 15 s. After this, we measured participants’ food waste intention following Gao et al. (2024). Specifically, We asked the participants what they would do with any leftover food after finishing their meal using three items: “I would throw away the remaining food” (1 = “strongly disagree,” 7 = “strongly agree”), “I would save the remaining food to eat tomorrow” (1 = “strongly disagree,” 7 = “strongly agree”; reverse-coded), and “Think more precisely about the scenario and your potential actions, what would you do?” (1 = “I would definitely throw away what’s left of the lunch,” 7 = “I would definitely save the leftovers to eat tomorrow”; reverse-coded).

We averaged the three items to form a food waste index (α = 0.89). Next, we measured participants’ moral justification by adapting the scale from Moore et al. (2012). Participants were asked to respond to three items: “It is okay to throw away leftovers to keep up appearances,” “It is alright to throw away the remaining food to keep a good dining experience,” and “Throwing away leftovers is sometimes necessary in order to avoid trouble.” (1 = strongly disagree; 7 = strongly agree; α = 0.84). And then we measured self-efficacy as a potential alternative explanation. Following Garvey et al. (2016), individuals provided a three-item measure of food saving self-efficacy (“On a scale from 1 to 7, how capable are you at food saving?,” “On a scale from 1 to 7, how experienced are you in reducing food waste?,” “On a scale from 1 to 7, How confident are you in your skills to reduce food waste?”; α = 0.75). To avoid interference from order effects, all items were presented in a random order. Subsequently, participants reported their gender, age, education level, income, and frequency of dining out as control variables. Finally, we conducted a manipulation check using the same items as in Study 1.

4.2 Results

4.2.1 Manipulation check

One-way ANOVA results showed that participants in the concrete prompt group rated the concreteness of the prompt significantly higher [M_concrete = 5.42, SD_concrete = 1.43; M_abstract = 2.75, SD_abstract = 1.41; F(1, 191) = 168.30, p < 0.001]. There were no significant differences in the ratings for the visual appeal (p = 0.115) and comprehensibility (p = 0.509) of the prompts.

4.2.2 Food waste intention

We conducted a one-way ANOVA on food waste intention with prompt concreteness (concrete vs. abstract) as the independent variable. The results showed that participants in the concrete prompt group had significantly lower food waste intention than those in the abstract prompt group [M_concrete = 2.29, SD_concrete = 1.15; M_abstract = 2.93, SD_abstract = 1.33; F (1, 191) = 12.93, p<0.001, η_p² = 0.063]. When gender, age, education level, income, and frequency of dining out were included as covariates in the ANOVA, the main effect remained significant [F (1, 188) = 9.34, p = 0.003, η_p² = 0.048] (see Table 2).

4.2.3 Mediation analyses

To test H2, we conducted 5,000 bootstraps using the PROCESS, Model 4. This is a commonly used non-parametric inference method, especially when the sample size is small, and it effectively tests whether hypotheses about mediation hold (Hayes, 2013). We defined prompt concreteness (1 = concrete, 2 = abstract) as the independent variable, moral justification as the mediator, food waste intention as the dependent variable, gender, age, education level, income, and frequency of dining out as the covariates. The results shown in Figure 4 indicated that prompt concreteness significantly affects moral justification (a = 0.41, SE = 0.19, t = 2.18, p = 0.030). Moral justification significantly influences the food waste intention (b = 0.63, SE = 0.05, t = 11.58, p < 0.001). Moral justification plays an indirect mediating role in the effect of prompt concreteness on consumer food waste [Direct effect = 0.30, Boot SE = 0.14, t = 2.12, 95% CI = (0.0212, 0.5778); Indirect effect = 0.26, Boot SE = 0.12, 95% CI = (0.0238, 0.5078)].

Figure 4

4.2.4 Alternative explanation analyses

We further conducted a mediation analysis (PROCESS Model 4; 5,000 bootstraps) to test the alternative explanations of self-efficacy. The results did not identify the mediating effects of self-efficacy [Indirect effect = 0.07, Boot SE = 0.08, 95% CI = (−0.0800, 0.2371)].

