ABSTRACT. The fast fashion industry has experienced a significant digital transformation, with social media platforms, influencers, live shopping, and scarcity-driven promotional strategies becoming central to consumer engagement and sales growth (Weber & Ritch, 2023). These digital tools are particularly effective at speeding up purchasing decisions and stimulating impulse buying, especially among younger consumer groups (Rajput & Gandhi, 2024). Whilst impulse purchasing has long been recognised as a defining characteristic of fast fashion consumption, its psychological and emotional impact within the digital environment remains poorly understood (Susanto et al., 2023). Existing research on impulse buying within the fashion and retail sectors has primarily focused on its antecedents. In recent years, fear of missing out (FOMO) has been identified as a key psychological driver of impulse buying within digital environments (Akram et al., 2018; Çelik et al., 2019). However, much of this literature treats impulse buying as an endpoint rather than the starting point of a broader emotional and behavioural process (Obukhovich et al., 2024; Redine et al., 2023). This limit understanding of how impulsive decisions shape subsequent emotions. Post-purchase emotional responses, such as regret, have long been recognised within consumer psychology (Chen et al., 2021). However, these theories were largely developed within traditional retail settings and may not fully capture the emotional dynamics of digitally accelerated consumption. Within fast-fashion e-commerce ecosystems, consumers are frequently exposed to algorithmically driven content, influencer persuasion, and time-limited promotions (Qi, 2024). These factors compress decision-making processes and intensify emotional fluctuations. This study addresses these gaps by examining post-purchase emotional responses and coping behaviours in digital fast fashion contexts, with a focus on generational differences.

This research adopts a dual theoretical framework grounded in cognitive dissonance theory and regret theory, which collectively form a powerful foundation for understanding consumers' post-purchase emotional responses. The theory of regret emphasises counterfactual thinking, whereby consumers compare actual outcomes with imagined alternatives, generating negative emotions upon realising they could have made a superior choice (Barta et al., 2022). Cognitive dissonance theory (Festinger, 1957) posits that individuals experience psychological discomfort when their actions conflict with their beliefs, values, or self-concept (Powers & Jack, 2013). In consumption contexts, this discomfort typically arises when consumers question the necessity, appropriateness, or value of their purchase. Digital fast fashion consumption is characterised by speedier decision-making, heightened emotional stimulation, and constant engagement with highly persuasive marketing signals (Parker-Strak et al., 2020). These features reduce careful consideration and increase emotionally driven impulse purchases (Cook & Yurchisin, 2017). This research introduces the concept of “digital regret” as an extension of existing theory within specific contexts. ‘Digital regret’ reflects the negative emotions consumers experience upon realising their purchasing decisions were not well considered but rather induced by digital channels. This research identifies digital marketing stimuli, including influencer marketing, live-stream shopping, and scarcity promotions, as key factors intensifying fear of missing out (FOMO). FOMO, subsequently, heightens consumers' propensity for impulse purchases within fast-fashion digital environments. Post-purchase, consumers experience varying degrees of digital regret, the severity of which depends on individual and situational factors. This research conceptualises coping strategies as behavioural and psychological mechanisms consumers employ to reduce post-purchase emotional discomfort. These strategies include returns, rationalising purchase decisions, and avoiding future exposure to similar marketing stimuli or brands. Rather than viewing returns as purely logistical or operational outcomes, this research positions them as emotionally motivated coping mechanisms aimed at reducing cognitive dissonance and regret (Powers & Jack, 2013). Generational differences between Millennials and Generation Z form a core analytical perspective within the conceptual model. Although both groups are digitally native, they differ in their social interactions within digital environments, emotional regulation strategies, and attitudes towards consumption and sustainability. By integrating emotional, behavioural, and generational insights, this conceptual model offers a comprehensive framework for understanding post-purchase processes within digital fast fashion consumption.

