The Effect of Reward Strategies on Consumers’ Continuance Intention Toward Digital Low-Carbon Applications

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors, Thank you for the opportunity to review this article, which is the result of a great deal of work.

The authors address a topic that is extremely relevant today, both in the scientific and practical fields.

The language used in the article is correct and raises no objections.

The conclusions of the work are supported by both the literature on the subject and the authors' own research. They are also correctly presented and embedded in the structure of the work.

I have no critical comments and recommend that the article be published.

Author Response

Thank you for your insightful comment.

 

Comments: The language used in the article is correct and raises no objections.The conclusions of the work are supported by both the literature on the subject and the authors' own research. They are also correctly presented and embedded in the structure of the work. I have no critical comments and recommend that the article be published.

Response: We sincerely appreciate your thorough review and endorsement of this manuscript. In response to your positive evaluation, we have refined the language throughout to better align with the stylistic conventions of this journal. We wish you continued success in your scholarly endeavors.

 

Once again, we would like to thank you for the above mentioned highly constructive suggestions regarding our manuscript. Besides the suggestions you highlighted above, we have also conducted a comprehensive review and streamlined the manuscript. The introduction, theoretical background and hypothesis development part of the original paper were written up to line 391. After revision, these sections now have a total length of 298 lines.

Reviewer 2 Report

Comments and Suggestions for Authors

The presented manuscript is a rigorous statistical study of the impact of reward strategies on users of digital low-carbon apps, with the noble goal of identifying ways to sustain customer/user interest in the corresponding project (e.g., using the service equals planting a tree). The introductory reasoning is very clear, accompanied by numerous examples, and conveys the research goal to the reader. The literature review (aka the theoretical concepts) is equally adequate, as the authors choose a relatively straightforward psychological model and gradually embed hypotheses within, resulting in a transparent and testable framework. The results are clearly presented, and the manuscript offers a concise discussion section that compares the obtained results with the previous literature.

In its current form, the paper has only two significant flaws. First of all, it is the absence of a coherent methodological section, which would explain by summarizing a) what studies (experiments) were conducted, b) in which cities and under what conditions, c) was there a control group, d) what data was collected (responses from forms, if so, which ones), e) what proxies were used for the theoretical variables, and, finally, f) what statistical methods were used in all and each case (t-tests, Chi-square tests, ANCOVA, etc.). Technically, most of this information is already present in the text, but it is scattered across studies, not forming a clear, consistent, and coherent research picture. The reviewer, therefore, asks that this issue be addressed in the final version.

The second issue is the abstract. It equally lacks a clear methodology, but most of all, its claims are too bold. The reviewer asks the authors to be more "modest" and limit their claims to evidence for consumers in certain provinces/cities, not the whole world.

Author Response

We greatly appreciate your constructive suggestions. Below, we provide a detailed account of our revisions.

 

Comments 1: It is the absence of a coherent methodological section, which would explain by summarizing a) what studies (experiments) were conducted, b) in which cities and under what conditions, c) was there a control group, d) what data was collected (responses from forms, if so, which ones), e) what proxies were used for the theoretical variables, and, finally, f) what statistical methods were used in all and each case (t-tests, Chi-square tests, ANCOVA, etc.). Technically, most of this information is already present in the text, but it is scattered across studies, not forming a clear, consistent, and coherent research picture.

Response 1: We concur with your assessment and have accordingly incorporated a comprehensive methodological overview at the conclusion of the Introduction section. This newly added section systematically delineates the research questions, methodological approaches, statistical techniques, participant recruitment procedures and criteria, operational metrics and variables, as well as the objectives and experimental designs of all three studies.