4.3 Discussion

Study 2 confirmed that prompt concreteness influences consumers’ food waste intentions, providing further evidence for the main effect (H1). Additionally, Study 2 showed that moral justification mediates the relationship between prompt concreteness and consumers’ food waste, while ruling out self-efficacy as an alternative explanation. We consider that there may be a potential gap between intention and actual behavior, and previous studies suggesting that, in addition to takeout decisions, over-ordering is another key behavioral factor contributing to food waste (Silvennoinen et al., 2015; Li et al., 2022). Based on this, study 3 will directly measure the effect of prompt concreteness on consumers’ ordering behavior to test the research hypotheses.

5 Study 3: the moderating role of trait reactance

Study 3 aims to further test the main effect and the mediating role of moral justification by changing the dining context and prompt labels. It also tests whether trait reactance moderates the effect of prompt concreteness on consumers’ food waste behavior in restaurants.

Previous studies have shown that a large proportion of food waste in restaurants results from over-ordering by customers (Silvennoinen et al., 2015). A lack of awareness of food portions and insufficient food-saving consciousness are two key reasons for over-ordering (Li et al., 2022). In other words, when customers are aware of the portion sizes, those who aim to avoid food waste will consciously reduce their order quantities (Yu et al., 2021). Therefore, in Study 3, we provided participants with a menu that listed the weight of each food item and directly measured their ordering behavior to assess the potential for food waste.

5.1 Method

The study followed a 2 (prompt concreteness: concrete vs. abstract) × 2 (trait reactance: high vs. Low) between-subjects design. We conducted a power analysis using G*Power 3.1, setting α = 0.05 and power = 0.95. The results indicated that at least 210 participants were required for a two-way ANOVA with four levels. Then we recruited 360 participants through the Credamo platform, and 357 participants passed attention check (67.5% female; M_age = 28.27, SD = 7.95).

To avoid participants guessing the purpose of the experiment or being influenced by unrelated factors, participants were informed that they needed to complete two unrelated tasks. Task one measured the participants’ trait reactance. Considering that too many questionnaire items might cause fatigue and affect the accuracy of the results, following Wendlandt and Schrader (2007), we selected four core items from the scale developed by Hong and Faedda (1996) to measure participants’ trait reactance: “I consider advice from others to be an intrusion”, “I become angry when my freedom of choice is restricted”, “When something is prohibited, I usually think that is exactly what I am going to do,” “I resist the attempts of others to influence me” (1 = strongly disagree; 7 = strongly agree; α = 0.76).

In task two, we manipulated the prompt concreteness and measured participants’ ordering behavior and moral justification. Similar to Study 2, participants were randomly assigned to either the concrete prompt group or the abstract prompt group. First, both groups were asked to imagine entering a certain brand of spicy hotpot restaurant (one of the most popular types of Chinese fast food). To help participants better immerse themselves in the dining scenario, both groups were shown the same image of the restaurant’s interior. Next, participants were informed that a prompt label was displayed on the restaurant’s table. Since Study 3 focuses on the impact of prompt concreteness on consumers’ over-ordering behavior, we redesigned the content of the concrete prompts. To avoid potential influence from irrelevant factors, we also adjusted the content of the abstract prompts to ensure both types had the same word count and layout. Specifically, the concrete prompt group saw the message “Ordering appropriately, start with me,” while the abstract prompt group saw the message “Saving food, start with me.” Following this, participants were provided with a menu labeled with prices and weights (e.g., mushrooms, 50 g, 2.28 RMB) and were asked to simulate ordering. After completing the ordering task, participants were asked to evaluate the reasonableness of the menu design (“Do you think the design of this menu is reasonable? 1 = Very unreasonable, 7 = Very reasonable”). Next, we measured participants’ moral justification using a scale adapted from Moore et al. (2012) (α = 0.74), and asked participants to report their gender, age, education level, income, and frequency of dining out as control variables. Finally, we conducted a manipulation check using the same items as in Study 1.

5.2 Results

5.2.1 Manipulation check

Manipulation check was carried out with a One-way ANOVA and the result showed that participants in the concrete prompt group rated the concreteness of the prompt significantly higher [M_concrete = 4.43, SD_concrete = 1.42; M_abstract = 3.34, SD_abstract = 1.41; F(1, 355) = 53.30, p < 0.001]. There were no significant differences in the ratings for the visual appeal (p = 0.768) and comprehensibility (p = 0.180) of the prompts. In addition, the evaluation of the reasonableness of the menu design was generally high, with no significant differences between the groups [M_concrete = 5.42, SD = 1.11; M_abstract = 5.34, SD = 1.21; F(1, 355) = 0.37, p < 0.542].