This research adopts a pragmatic research paradigm, recognising that the research question requires both interpretative insights into consumers' emotional experiences and empirical testing of theory-based relationships (Maarouf, 2019). Consequently, this research employs a mixed-methods design guided by abductive reasoning, permitting iteration between theoretical and empirical observations. During the qualitative research phase, in-depth semi-structured interviews are conducted with consumers who have recently engaged in impulse purchases on fast fashion e-commerce platforms. Findings from this phase inform the refinement of the conceptual framework and adapt existing measurement scales for application within the fast-fashion e-commerce platform context. The quantitative research phase incorporated validated scales to measure fear of missing out (FOMO), impulse buying tendencies, regret, and coping strategies, with adjustments made in line with qualitative findings. This mixed-methods design enables the study to capture the complex emotional processes underlying digital impulse purchases while delivering evidence-based insights, thereby providing a reference for theoretical development and managerial practice.

This research extends cognitive dissonance theory and regret theory to the digital fast fashion context, conceptualising digital dissonance and digital regret as manifestations of post-purchase emotional discomfort within specific situations. The research deepens our understanding of how established psychological mechanisms operate within accelerated and highly persuasive digital environments. Secondly, this research contributes to the impulse buying literature by shifting the analytical focus from the purchase decision itself to the post-purchase emotional processes and coping behaviours. Thirdly, this research examines Millennials and Generation Z as distinct analytical groups rather than treating age as a control variable, thereby contributing to generational studies. This research provides a theoretical deep dive into how digital socialisation and value orientations influence post-purchase processes. From a practical perspective, this research holds promise in providing insights for fashion brands and digital marketers, assisting them in enhancing marketing effectiveness whilst prioritising consumer wellbeing. By highlighting the emotional consequences of digitally driven impulse purchases, minimise post-purchase regret and dissatisfaction. This research contributes to the development of more empathetic return policies and post-purchase communication strategies by framing product returns as an emotionally driven coping mechanism rather than a purely logistical outcome. By clarifying the emotional mechanisms linking digital influence, impulse purchases, and post-purchase responses, this research establishes a foundation for strategies aiming to reduce overconsumption, return-related waste, and long-term consumer dissatisfaction.

ABSTRACT. Sustainable fashion intentions among Gen Z consumers are often fragile in digital retail contexts. This study examines how message framing and persuasive technology influence Sustainable Fashion Behaviour Intention, and how cognitive effort moderates these effects. A 2×2 experiment with Gen Z consumers in Nigeria uses simulation methods.