Specifically, we have articulated this in the manuscript as follows:

Accordingly, This research investigates: (1) whether material or immaterial rewards more effectively enhance continuance intention in DLCA contexts; (2) whether reward timing and orientation moderate the effect of reward type, and through what mechanisms. Using a quantitative experimental design, we recruited 1,062 Chinese participants through Credamo (https://www.credamo.com/#/, accessed on 29 January 2026). Participants were drawn from provinces across China. Upon completion of the experiment, each participant received compensation of 1 RMB. Following exposure to experimental stimuli, participants completed manipulation check items, control variable items, attitude items, DLCAs continuance intention items, and demographic items. Statistical analyses included t-tests, chi-square tests, ANCOVA, and PROCESS. Experiment 1 (N = 257) employed a between-group design with reward type (immaterial rewards vs. material rewards). Experiment 2 (N = 268) utilized a 2 (reward type: immaterial rewards vs. material rewards) × 2 (reward timing: delayed reward vs. immediate reward) between-subjects experiment. Experiment 3 (N = 409) conducted a 2 (reward type: immaterial rewards vs. material rewards) × 2 (reward timing: delayed reward vs. immediate reward) × 2 (reward orientation: altruistic rewards vs. self-oriented rewards) between-subjects experiment. Effects replicated across three distinct DLCA contexts. This research elucidates how reward strategies influence DLCAs continuance intention, extends construal level theory , and provides actionable guidance for firms designing sustainable incentive systems.

 

Comments 2: Abstract equally lacks a clear methodology, but most of all, its claims are too bold. The reviewer asks the authors to be more “modest” and limit their claims to evidence for consumers in certain provinces/cities, not the whole world.

Response 2: We appreciate this valuable feedback and have revised the Abstract to explicitly specify that the scenario-based experiments were conducted with Chinese consumers. Given that the detailed participant recruitment procedures and conditions are now comprehensively elaborated in the methodological section of the Introduction, we have opted to avoid redundancy in the Abstract.

 

Once again, we would like to thank you for the above mentioned highly constructive suggestions regarding our manuscript. Besides the suggestions you highlighted above, we have also conducted a comprehensive review and streamlined the manuscript. The introduction, theoretical background and hypothesis development part of the original paper were written up to line 391. After revision, these sections now have a total length of 298 lines.

Reviewer 3 Report

Comments and Suggestions for Authors

Thanks for the opportunity to review this article. Kindly look at below points.

1. I advise rephrasing the title to look like “The Effect of Reward Strategies on Consumers’ Continuance Intention toward Low-Carbon Digital Applications”.
2. Is there a need to use the word, study1, study2, study3, they are confusing.
3. The article need to present their findings, result and reliability and validity and other important information. It is very difficult to trace the detailed information this way.
4. Discussion should always be split from the result, these are two separate sections.
5. The figure of the model needs to show directions of the hypotheses, name hypotheses on each arrow to enable readers to better understand the directions of the hypotheses.
6. I wonder if the term “application” can be substituted by “systems” as it implies greater type of work unlike applications.
7. If possible develop a separate section for the operational definitions of the concepts of the study.
8. Last section of the introduction section should include an organization of the research.
9. I would like to see the argument of the authors at the last part of every section developed and before the established hypotheses. It is important to see how authors anticipate.
10. The research methodology needs further elaboration on key elements of the research, type of the research, sample, location and other important elements.
11. Please show the slop graphs of the moderation if possible.
12. Please elaborate more on the limitations of the study and offer valid future research guidelines.
13. Important question is that, why is attitude the only mediator? Why not perceived value, intrinsic motivation, moral satisfaction, or habit?
14. Why does reward type directly influence continuance without passing through attitude or cognition? This violates most behavioral theories. Please defend the model.
15. Why should altruistic + delayed be better than self-oriented + immediate? Which combinations are optimal and why? Please support.

All the best

Author Response

We greatly appreciate your constructive suggestions. Below, we provide a detailed account of our revisions.

 

Comments 1: I advise rephrasing the title to look like “The Effect of Reward Strategies on Consumers’ Continuance Intention toward Low-Carbon Digital Applications”.

Response 1: We concur with your recommendation to recast the title as a declarative statement, which indeed more clearly communicates the research focus.

 

Comments 2: Is there a need to use the word, study1, study2, study3, they are confusing.

Response 2: We acknowledge this concern regarding potential reader confusion. In the revised Abstract, we have replaced the “Study 1,” “Study 2,” and “Study 3” nomenclature with transitional phrases such as “Furthermore” and “Additionally” to more effectively convey the logical progression among the three findings. Additionally, throughout the main text, we have substituted “Experiment” for the previous “Study” terminology to enhance clarity.

 

Comments 3: The article need to present their findings, result and reliability and validity and other important information.