5.2.2 Over-ordering behavior

We conducted a one-way ANOVA on the weight of the food ordered with prompt concreteness (concrete vs. abstract) as the independent variable, participants’ gender, age, education level, income, and frequency of dining out as covariates. The results showed that participants in the concrete prompt group ordered significantly less than those in the abstract prompt group [M_concrete = 669.72 g, SD_concrete = 263.49; M_abstract = 748.66 g, SD_abstract = 297.46; F(1, 355) = 7.04, p = 0.008, η_p² = 0.019] (see Table 3).

5.2.3 Mediation analyses

We conducted a mediation analysis (PROCESS Model 4; 5,000 bootstraps) to test H2. We defined prompt concreteness (1 = concrete, 2 = abstract) as the independent variable, moral justification as the mediator, the weight of the food ordered as the dependent variable, gender, age, education level, income, and frequency of dining out as the covariates. The results showed that prompt concreteness significantly affects moral justification (a = 0.26, SE = 0.13, t = 2.06, p = 0.040). Moral justification significantly influences the weight of the food ordered (b = 65.89, SE = 11.90, t = 5.54, p < 0.001). Moral justification plays an indirect mediating role in the effect of prompt concreteness on over-ordering behavior (Direct effect = 57.60, Boot SE = 28.67, t = 2.01, 95% CI = [1.2154, 113.9921]; Indirect effect = 17.37, Boot SE = 8.82, 95% CI = [1.0980, 35.7835]).

5.2.4 Moderation analysis

To test H3, we conducted a moderation analysis (PROCESS Model 1; 5,000 bootstraps). We prompt concreteness (1 = concrete, 2 = abstract) as the independent variable,the weight of the food ordered as the dependent variable, and trait reactance as the moderator. The results indicated that trait reactance significantly moderates the effect of prompt concreteness on the weight of the food ordered (Effect = −71.34; LLCI = −124.6241, ULCI = −18.0592). Subsequently, we conducted the Johnson-Neyman test, and the results showed that when participants’ trait reactance was low (below the critical value of 3.55), prompt concreteness had a significant positive effect on the weight of the food ordered. However, when participants’ trait reactance was high (above the critical value of 3.55), the effect of prompt concreteness on the weight of the food ordered was not significant (see Figure 5).

Figure 5

5.3 Discussion

By changing the dining context and prompt labels, Study 3 replicate the results of prior studies, which showed that prompt concreteness influences consumer over-ordering behavior and moral justification plays a mediating role in it. In addition, Study 3 identified the moderating role of trait reactance, the effect of prompt concreteness on consumer over-ordering behavior was magnified for consumers’ trait reactance is low, while such effect disappeared for consumers’ trait reactance is high, supporting H3.

6 General discussion and implications

Prompts are considered an important measure to reduce consumer food waste in restaurants (Stöckli et al., 2018). However, we still know little about how to design the content of prompts to maximize their effectiveness in reducing food waste. To address the current research gap, this study reviews the literature on anti-food waste prompts. Specifically, previous research, based on the consistency principle of cognitive dissonance theory, suggests that prompts can help consumers build cognitive awareness of behavioral norms (Goldstein et al., 2008), thereby encouraging them to correct their behavior to avoid cognitive dissonance (Kallbekken and Sælen, 2013). However, this pathway struggles to explain why, at times, consumers remain indifferent to reducing food waste even after receiving prompts (Yu et al., 2016). We find that another path proposed by cognitive dissonance theory for avoiding dissonance can help explain this phenomenon: people can also align their cognition and behavior by justifying their actions at the cognitive level, which weakens their motivation to make behavioral changes (Mazar et al., 2008). Based on this, we hypothesize that, compared with concrete prompts, abstract prompts are often more general and broad, lacking clear behavioral requirements. This gives consumers more room to justify their food waste behaviors, which may lead to less effective outcomes in reducing food waste.