ABSTRACT. ABSTRACT This study investigates how AI chatbots’ message tone (warm vs. competent), recommendation type (attribute vs. benefit), and personalization jointly shape consumer continuance intention in fashion e-commerce. Across two experiments, findings show that competent tones are most effective when paired with benefit-focused recommendations, while personalization serves as a key boundary condition by influencing perceived task-solving competence and psychological reactance. Overall, chatbot tone operates as a strategic communication cue whose effectiveness depends on message framing and personalization design. INTRODUCTION AI chatbots are increasingly embedded in digital retail environments, providing conversational support and product recommendations that influence consumer decision-making and long-term engagement (Luo et al., 2019). In fashion e-commerce, where shopping involves both functional problem solving and identity expression, chatbot communication style may play a particularly important role in determining whether consumers continue using AI-based services (Ashfaq et al., 2020). Although prior research highlights anthropomorphism, warmth, and personalization as drivers of positive consumer responses, socially rich chatbot cues may also generate discomfort or resistance when perceived as intrusive or autonomy threatening (Araujo, 2018; White et al., 2008). These mixed findings suggest the need for clearer theoretical explanations regarding when particular chatbot message strategies are most effective. To address this gap, this study focuses on chatbot message tone as a core communication cue. Drawing on Construal Level Theory (CLT), we argue that warm versus competent tones influence psychological distance and shape how recommendations are processed. By introducing construal congruence as a guiding mechanism, this research clarifies why certain tone–recommendation combinations enhance continuance intention in fashion chatbot contexts (Lee et al., 2010). CONCEPTUAL BACKGROUND Chatbot Message Tone: Warmth vs. Competence The Stereotype Content Model suggests that social actors are evaluated along warmth and competence dimensions (Fiske et al., 2007). Warmth reflects friendliness and relational intent, whereas competence conveys expertise and capability (Cuddy et al., 2008). In AI-mediated interactions, consumers rely on such linguistic cues to infer whether a chatbot is socially supportive or professionally reliable (van Doorn et al., 2017). Warm tones may foster comfort, while competent tones may strengthen credibility and task-oriented effectiveness. Construal Level and Recommendation Type Construal Level Theory proposes that psychological distance shapes whether individuals process information concretely or abstractly (Trope & Liberman, 2010). In fashion e-commerce, attribute-focused recommendations emphasize concrete product features, whereas benefit-focused recommendations highlight abstract lifestyle outcomes. Persuasion should be strongest when chatbot tone aligns with the abstraction level of recommendation content, enhancing processing fluency through construal congruence (Lee et al., 2010). Personalization as a Boundary Condition Personalization increases perceived self-relevance and relational closeness, potentially amplifying the effectiveness of warm communication (Ashfaq et al., 2020). However, highly tailored messages may also heighten perceptions of intrusiveness, eliciting psychological reactance and reducing engagement (White et al., 2008). Thus, personalization may determine when warmth versus competence cues foster continuance intention. HYPOTHESES DEVELOPMENT Based on CLT, warm tones signal psychological proximity and may align with concrete, attribute-focused recommendations, whereas competent tones convey authority and may fit abstract, benefit-focused recommendations. Message effectiveness should increase when tone matches recommendation framing through construal congruence. Personalization is also expected to shape tone effectiveness. Warm tones may become especially persuasive under high personalization, while competent tones may remain robust when personalization is low. Finally, perceived task-solving competence should enhance continuance intention, whereas psychological reactance should reduce it. H1. Warm (vs. competent) tones enhance continuance intention more for attribute-focused (vs. benefit-focused) recommendations. H2. Warm (vs. competent) tones are more effective under high (vs. low) personalization. H3. Task-solving competence mediates the interaction of tone and personalization on continuance intention. H4. Psychological reactance mediates the interaction of tone and personalization on continuance intention. METHODS Survey data were collected from consumers aged 20–40 with experience in online fashion shopping and chatbot usage. Across two experiments, 582 valid responses were analyzed using SPSS 29. Study 1 (n = 225) employed a 2 × 2 between-subjects design manipulating chatbot tone (warm vs. competent) and recommendation type (attribute vs. benefit). Participants viewed a simulated fashion shopping scenario and reported continuance intention. Study 2 (n = 357) used a 2 × 2 design manipulating chatbot tone and personalization level (high vs. low) within a more interactive NLP-based chatbot conversation. Participants reported continuance intention, perceived task-solving competence, and psychological reactance. RESULTS AND DISCUSSION In Study 1, a significant interaction emerged between chatbot tone and recommendation type on continuance intention (F(1,225) = 8.261, p < .01). Competent tones produced higher continuance intention when paired with benefit-focused recommendations, whereas warm tones showed no significant differences across recommendation types. Study 2 demonstrated that personalization functions as an important boundary condition (F(1,357) = 5.283, p < .05). Warm tones were most effective under high personalization, while competent tones remained relatively stable across personalization contexts. Moderated mediation analyses indicated that perceived task-solving competence positively influenced continuance intention, and psychological reactance emerged as a negative mediator under high personalization. Overall, chatbot tone is not universally beneficial but operates as a strategic communication lever whose effectiveness depends on recommendation framing and personalization design. Fashion retailers should deploy competent tones for benefit-oriented guidance and apply warm tones carefully in highly personalized interactions to support long-term consumer engagement.

ABSTRACT. This PhD project investigates the paradoxical relationship between sustainability and luxury in the fashion industry, focusing on how sustainability is perceived, managed, and communicated across different market and organizational contexts.

Building on prior literature, the project starts from the observation that luxury and sustainability are often perceived by consumers as contradictory concepts. While luxury is associated with exclusivity, heritage, and symbolic value, sustainability is linked to ethics, moderation, and responsibility. This tension creates challenges for luxury brands, as explicit sustainability communication may dilute brand desirability and trigger consumer skepticism. As a result, many luxury brands adopt implicit or restrained communication strategies, a phenomenon increasingly described as green-concealing. The first part of the project examines how luxury brands navigate this paradox in practice, with particular attention to managerial decision-making regarding whether, when, and how sustainability should be communicated without undermining symbolic value.