Response 3: We concur with this observation and have supplemented the concluding paragraph of the Introduction with relevant elaboration. Specifically, to ensure the reliability and validity of our experimental findings, we designed 3 experiments that simulate distinct usage contexts, namely, a bank’s personal carbon account platform, a Mobility-as-a-Service (MaaS), and a food surplus blind box platform. This variation in experimental contexts enabled cross-validation of our hypotheses. For instance, Experiment 2, conducted in the MaaS context, successfully replicated the findings from Hypotheses 1 and Hypotheses 2 initially validated in the personal carbon account platform context in Experiment 1.

 

Comments 4: Discussion should always be split from the result, these are two separate sections.

Response 4: We concur with your recommendation. Accordingly, the former “7. Discussion and Conclusions” section has been restructured into two distinct subsections: “7.1. Conclusions” and “7.2. Discussion,” thereby clearly delineating the presentation of results from their interpretive discussion.

 

Comments 5: The figure of the model needs to show directions of the hypotheses, name hypotheses on each arrow to enable readers to better understand the directions of the hypotheses.

Response 5: We concur with your suggestion. Accordingly, we have annotated each directional pathway in “Figure 1. Conceptual Framework” with its corresponding hypothesis label.

 

Comments 6: I wonder if the term “application” can be substituted by “systems” as it implies greater type of work unlike applications.

Response 6: We sincerely appreciate this thoughtful observation. We fully recognize the broader applicability that “systems” offers, as it indeed encompasses a wider array of information system types. However, following careful deliberation, we have determined that retaining the term “application” is more theoretically precise for our research context. Our rationale centers on the fact that our research context and theoretical framework are integrally tied to the distinctive characteristics of mobile applications. Unlike the generic term “systems,” “application” captures the mobile, instantaneous, and personalized nature of the technology, which directly influences user engagement experiences and the design of reward strategies. For example, location-based check-in incentives, social sharing bonus points, and real-time carbon reduction notifications represent intervention mechanisms uniquely enabled by mobile applications. Substituting “systems” might obscure these critical contextual boundaries and diminish the situational specificity of our theoretical model. Notably, in extant literature, Paschmann et al. (2024), published in the Journal of Marketing Research, investigated the effects of different reward types (gamification rewards versus value-based rewards) on both the quality and quantity of user engagement in the context of enhancing app user participation. Thus, employing “application” as our focal term ensures better conceptual alignment with the reward strategy literature.

 

Comments 7: If possible develop a separate section for the operational definitions of the concepts of the study.

Response 7: We appreciate this recommendation. To enhance conceptual clarity and theoretical rigor, we have restructured the original “2. Theoretical Framework and Hypotheses Development” into two discrete sections: “2. Theoretical Background” and “3. Hypotheses Development.”

 

Comments 8 and 10: Last section of the introduction section should include an organization of the research. 3.The research methodology needs further elaboration on key elements of the research, type of the research, sample, location and other important elements.

Response 8 and 10: We appreciate this constructive feedback. Accordingly, we have incorporated a comprehensive methodological overview at the conclusion of the Introduction section. This newly added section systematically presents the research questions, methodological approaches, statistical techniques, participant recruitment procedures and criteria, operational metrics and variables, as well as the objectives and experimental designs of all three studies.

Specifically, we have articulated this in the manuscript as follows:

Accordingly, This research investigates: (1) whether material or immaterial rewards more effectively enhance continuance intention in DLCA contexts; (2) whether reward timing and orientation moderate the effect of reward type, and through what mechanisms. Using a quantitative experimental design, we recruited 1,062 Chinese participants through Credamo (https://www.credamo.com/#/, accessed on 29 January 2026). Participants were drawn from provinces across China. Upon completion of the experiment, each participant received compensation of 1 RMB. Following exposure to experimental stimuli, participants completed manipulation check items, control variable items, attitude items, DLCAs continuance intention items, and demographic items. Statistical analyses included t-tests, chi-square tests, ANCOVA, and PROCESS. Experiment 1 (N = 257) employed a between-group design with reward type (immaterial rewards vs. material rewards). Experiment 2 (N = 268) utilized a 2 (reward type: immaterial rewards vs. material rewards) × 2 (reward timing: delayed reward vs. immediate reward) between-subjects experiment. Experiment 3 (N = 409) conducted a 2 (reward type: immaterial rewards vs. material rewards) × 2 (reward timing: delayed reward vs. immediate reward) × 2 (reward orientation: altruistic rewards vs. self-oriented rewards) between-subjects experiment. Effects replicated across three distinct DLCA contexts. This research elucidates how reward strategies influence DLCAs continuance intention, extends construal level theory , and provides actionable guidance for firms designing sustainable incentive systems.