To test the proposed hypotheses, we conducted one field experiment and two laboratory experiments.

Study 1 is a field experiment aimed at gradually verifying whether prompt concreteness influences consumer food waste in restaurants. Different types of prompt labels are placed in a real restaurant, and the weight of leftover food is measured. The results show that, compared with no prompt, abstract prompts are somewhat effective in reducing consumer food waste; however, concrete prompts show better results in this regard, providing preliminary support for H1.

Study 2 is a laboratory experiment designed to examine the impact of prompt concreteness on consumers’ willingness to take leftovers and the mediating role of moral justification. A fictional scenario is created where participants dine at a restaurant with different types of prompts placed on their tables. The results indicate that compared with abstract prompts, participants who receive concrete prompts have a higher intention to take leftovers, and moral justification plays a mediating role in it, supporting H1 and H2.

Study 3 is a laboratory experiment aimed at examining the impact of prompt concreteness on consumers’ over-ordering behavior and the moderating role of trait reactance. The prompt labels are changed, and participants are provided a menu to measure their over-ordering behavior. The results show that prompt concreteness influences consumer over-ordering behavior, and moral justification plays a mediating role in it. Moreover, Study 3 finds that the effect of prompt concreteness on consumer over-ordering behavior is magnified for consumers with low trait reactance, while such effect disappears for consumers with high trait reactance, providing converging evidence for H3.

6.1 Theoretical contribution

This study makes three main contributions. Firstly, Although some scholars recognize that the design features of prompts may influence the effectiveness of reducing consumer food waste (Jiang et al., 2024), only a few studies have examined the impact of different information frameworks on prompt effectiveness (Dai and Gong, 2024). The potential impact of prompt concreteness on consumer food waste remains unclear. This study, based on the consistency principle of cognitive dissonance theory, explores the impact of prompt concreteness on consumer food waste in restaurants. By doing so, this study enriches the research on antecedents of consumer food waste in the field of marketing, and extends the research context of message concreteness.

Second, previous research has found that moral justification allows individuals to feel good about themselves even after engaging in immoral or norm-violating behaviors (Shalvi et al., 2015), which can further intensify such improper behaviors (Mazar et al., 2008). However, most studies have focused on topics such as adolescent violent behavior (Bandura et al., 1996) or unethical behavior in the workplace (Moore et al., 2012). Scant research has studied moral justification from a marketing or consumer perspective. This study reveals that moral justification is a key mechanism through which prompt concreteness influences consumer food waste in restaurants, thus enriching the existing literature on moral justification.

Third, this study confirms that the effect of prompt concreteness on consumer food waste in restaurants was magnified for consumers’ trait reactance is low, while such effect disappeared for consumers’ trait reactance is high. It defines the boundary for the impact of prompt concreteness on consumer food waste. This also provides an explanation for the phenomenon identified by Baca-Motes et al. (2013), where abstract commitments are more likely to be accepted by consumers than concrete commitments. This study provides a broader theoretical understanding of why consumers resist prompts.

6.2 Practical implications

Food waste has been a significant issue that governments worldwide have been working to address for many years. The United Nations has set a clear goal to halve global per capita food waste in retail and consumption by 2030 (UNEP, 2021). Consumer food waste in restaurants is a major contributor to food waste at the consumption stage (Xue et al., 2021), and prompts are considered an important measure to curb this source of waste (Stöckli et al., 2018). This study provides practical recommendations for designing prompts to help reduce consumer food waste in restaurants effectively.

Firstly, One common design feature of prompts-prompt concreteness-can influence the effectiveness of reducing consumer food waste in restaurants. The results of a field experiment show that in casual dining restaurants, concrete prompts can reduce consumer food waste by an additional 17.2% compared with abstract prompts. Two laboratory experiments, one in a casual dining setting and the other in a fast food setting, further indicate that, compared with abstract prompts, concrete prompts significantly increase consumers’ willingness to take leftovers and help curb over-ordering behavior. Therefore, in anti-food waste campaigns, marketers should ensure that the prompt content clearly involves specific behavioral reminders, such as packing leftovers or ordering appropriately, rather than merely making broad appeals like “cherish food” or “reduce waste.” This simple change in wording incurs little additional cost but can significantly enhance the effectiveness of the prompts in reducing consumer food waste in restaurants.