The second part of the project shifts the focus to the consumer side by exploring how sustainable luxury is perceived in developed versus emerging markets, with a comparative focus on Italy and Turkey. Prior research suggests that consumer responses to sustainability vary across cultural and market contexts, and that luxury consumption in emerging markets is more strongly driven by social signaling and status display. This raises questions about whether sustainability attributes are interpreted as added value or as a potential threat to luxury’s symbolic meaning. Through in-depth qualitative consumer interviews, this study investigates how consumers in different market contexts interpret sustainability cues in luxury, and how cultural, social, and economic factors shape these perceptions.

The third part of the project examines the Digital Product Passport (DPP) as an emerging regulatory and technological tool introduced by the European Union to enhance product traceability and sustainability transparency. While DPPs are designed to communicate standardized sustainability information to consumers, little is known about how fashion brands are preparing for this new regulation and how data will be generated, coordinated, and shared across supply chain actors. This study approaches the DPP not only as a communication device but also as a coordination mechanism that requires alignment between brands, suppliers, and service providers. It investigates the organizational and relational implications of DPP implementation in luxury fashion, particularly in relation to tensions between transparency, heritage, and exclusivity.

Taken together, the project adopts a multi-level perspective on the sustainability–luxury paradox, combining brand strategies, consumer interpretations, and emerging regulatory tools. By integrating B2C and B2B perspectives, the research aims to contribute to a better understanding of how sustainability can be integrated into the luxury fashion industry without eroding its symbolic foundations, and how new forms of traceability and information disclosure may reshape both communication practices and inter-organizational relationships.

ABSTRACT. As consumers increasingly delegate creative autonomy to Agentic AI, the dynamics of human-AI symbiosis demand deeper investigation. This study compares the psychological mechanisms driving user value co-creation when interacting with highly autonomous AI versus human experts. Employing a mixed-methods design involving social media text analysis and scenario experiments, we reveal a dual-pathway mechanism: human agents facilitate co-creation via outward "self-expansion," whereas Agentic AI induces inward "inclusion of other in self" (IOS). Furthermore, we uncover how task criticality and service failure uniquely moderate these pathways, offering novel insights into service resilience in AI-driven co-creation.

ABSTRACT. 8 to 10 percent of global carbon emissions are attributable to the fashion industry, largely due to energy-intensive manufacturing processes and waste. Digital technologies such as artificial intelligence, virtual sampling, and predictive analytics offer improved efficiency, operational resilience, and ostensibly enhance sustainability. However, they also entail substantial and increasing carbon costs. Notably, the environmental impacts of these technologies are often neglected. Specifically, energy consumption associated with data storage, processing, and transmission, especially using emerging tools like artificial intelligence and cloud computing, can exacerbate carbon emissions. Consequently, digital decarbonisation, reducing the carbon footprint of digital technologies, has emerged as a critical focus.

The fashion industry’s rapid digitalisation, particularly within the production stage, presents a critical window of opportunity to embed digital sustainable practices from the ground up. Developing an understanding of digital decarbonisation within fashion can enable targeted solutions that resonate across the wider creative sector, thereby influencing environmental strategies in design, media, and entertainment. This provides a unique opportunity for the creative sector to build genuine climate resilience long-term, beyond just optimising operations.

This paper introduces the concept of the digital sustainability paradox: the simultaneous pursuit of environmental goals through technologies that expand carbon footprints. Drawing on fashion as a case study, the paper conducts a comprehensive literature review to identify the existing research gap in digital decarbonisation within the fashion industry and critiques existing policy frameworks. This includes evaluating the UK’s DBT supply chain resilience model, arguing that it inadequately accounts for the environmental impacts of digital infrastructure. With the upcoming introduction of EU Digital Product Passports, digitalisation is no longer optional but mandatory, intensifying energy demands across fashion systems.

The findings indicate a notable gap in the literature concerning digital decarbonisation in fashion. While considerable attention has been directed toward the potential of digital technologies for sustainability, a thorough examination of their environmental costs is necessary. This paper advocates for further research dedicated to decarbonising digital tools in the fashion sector, ensuring that innovations contribute positively to sustainability without escalating the industry’s carbon footprint.