 

Comments 9: I would like to see the argument of the authors at the last part of every section developed and before the established hypotheses. It is important to see how authors anticipate.

Response 9: We sincerely appreciate this valuable suggestion. Your observation regarding the necessity of strengthening the theoretical argumentation preceding each hypothesis is indeed critical. In accordance with your recommendation, we have established a dedicated “Hypotheses Development” section, which is organized into three subsections that systematically elaborate on: (1) the impact of reward type on DLCAs continuance intention ; (2) the moderating role of reward timing; and (3) the re-moderating effect of reward orientation. This restructuring yields 6 hypotheses in total. To substantiate these hypotheses, our theoretical framework is primarily grounded in Construal Level Theory (CLT).

 

Comments 11: Please show the slop graphs of the moderation if possible.

Response 11: We greatly appreciate this constructive suggestion. In the extant literature, scholars predominantly employ bar charts to report moderation analysis results. Consistent with this convention, we have presented bar charts illustrating the moderating effect of reward timing and the re-moderating effect of reward orientation in Figure 3 and Figure 4, respectively. To enhance the clarity of the moderation direction, we have additionally annotated the statistical significance levels between groups within each bar chart. Should you find the current bar chart presentation insufficient for demonstrating the moderation direction, we would be pleased to supplement the analysis with slope graphs upon request.

 

Comments 12: Please elaborate more on the limitations of the study and offer valid future research guidelines.

Response 12: We sincerely appreciate this valuable suggestion. We acknowledge that the original limitations and future research directions section was unduly brief and insufficiently comprehensive. Accordingly, we have substantially expanded this section at the conclusion of the manuscript.

Specifically, we have articulated this in the manuscript as follows:

While this study systematically examines the mechanisms through which reward strategies influence DLCAs continuance intention, several limitations warrant acknowledgment. First, regarding methodology, this study employs experimental scenarios to simulate user interactions with DLCAs. Although this approach effectively controls for extraneous variables and ensures internal validity, certain discrepancies may exist between experimental scenarios and authentic usage contexts. Future research could complement experimental findings with field studies using actual DLCAs, employing surveys, in-depth interviews, and behavioral tracking to collect data, thereby providing a comprehensive analysis of reward strategy effectiveness in naturalistic settings and enhancing external validity and practical applicability.

Second, concerning variable measurement, this study assessed continuance intention through self-report measures. While this approach facilitates efficient data collection, self-reports are susceptible to social desirability bias—participants may overestimate their continuance intention to present themselves as environmentally conscious, resulting in potential discrepancies between reported intentions and actual behaviors. Future research could employ behavioral tracking methodologies, recording actual usage data to measure continuance behavior, thereby enhancing measurement precision.

Finally, with respect to the theoretical model, this research focuses on three core dimensions of reward strategy without incorporating other potentially influential variables beyond reward strategies (e.g., environmental consciousness intensity, app usage habits, personal values, professional background, and personal interests). Future research could integrate these variables into the theoretical framework to further enrich our understanding of DLCAs continuance intention.

 

Comments 13: Important question is that, why is attitude the only mediator? Why not perceived value, intrinsic motivation, moral satisfaction, or habit?

Response 13: We greatly appreciate this highly insightful question. Our decision to designate attitude as the core and sole mediating variable is predicated upon a comprehensive consideration of two dimensions: the central research logic and theoretical rationale of this study. The detailed justification is provided below.

(1) The core logic of this research is structured around the "Reward Strategy (S) – Attitude (O) – Continuance Intention (R)" framework, focusing on how external reward stimuli influence consumers’ subsequent continuance intention through their holistic psychological evaluation of DLCAs. Within the domain of behavioral intention research, attitude constitutes a classical mediating variable linking external stimuli to behavioral intention—a theoretical logic firmly supported by the Theory of Planned Behavior (TPB).