In addition, moral justification is the key mechanism through which prompt concreteness influences consumer food waste behavior in restaurants. This is fundamentally due to the lack of a clear understanding of the immorality of food waste in consumers’ common sense. Therefore, further public campaigns opposing food waste should be promoted to help consumers clearly understand the consumption behaviors that society advocates, and which improper behaviors lead to food waste. These efforts aim to make people internalize the norms and values regarding food waste, which will serve as an internal benchmark for individuals to compare their consumption behavior in restaurants.

Also, the effect of prompt specificity on consumer food waste primarily occurs among consumers with low trait reactance. Research in clinical psychology indicates that although trait reactance varies from individual to individual, it follows a trimodal trajectory throughout the life span (Miller et al., 2007). In other words, compared to adolescents or the elderly, middle-aged individuals tend to have lower trait reactance. Therefore, restaurants targeting middle-aged individuals as the main audience should adopt concrete prompts to reduce food waste.

6.3 Limitations

First, all the experiments in this study were conducted within the context of a single country, which introduces certain cultural limitations. For example, previous research suggests that the collectivist culture prevalent in many Asian regions may make consumers more accepting of external control (Miron and Brehm, 2006), potentially amplifying the impact of prompt concreteness on reducing food waste. In contrast, consumers in individualistic cultures may be less inclined to accept constraints (Miron and Brehm, 2006), and more likely to resist concrete prompts, which could weaken the difference in effectiveness between concrete and abstract prompts in reducing food waste. Second, according to the definition by Stöckli et al. (2018), prompts can be delivered to consumers not only in written form but also verbally by service staff. From the perspective of impression management theory, it can be inferred that the impact of prompts delivered by service staff might differ from that of written prompts. Thus, whether the conclusions of this study apply to verbal prompts remains to be further explored. Additionally, while this study controlled for the potential influence of consumers’ personal traits on food waste, existing research has shown that external factors such as restaurant size (Wang et al., 2017) or service quality (Li et al., 2022) can also affect consumer food waste. Future research should consider incorporating these factors into the control variables for a more comprehensive investigation. Moreover, the prompts used in this study’s experiments were static images and text. However, in real-life settings, prompts are also presented in dynamic formats such as animations or short videos. The impact of different delivery mediums on the effectiveness of these prompts warrants further exploration.

References

• 1

  AgarwalA. (2018). A pioneering tactic for sustainable waste management: design of a cost-effective recycling machine. In MATEC Web of Conferences (172, 03003). EDP Sciences. doi: 10.1051/matecconf/201817203003

  • CrossRef
  • Google Scholar

• 2

  Baca-MotesK.BrownA.GneezyA.KeenanE. A.NelsonL. D. (2013). Commitment and behavior change: evidence from the field. J. Consum. Res.39, 1070–1084. doi: 10.1086/667226

  • CrossRef
  • Google Scholar

• 3

  BanduraA.BarbaranelliC.CapraraG. V.PastorelliC. (1996). Mechanisms of moral disengagement in the exercise of moral agency. J. Pers. Soc. Psychol.71, 364–374. doi: 10.1037/0022-3514.71.2.364

  • CrossRef
  • Google Scholar

• 4

  BeutlerL. E.HarwoodT. M.MichelsonA.SongX. (2011). Resistance/reactance level. J. Clin. Psychol.67, 133–142. doi: 10.1002/jclp.20753,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  BuboltzW. C.Jr.ThomasA.DonnellA. J. (2002). Evaluating the factor structure and internal consistency reliability of the therapeutic reactance scale. J. Counsel. Dev.80, 120–125. doi: 10.1002/j.1556-6678.2002.tb00174.x

  • CrossRef
  • Google Scholar

• 6

  China (2021) Anti-food waste law of the People's Republic of China Available online at: https://www.gov.cn/xinwen/2021-04/29/content_5604029.htm (Accessed September 20, 2025).

