Theoretical background
The increased use of service robots has caused remarkable changes in service delivery across various contexts. Service robots are “system-based autonomous and adaptable interfaces that interact, communicate and deliver service to an organization’s customers” (Wirtz et al. 2018, p. 4). Unlike self-service technology, service robots are empowered with artificial intelligence to deliver socially and emotionally interactive services. In human-robot interactions, consumers can feel they are interacting with another social agent (Doorn et al., 2017). Thus, robots’ adoption can not only be for their competence but also their emotional and social presence (Wirtz et al. 2018). The global market for robots is growing (Xiao and Kumar 2021) and robots are expected to be used massively in delivering frontline hospitality services by 2025 (Huang and Rust, 2018). However, research on service robots is mainly conceptual and focuses on the drivers of their adoption (Filieri et al., 2022). Empirical research on how consumers respond to unintended human-robot interaction failure and how firms can mitigate these unintended consequences is scant (Doorn, 2017; Honig and Oron-Gilad 2018). Thus, understanding how consumers respond to different attributes of human-robot interaction failure is needed. Human-robot interaction failure can influence consumers’ intentions to reuse the service and perceptions of service robots (Gompei and Umemuro, 2015). Thus, mitigating these unintended consequences can have substantial managerial and academic contributions. Human-robot interactions failure has rarely been studied in services marketing and is considered a future developmental plan for services research.
Research questions
RQ1: How would consumers respond to unintended service failure caused by a robot vs caused by a human misuse of a robot?
RQ2: Do consumers’ responses to an unintended human-robot interaction failure differ based on service failure type, service failure stability, robot design and the social presence of others?
RQ3: What is the most effective recovery strategy in handling service failures caused by a robot vs caused by human misuse of a robot?
Expected contributions
In line with the premises of attribution theory, we argue that failure causal attributions are related to consumers’ cognitive and emotional reactions (Vaerenbergh et al., 2014). Consumers’ reactions to failure caused by robots will be stronger than those caused by human misuse of robots due to the external vs internal locus of causality (Weiner 1985). We expect to make several theoretical and managerial contributions to the literature on human-robot interactions in the service failure and recovery context. Specifically, this study responds to services scholars’ calls (Pitardi et al., 2022) for additional research on the unintended consequences of service delivered by smart technologies such as service robots to improve consumers’ overall service experience. Meaningful technological engagement that can develop lasting relationships with humans has substantial implications for customers’ experiences (Doorn, 2017).
Methodology
Figure 1 provides an organizing framework that captures our conceptual and empirical work. Study 1 examines how failure caused by a robot versus a human misuse of a robot affects consumers’ cognitive, emotional and behavioral reactions following a technical versus interactional failure (testing Hypotheses 1). Study 2 examines how failure caused by a robot versus a human misuse of a robot) affects consumers’ reactions following a permanent versus temporary failure (testing Hypothesis 2). Study 3A examines the moderating effect of interacting with a very human-like vs a mechanical robot (testing hypothesis 3a). Study 3B examines the moderating effect of interacting with thinking vs relational robot (testing hypothesis 3b). Study 3C tests the moderating effect of the social presence of others (testing hypothesis 3c). Study 4A examines the moderating effect of expressing regret by a robot vs a human (testing hypothesis 4a). Study 4B examines the moderating effect of offering options to the consumers by a robot vs a human (testing hypothesis 4b), and study 4C tests the moderating effect of offering a collaborative solution with the help of a human (testing hypothesis 4c).
Studies 1, 2, 3 and 4
Participants and procedures
Study 1 design is a 2 (service failure source: robot vs. human misuse of a robot) x 2 (failure type: technical vs. interactional). Study 2 design is a 2 (service failure source: robot vs. human misuse of a robot) x 2 (failure stability: permanent vs temporary) between-subjects design. Study 3A design is a 2 (service failure source: robot vs. human misuse of a robot) x 2 (robot appearance: very human-like vs. mechanical) between-subjects design. Study 3B design is a 2 (service failure source: robot vs. human misuse of a robot) x 2 (robot type: relational vs. thinking) between-subjects design. Study 3C design is a 2 (service failure source: robot vs. human misuse of a robot) x 2 (social presence of others: present vs. absence) between-subjects design. Studies 4A, 4B and 4C designs are a 2 (recovery source: robot vs. human) x 3 (recovery type: expressing regret vs. offering options for the consumer vs. collaborative solutions with the help of a human). Participants will first read a scenario describing a service failure at a restaurant. Failure source, failure type, and failure stability will be manipulated by pictorial stimuli that will be presented along with written scenarios adopted from previous empirical studies (e.g., Choi et al. 2021). These scenarios will be verified and pretested in separate pretests before using them to manipulate our stimuli. A random sample of UK adult consumers will be recruited from Amazon MTurk’s consumer panel and randomly assigned to one of our experimental conditions. For example, participants in study 1 will be asked to imagine themselves in a restaurant scenario where either a technical failure (robot malfunction caused by a robot vs a human misuse of a robot) or social norm violation (inappropriate social behavior by a robot vs. a human) has occurred. Participants will be asked to indicate their perceived cognitions, emotions and behavioral outcomes. They will also be asked to respond to background information, including the frequency of dining out, failure controllability, failure severity and consumers’ technology readiness (used as potential control variables in a subsequent analysis).