(2) As a comprehensive psychological evaluative construct, “attitude” is particularly well-suited to accommodate the multi-boundary interaction analytical framework of this study. Our research constructs a three-way moderated mediation model involving “reward type × reward timing × reward orientation,” with the central objective of investigating how contextual variables alter the influence on consumer attitudes, which subsequently transmits to continuance intention. In contrast, constructs such as perceived value, intrinsic motivation, and moral satisfaction represent more unidimensional psychological perceptions. For instance, perceived value focuses on benefit-cost evaluation, while moral satisfaction centers on the moral experience of prosocial behavior—neither of which can adequately integrate the comprehensive psychological influences arising from the multi-contextual interactions in our study. Habit, as a behavioral-level variable, requires long-term usage behavior to form. Given that our study focuses on “continuance intention” rather than “habit formation following actual usage behavior,” this variable does not align with the dependent variable stage of our research and was therefore excluded from the mediation framework.

 

Comments 14: Why does reward type directly influence continuance without passing through attitude or cognition? This violates most behavioral theories. Please defend the model.

Response 14: We sincerely appreciate this profound and theoretically significant question. Our model does not exclude attitude as a mediator; rather, it simultaneously posits 2 hypotheses: the direct effect of reward type on DLCAs continuance intention (H1) and the indirect effect mediated through attitude (H2). In both Experiment 1 and Experiment 2, our findings consistently demonstrate that the direct effect of reward type on DLCAs continuance intention is statistically significant, while attitude serves as a partial mediator.

 

Comments 15: Why should altruistic + delayed be better than self-oriented + immediate? Which combinations are optimal and why? Please support.

Response 15: We greatly appreciate this critically important question. The conclusion derived from our model does not advocate for a singular “optimal combination”; rather, it reveals that the optimal combination is contingent upon reward orientation (self-oriented rewards vs. altruistic rewards). According to our Hypotheses H5a and H5b, no universally superior “absolute optimal” combination exists. Instead, reward effectiveness depends on whether the reward type, timing, and orientation are congruent—congruence enhances effectiveness, while incongruence diminishes it.

Under self-oriented reward conditions, we hypothesize that “immaterial rewards + delayed reward + self-oriented rewards” and “material rewards + immediate reward + self-oriented rewards” constitute the optimal combinations. The rationale is as follows: When delayed reward strategies are adopted, the abstract value of immaterial rewards fits the greater temporal distance of the delayed context, thereby enhancing continuance intention by reinforcing long-term value identification with low-carbon goals. Conversely, immediate rewards emphasize short-term gains, and material rewards—owing to their direct economic value—better satisfy consumers’ demand for immediate benefits and enhance instant gratification, thus demonstrating greater advantage in immediate reward contexts. Therefore, when the level of construal of external rewards fits consumers’ psychological distance, consumers exhibit greater willingness to continue using DLCAs.

Conversely, under altruistic reward conditions, “altruistic rewards + immaterial rewards” represents the optimal combination—that is, regardless of whether the reward is delayed or immediate, immaterial rewards more effectively stimulate DLCAs continuance intention. The rationale is as follows: Since both altruistic rewards and immaterial rewards constitute high-level-construal stimuli, this construal fit strengthens consumers’ social identification and goal commitment. Existing findings also demonstrate that immaterial rewards create synergistic effects with the abstract value of altruistic goals, significantly enhancing consumer loyalty. Under the delayed–immaterial reward condition, both elements fit at high-level construals, naturally enhancing consumers’ continuance intention. Under the immediate–immaterial reward condition, although immediate rewards constitute a low-level-construal stimulus, immaterial rewards inherently carry symbolic meaning at the spiritual level, which reinforces consumers’ altruistic motivation and social identification, thereby sustaining strong continued usage intention.  Griffioen et al. (2019) energy-saving experiment demonstrated that high-level-construal interventions were most effective for participants with greater social distance and were unaffected by temporal manipulations. This occurs because at high-level construals, consumers attend more to the essence of behavior rather than temporal details, thereby attenuating differences in reward timing. Therefore, under altruistic orientation, immaterial rewards—because their construal fits that of altruistic rewards—foster more positive continuance intention. In contrast, when material rewards are employed under altruistic orientation, the construal misfit diminishes DLCA continuance intention. In this case, material rewards convey profit-seeking signals that undermine consumers’ perception of the purity of altruistic behavior. Material rewards may also trigger motivation crowding-out effects, whereby consumers attribute their behavior to external incentives rather than intrinsic low-carbon beliefs, consequently reducing continuance intention.