  • Google Scholar

• 7

  China (2023) China food and nutrition development report 2022 Available online at: https://ifnd.caas.cn/kxyj/kycg/35b045b8eb0a4d48aa5f147448ef1114.htm

  • Google Scholar

• 8

  DaiJ.GongS. (2024). Sustainable messaging strategies and consumer food waste: the congruence effect between message framing and state anxiety. J. Retail. Consum. Serv.79:103817. doi: 10.1016/j.jretconser.2024.103817

  • CrossRef
  • Google Scholar

• 9

  EllisonB.LuskJ. L. (2018). Examining household food waste decisions: a vignette approach. Appl. Econ. Perspect. Policy40, 613–631. doi: 10.1093/aepp/ppx059

  • CrossRef
  • Google Scholar

• 10

  FestingerL. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press.

  • Google Scholar

• 11

  FilimonauV.MatyakubovU.AllonazarovO.JurayevA. (2022). Food waste and its management in restaurants of a transition economy: an exploratory study of Uzbekistan. Sustain. Prod. Consum.29, 25–35. doi: 10.1016/j.spc.2021.09.018

  • CrossRef
  • Google Scholar

• 12

  GaoH.JiaH.GuoB. (2024). Resources available for me versus us: implications for mitigating consumer food waste. J. Mark. Res.61, 619–637. doi: 10.1177/00222437231162615

  • CrossRef
  • Google Scholar

• 13

  GarveyA. M.GermannF.BoltonL. E. (2016). Performance brand placebos: how brands improve performance and consumers take the credit. J. Consum. Res.42, 931–951. doi: 10.1093/jcr/ucv094

  • CrossRef
  • Google Scholar

• 14

  GoldsteinN. J.CialdiniR. B.GriskeviciusV. (2008). A room with a viewpoint: using social norms to motivate environmental conservation in hotels. J. Consum. Res.35, 472–482. doi: 10.1086/586910

  • CrossRef
  • Google Scholar

• 15

  GrandpreJ.AlvaroE. M.BurgoonM.MillerC. H.HallJ. R. (2003). Adolescent reactance and anti-smoking campaigns: a theoretical approach. Health Commun.15, 349–366. doi: 10.1207/S15327027HC1503_6,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  HamermanE. J.RudellF.MartinsC. M. (2018). Factors that predict taking restaurant leftovers: Strategies for reducing food waste. J. Consum. Behav.17, 94–104. doi: 10.1002/cb.1700

  • CrossRef
  • Google Scholar

• 17

  HayesA. F. (2013). Introduction to mediation, moderation, and conditional process analysis. J. Educ. Meas.51, 335–337. doi: 10.1111/jedm.12050

  • CrossRef
  • Google Scholar

• 18

  HongS. M.FaeddaS. (1996). Refinement of the Hong psychological reactance scale. Educ. Psychol. Meas.56, 173–182. doi: 10.1177/0013164496056001014

  • CrossRef
  • Google Scholar

• 19

  HseeC. K.YuF.ZhangJ.ZhangY. (2003). Medium maximization. J. Consum. Res.30, 1–14. doi: 10.1086/374702

  • CrossRef
  • Google Scholar

• 20

  JiangS.ChenH.ShanP.ShangZ. (2024). Efficacy of informational intervention on food waste: evidence from a randomized controlled trial. J. Clean. Prod.443:141106. doi: 10.1016/j.jclepro.2024.141106

  • CrossRef
  • Google Scholar

• 21

  KallbekkenS.SælenH. (2013). Nudging’ hotel guests to reduce food waste as a win–win environmental measure. Econ. Lett.119, 325–327. doi: 10.1016/j.econlet.2013.03.019

  • CrossRef
  • Google Scholar

• 22

  KamanG. S.BozkurtI.BölükbaşR.DedeoğluB. B. (2024). The strategy food waste in restaurants: a systematic literature review. Trends Food Sci. Technol.151:104625. doi: 10.1016/j.tifs.2024.104625

  • CrossRef
  • Google Scholar

• 23

  KarnoM. P.LongabaughR. (2005). Less directiveness by therapists improves drinking outcomes of reactant clients in alcoholism treatment. J. Consult. Clin. Psychol.73, 262–267. doi: 10.1037/0022-006X.73.2.262,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  LiC.BremerP.HarderM. K.MiJ. (2022). A systematic review of food loss and waste in China: quantity, impacts and mediators. J. Environ. Manag.303:114092. doi: 10.1016/j.jenvman.2021.114092,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  LiaoC.LuoJ.WangS.LiR. (2025). How live-streaming commerce influences consumers' impulse buying of near expiry-date food: implications for reducing food waste. Front. Sustain. Food Syst.9:1570483. doi: 10.3389/fsufs.2025.1570483