In essence, our model adheres to “construal fit effect”: When the reward orientation is self-oriented, the optimal strategy involves employing immaterial-delayed and material-immediate reward combinations to satisfy instrumental needs and maximize immediate utility. When the reward orientation is altruistic, the optimal strategy entails adopting immaterial rewards combinations to preserve and reinforce intrinsic motivation and moral identity.

 

Once again, we would like to thank you for the above mentioned highly constructive suggestions regarding our manuscript. Besides the suggestions you highlighted above, we have also conducted a comprehensive review and streamlined the manuscript. The introduction, theoretical background and hypothesis development part of the original paper were written up to line 391. After revision, these sections now have a total length of 298 lines.

Reviewer 4 Report

Comments and Suggestions for Authors

This is an interesting study on DLCAs. The authors have great ideas and suggestions for addressing global environmental challenges. Some revisions/clarifications are needed as follows.

1. The length of each part of the paper is not appropriate. It feels like the introduction of the authors' work starts from line 379. The content before this can be appropriately compressed or streamlined.
2. In the part of Introduction, before DLCAs, were there similar applications? What are the characteristics? Will it be replaced by DLCAs?
3. The professional background and hobbies of the participants in the experiment will also have an impact on the results. Have the authors considered these factors?
4. In Figure 1, are these factors mutually influential? Will it also affect the higher-level factors?
5. People of different ages and professional backgrounds have different preferences. Have the authors considered designing different types of rewards for different groups of people?
6. Part 6 should distinguish between discussion and conclusion. The conclusion can include the results obtained in this paper and prospects for the future.

Author Response

We greatly appreciate your constructive suggestions. Below, we provide a detailed account of our revisions.

 

Comments 1: The length of each part of the paper is not appropriate. It feels like the introduction of the authors' work starts from line 379. The content before this can be appropriately compressed or streamlined.

Response 1: We sincerely appreciate your precise and constructive feedback. Your observation regarding the disproportionate length allocation across sections and the delayed introduction of our core research constitutes a critical optimization point for this manuscript. We fully concur with your assessment and have undertaken a comprehensive review and streamlining of the paper in response to this concern. The introduction, theoretical basis and hypothesis deduction part of the original paper were written up to line 391. After revision, these sections now have a total length of 298 lines.

 

Comments 2: In the part of Introduction, before DLCAs, were there similar applications? What are the characteristics? Will it be replaced by DLCAs?

Response 2: We greatly appreciate this valuable suggestion. In accordance with your recommendation, we have supplemented the Introduction with a systematic review of analogous low-carbon applications and research that preceded the emergence of DLCAs. We have clarified their distinctive characteristics and elucidated that DLCAs do not constitute a replacement for traditional counterparts, but rather represent an iterative upgrade and paradigm reconstruction enabled by technological advancement. The specific elaboration is provided below:

Analogous Low-Carbon Applications Prior to DLCAs and Their Core Characteristics. Before DLCAs achieved scalable and intelligent low-carbon services through digital technology, academia and industry had developed numerous consumer-oriented low-carbon applications and service models. These initiatives fundamentally focused on guiding consumer low-carbon behavior and reducing carbon emissions in production and daily life. However, constrained by technological limitations of the era, they failed to form systematic and intelligent service ecosystems. Such applications predominantly relied on offline promotional campaigns, paper-based manuals, and rudimentary online check-in mini-programs, lacking the support of big data, Internet of Things (IoT), and blockchain technologies, and were consequently unable to achieve precise resource matching and intelligent energy consumption regulation. Early examples include low-carbon transportation awareness campaigns, community-based offline waste sorting guidance services, and basic household energy consumption tracking spreadsheets—all of which could only disseminate low-carbon concepts and record fundamental behaviors, falling short of enabling precise and efficient interventions in consumer behavior. In essence, low-carbon applications predating DLCAs represented passive, single-scenario service models without deep digital technology empowerment, with their core function limited to fostering low-carbon awareness rather than restructuring low-carbon production and lifestyle patterns.