  • CrossRef
  • Google Scholar

• 26

  LockeE. A.ChahD. O.HarrisonS.LustgartenN. (1989). Separating the effects of goal specificity from goal level. Organ. Behav. Hum. Decis. Process.43, 270–287. doi: 10.1016/0749-5978(89)90053-8

  • CrossRef
  • Google Scholar

• 27

  MazarN.AmirO.ArielyD. (2008). The dishonesty of honest people: a theory of self-concept maintenance. J. Mark. Res.45, 633–644. doi: 10.1509/jmkr.45.6.633

  • CrossRef
  • Google Scholar

• 28

  MillerC. H.LaneL. T.DeatrickL. M.YoungA. M.PottsK. A. (2007). Psychological reactance and promotional health messages: the effects of controlling language, lexical concreteness, and the restoration of freedom. Hum. Commun. Res.33, 219–240. doi: 10.1111/j.1468-2958.2007.00297.x

  • CrossRef
  • Google Scholar

• 29

  MironA. M.BrehmJ. W. (2006). Reactance theory-40 years later. Z. Sozialpsychol.37, 9–18. doi: 10.1024/0044-3514.37.1.9

  • CrossRef
  • Google Scholar

• 30

  MooreC.DetertJ. R.TreviñoL. K.BakerV. L.MayerD. M. (2012). Why employees do bad things: moral disengagement and unethical organizational behavior. Pers. Psychol.65, 1–48. doi: 10.1111/j.1744-6570.2011.01237.x

  • CrossRef
  • Google Scholar

• 31

  PackardG.BergerJ. (2021). How concrete language shapes customer satisfaction. J. Consum. Res.47, 787–806. doi: 10.1093/jcr/ucaa038

  • CrossRef
  • Google Scholar

• 32

  ReynoldsC.GoucherL.QuestedT.BromleyS.GillM.ParryA. (2019). Consumption-stage food waste reduction interventions–what works and how to design better interventions. Food Policy83, 7–27. doi: 10.1016/j.foodpol.2019.01.009

  • CrossRef
  • Google Scholar

• 33

  Reynolds-TylusT. (2019). Psychological reactance and persuasive health communication: a review of the literature. Front. Commun.4:56. doi: 10.3389/fcomm.2019.00056

  • CrossRef
  • Google Scholar

• 34

  RoeB. E.QiD.BenderK. E. (2020). Some issues in the ethics of food waste. Physiol. Behav.219:112860. doi: 10.1016/j.physbeh.2020.112860,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  SchellekensG. A. C.VerleghP. W. J.SmidtsA. (2010). Language abstraction in word of mouth. J. Consum. Res.37, 207–223. doi: 10.1086/651240

  • CrossRef
  • Google Scholar

• 36

  ScherhauferS.MoatesG.HartikainenH.WaldronK.ObersteinerG. (2018). Environmental impacts of food waste in Europe. Waste Manag.77, 98–113. doi: 10.1016/j.wasman.2018.04.038,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  SchultzP. W.OskampS.MainieriT. (1995). Who recycles and when? A review of personal and situational factors. J. Environ. Psychol.15, 105–121. doi: 10.1016/0272-4944(95)90019-5

  • CrossRef
  • Google Scholar

• 38

  ShalviS.GinoF.BarkanR.AyalS. (2015). Self-serving justifications: doing wrong and feeling moral. Curr. Dir. Psychol. Sci.24, 125–130. doi: 10.1177/0963721414553264

  • CrossRef
  • Google Scholar

• 39

  SilvennoinenK.HeikkiläL.KatajajuuriJ. M.ReinikainenA. (2015). Food waste volume and origin: case studies in the Finnish food service sector. Waste Manag.46, 140–145. doi: 10.1016/j.wasman.2015.09.010,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40

  StöckliS.DornM.LiechtiS. (2018). Normative prompts reduce consumer food waste in restaurants. Waste Manag.77, 532–536. doi: 10.1016/j.wasman.2018.04.047,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  UNEP (2021) Measuring progress: environment and the SDGs Available online at: https://www.unep.org/events/conference/measuring-progress-environment-and-sdgs (Accessed September 20, 2025).