The Relationship Between DLCAs and Traditional Low-Carbon Applications. DLCAs do not supplant traditional low-carbon applications; rather, against the backdrop of digital technology advancement, they constitute a paradigm reconstruction, scenario expansion, and efficiency enhancement of traditional approaches. The relationship between the two is one of inheritance and evolution, manifested in two dimensions: (1) Both share the core objective of reducing carbon emissions in consumers' production and daily activities. DLCAs perpetuate the fundamental mission of traditional low-carbon applications—"guiding public low-carbon behavior"—representing a continued exploration of implementing low-carbon principles at the consumer end. (2) Evolution in implementation pathways and effectiveness: Through three digitally-enabled mechanisms—resource sharing, intelligent control, and low-carbon behavior visualization—DLCAs address the core pain points of traditional applications: "weak technology, fragmented scenarios, and poor effectiveness." This transformation advances low-carbon services from "passive guidance" to "proactive intervention," from "single scenarios" to "cross-domain integration," and from "ambiguous effects" to "visualized quantification." For instance, while traditional bike-sharing could only achieve low-carbon short-distance travel, Mobility-as-a-Service (MaaS) within DLCAs integrates bike-sharing, public transit, car-sharing, and carpooling into a unified service through real-time big data analytics, achieving carbon emission reduction across the entire travel chain.

We have incorporated the aforementioned content into the research background section preceding the conceptual definition of DLCAs in the Introduction. By systematically reviewing the research and practice of analogous low-carbon applications prior to DLCAs, we have refined the logical coherence of the Introduction while further clarifying the innovative value and research significance of this study.

 

Comments 3: .In Figure 1, are these factors mutually influential? Will it also affect the higher-level factors?

Response 3: We sincerely appreciate this critical question. Our model does not assume simplistic pairwise random influences among the three factors; rather, grounded in “construal fit effect”, it specifically focuses on how “reward orientation” determines the optimal matching pattern between “reward timing” and “reward type.” Hypotheses H5a and H5b explicitly demonstrate that under self-oriented rewards versus altruistic rewards conditions, the interaction effects between “reward timing” and “reward type” are distinctly different, indeed opposite.

We recognize that your question may also point toward other potential relationships, such as “Does reward orientation directly influence reward type preference?” or “Does reward timing inversely affect perceptions of reward orientation?” In the current model, we have intentionally refrained from specifying these pathways as direct effects or bidirectional relationships, primarily based on two considerations: (1) The core objective of this research is to investigate the influence mechanisms of external intervention strategies (reward strategy design) on user intention. We have operationalized "reward orientation" as an exogenously given context established through corporate communication, rather than an endogenous mediating variable susceptible to modification by other reward attributes. This approach maintains model clarity and testability. (2) Simultaneously testing all bidirectional or mediating relationships among factors would render the model excessively complex, obscuring the examination of core hypotheses. Therefore, our current model specification achieves an appropriate balance between theoretical depth and empirical feasibility.

We fully concur that in real-world contexts, the interrelationships among these factors may be more dynamic and complex than our model depicts. Your question has illuminated highly valuable directions for future research. Subsequent studies could employ longitudinal designs to investigate whether reward timing retrospectively shapes user perceptions of reward orientation over time, or explore whether mediating relationships exist among reward type, timing, and orientation pairs, thereby constructing a more comprehensive conceptual model.

 

Comments 4: The professional background and hobbies of the participants in the experiment will also have an impact on the results. Have the authors considered these factors?

Response 4: We appreciate this valuable observation. The professional backgrounds and hobbies of participants that you have identified may indeed systematically influence their attitudes toward DLCAs and continuance intention, which is critical for enhancing the internal validity and generalizability of our findings.

Drawing upon extant research, we contend that participants' professional backgrounds and hobbies are unlikely to substantively affect continuance behavior. Vrain et al. (2022), employing repeated sampling methodology, investigated the dynamic changes in UK consumers’ adoption behavior across 16 DLCAs. Their findings indicate that individuals who discontinued DLCAs usage exhibited the prototypical characteristics of “early adopters” and “innovators”—specifically, young, employed, high-income, change-embracing, innovative, digitally proficient, and online-active. Consequently, participants’ individual characteristics do not determine whether they continue using DLCAs. Furthermore, the study advances an important conclusion: although DLCAs may initially attract consumers through word-of-mouth and novelty appeal, users may gradually disengage due to “Disenchantment of attributes”—a phenomenon wherein features fail to meet expectations or lose their appeal over time. This underscores the necessity of investigating effective incentive mechanisms to promote sustained DLCA usage, which directly motivates the core focus of this paper—examining the influence mechanisms of external intervention strategies (reward strategy design) on user continuance intention.

In our experimental design, while we did not directly measure “professional background” and “hobbies,” we partially captured or controlled for related potential influences through the following variables: (1) Environmental attitudes: We employed validated scales to measure participants' overall environmental concern and sense of responsibility. This variable substantially encompasses the deeper concern for low-carbon issues that may derive from hobbies or professional backgrounds. (2) Prior evaluations and usage experience with DLCAs: We measured participants' familiarity with such applications, past usage experience, and overall evaluation. These variables effectively reflect pre-existing cognitions and habits formed through interest- or profession-related exposure, serving as more direct behavioral antecedents than “hobbies” alone. (3) Among demographic variables, we controlled for educational attainment and age. Educational level correlates to some extent with the breadth and depth of professional knowledge, while age is associated with digital product acceptance and interest patterns. We maintain that the control variables measured in our study contribute to mitigating confounds arising from individual differences (including those partially attributable to professional background and interests) on core experimental treatment effects.

Your suggestion has illuminated valuable research directions. In the “Limitations and Future Research” section, we have explicitly proposed that future research may consider variables beyond reward strategies, such as consumers’ professional background, hobbies, and individual values, and their potential effects on the model.

 

Comments 5: People of different ages and professional backgrounds have different preferences. Have the authors considered designing different types of rewards for different groups of people?

Response 5: We sincerely appreciate this highly insightful and practically illuminating question. Your suggestion of “designing different types of rewards for different demographic groups (e.g., varying by age and professional background)” indeed represents a critical direction for achieving precision-targeted incentives and enhancing efficiency in digital marketing and user engagement. We strongly endorse this perspective.

First, the core objective of this research is to elucidate the universal psychological mechanisms through which the 3 key design dimensions—reward type, timing, and orientation—interact and influence user intention via attitudes. We endeavor to address a more fundamental scientific question: “Under different contexts (self-oriented vs. altruistic), which combinations of reward attributes (material vs. immaterial, immediate vs. delayed) are psychologically more effective, and why?” Accordingly, the current research model treats demographic variables (e.g., age, professional background) as control variables rather than moderators, with the purpose of isolating the interference of these individual differences to more cleanly test the existence and efficacy of the aforementioned core mechanisms.

Second, our research findings directly support and theoretically substantiate your suggestion. Our model validates attitude as a critical mediator. The “differential preferences” across demographic groups varying in age and professional background may initially manifest as differential attitudes toward rewards. Furthermore, your suggestion deepens the insight that reward strategies must also be matched with internal user characteristics (such as values and needs shaped by age and professional background). This essentially represents an application of the same “matching principle” at different levels. Our research framework provides the theoretical background for this—specifically, the need to identify reward attribute combinations that match the intrinsic "psychological context" of particular user segments.

To actively incorporate your valuable suggestion, in the “Limitations and Future Research” section, we have explicitly identified "exploring the role of consumers' professional background, hobbies, individual values, and other variables in the model" as an important future research direction.

 

Comments 6: Part 6 should distinguish between discussion and conclusion. The conclusion can include the results obtained in this paper and prospects for the future.

Response 6: We concur with your recommendation. Accordingly, the former “7. Discussion and Conclusions” section has been restructured with the addition of two distinct subsections: “7.1. Conclusions” and “7.2. Discussion,” thereby clearly delineating the presentation of results from their interpretive discussion. Furthermore, we have substantially expanded “7.4 Limitations and Future Research” at the conclusion of the manuscript.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

satisfied

Author Response

We would like to express our gratitude to the expert reviewers for their valuable time and professional suggestions during this process. We are ready to cooperate with the subsequent publication procedures at any time.

Reviewer 4 Report

Comments and Suggestions for Authors

It is recommended to accept and publish this paper.

Author Response

We would like to express our gratitude to the expert reviewers for their valuable time and professional suggestions during this process. We are ready to cooperate with the subsequent publication procedures at any time.