  • Google Scholar

• 42

  UNEP (2024) Food waste index report 2024 Available online at: https://www.unep.org/resources/publication/food-waste-index-report-2024 (Accessed September 20, 2025).

  • Google Scholar

• 43

  VasquezK.OswaldD. L.HammerA. (2019). Being dishonest about our prejudices: moral dissonance and self-justification. Ethics Behav.29, 382–404. doi: 10.1080/10508422.2019.1568877

  • CrossRef
  • Google Scholar

• 44

  VisschersV. H. M.GundlachD.BerettaC. (2020). Smaller servings vs. information provision: results of two interventions to reduce plate waste in two university canteens. Waste Manag.103, 323–333. doi: 10.1016/j.wasman.2019.12.046,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  VizzotoF.TessitoreS.TestaF.IraldoF. (2021). Plate waste in foodservice outlets: revealing customer profiles and their support for potentially contentious measures to reduce it in Italy. Resour. Conserv. Recycl.174:105771. doi: 10.1016/j.resconrec.2021.105771

  • CrossRef
  • Google Scholar

• 46

  WangL. E.LiuG.LiuX.LiuY. (2017). The weight of unfinished plate: a survey based characterization of restaurant food waste in Chinese cities. Waste Manag.66, 3–12. doi: 10.1016/j.wasman.2017.04.007

  • CrossRef
  • Google Scholar

• 47

  WendlandtM.SchraderU. (2007). Consumer reactance against loyalty programs. J. Consum. Mark.24, 293–304. doi: 10.1108/07363760710773111

  • CrossRef
  • Google Scholar

• 48

  WhiteK.HabibR.HardistyD. J. (2019). How to SHIFT consumer behaviors to be more sustainable: a literature review and guiding framework. J. Mark.83, 22–49. doi: 10.1177/0022242919825649

  • CrossRef
  • Google Scholar

• 49

  WhitehairK. J.ShanklinC. W.BrannonL. A. (2013). Written messages improve edible food waste behaviors in a university dining facility. J. Acad. Nutr. Diet.113, 63–69. doi: 10.1016/j.jand.2012.09.015,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  WRAP (2020) Resource pack for hospitality and food service sector Available online at: http://www.wrap.org.uk/sites/files/wrap/UK%20LFHWHospitalityResourcePack_0.pdf (Accessed September 20, 2025).

  • Google Scholar

• 51

  XiaoM.RazzaqA.QingP.IqbalW. (2024). Reducing food waste and promoting sustainable consumption: the role of message framing and controllability attributions in ugly produce marketing. Front. Sustain. Food Syst.8:1416142. doi: 10.3389/fsufs.2024.1416142

  • CrossRef
  • Google Scholar

• 52

  XuZ.ZhangZ.LiuH.ChenK. (2020). Food-away-from-home plate waste in China: preference for variety and quantity. Food Policy97:101918. doi: 10.1016/j.foodpol.2020.101918

  • CrossRef
  • Google Scholar

• 53

  XueL.LiuX.LuS.ChengG.HuY.LiuJ.et al. (2021). China’s food loss and waste embodies increasing environmental impacts. Nature Food2, 519–528. doi: 10.1038/s43016-021-00317-6,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  YuZ.JuX.BaiL.ZhaoG. (2021). Consumer's over-ordering behavior at restaurant: understanding the important roles of interventions from waiter and ordering habits. Appetite160:105092. doi: 10.1016/j.appet.2020.105092,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  YuW.WenY. U.AhmedA.ShangX. F. (2016). Food packing: a case study of dining out in Beijing. J. Integr. Agric.15, 1924–1931. doi: 10.1016/S2095-3119(15)61282-5

  • CrossRef
  • Google Scholar

• 56

  ZhangD.ZhangP.XueL.WuX. (2024). Food waste and its associated environmental impacts in workplace buffet-style canteens in Beijing, China. Front. Sustain. Food Syst.8:1455756. doi: 10.3389/fsufs.2024.1455756

  • CrossRef
  • Google Scholar