References
Choi, S., Mattila, A. S., & Bolton, L. E. (2021). To err is human (-oid): how do consumers react to robot service failure and recovery?. Journal of Service Research, 24(3), 354-371.
Filieri, R., Lin, Z., Li, Y., Lu, X., & Yang, X. (2022). Customer Emotions in Service Robot Encounters: A Hybrid Machine-Human Intelligence Approach. Journal of Service Research, 10946705221103937.
Gompei, T., & Umemuro, H. (2015, August). A robot’s slip of the tongue: Effect of speech error on the familiarity of a humanoid robot. In 2015 24th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN) (pp. 331-336). IEEE.
Honig, S., & Oron-Gilad, T. (2018). Understanding and resolving failures in human-robot interaction: Literature review and model development. Frontiers in psychology, 9, 861.
Huang, M. H., & Rust, R. T. (2018). Artificial intelligence in service. Journal of Service Research, 21(2), 155-172.
Pitardi, V., Heirati, N., Jayawardhena, C., Kunz, W., Paluch, S., Wirtz, J., Unanticipated and Unintended Consequences of Service Robots in the Frontline. Journal of Business Research [ accessed on 25 July 2022] https://www.journals.elsevier.com/journal-of-business-research/call-for-papers/unanticipated-and-unintended-consequences-of-service-robots-in-the-frontline
Van Doorn, J., Mende, M., Noble, S. M., Hulland, J., Ostrom, A. L., Grewal, D., & Petersen, J. (2017). Domo arigato Mr. Roboto: Emergence of automated social presence in organizational frontlines and customers’ service experiences. Journal of service research, 20(1), 43-58.
Van Vaerenbergh, Y., Orsingher, C., Vermeir, I., & Larivière, B. (2014). A meta-analysis of relationships linking service failure attributions to customer outcomes. Journal of Service Research, 17(4), 381-398.
Wirtz, J., & Zeithaml, V. (2018). Cost-effective service excellence. Journal of the Academy of Marketing Science, 46(1), 59-80.
Xiao, L., & Kumar, V. (2021). Robotics for customer service: a useful complement or an ultimate substitute?. Journal of Service Research, 24(1), 9-29.
INTRODUCTION
For retail owners, it is crucial to deliver high-quality and innovative service to stay competitive. High-quality service is a valuable asset since it leads to customer satisfaction and loyalty, which contributes to the financial performance of the company . The so-called service encounter, concerning the direct interaction between customer and organization at the moment of service delivery, is central to high-quality service .
Humanoid service robots are one of the most dramatic innovations in the retail service encounter . Softbanks’ Pepper is a typical example of a possible retail innovation: a 120 centimeters humanoid with a friendly face, moving arms including hands and fingers, and a human-like locomotion. Even though humanoid service robots like Pepper could play an important role in a high-quality service encounter , empirical knowledge about the effects on customers and store employees is still lacking .
Previous studies show that the vast majority of research done on robots in retail has an conceptual nature, mainly focuses on the first adoption phase of consumers, and has therefore paid little or no attention to, for example, the role of store employees . The empirical research that has been done has mostly been conducted in controlled environments, which makes the practical usefulness for retailers limited and misses the complexity of a real-life setting . Therefore, a field experiment in an actual store is crucial.
RESEARCH QUESTION
What happens if you measure the added value of a service robot in an uncontrolled store setting?
METHODOLOGY
We conducted a field experiment where we placed Pepper in an actual grocery store during a 3-week period. Pepper was located at the entrance of the store, welcoming customers and suggesting information about the app, store discounts or meal inspiration. We asked store visitors about their robot experience: if they noted the robot, how they perceived the interaction (if applicable), and their idea of how Pepper could add value to the store.
To complement the data, we asked store employees about their experience with working next to Pepper. We used a structured interview to obtain this data.
FINDINGS
– Customers
Out of 139 customers, 46 customers (33.1%) did not notice the robot, 43 customers (30.9%) noticed the robot but did not stop at the robot, the other 50 customers (36%) noticed the robot and stopped. It is important to mention that this percentages only contain the customers we interviewed. To obtain a more accurate stopping rate, we counted 8 times for 15 minutes the customers entering the store, and how many of them interacted with Pepper.
The average stopping rate during these days was 14,7%.
The customers interviewed gave us interesting insights. 35 (25.2%) of them mentioned that Pepper could be valuable for entertainment, 22 (15.8%) customers thought Pepper can mean something for children and another 22 (15.8%) did not see any perspective for a service robot in the store. Lastly, 12 customers (8.6%) had no clue about what Pepper could mean for them or the store.
In addition, 15 (10.8%) people mentioned not to be interested in the robot because they were in a hurry.
– Employees
The three employees we interviewed all seemed quite skeptical about Pepper in the store. They did not experience a real decrease in workload and could not really see great advantages yet. Again, the effect on children was mentioned. Another employee pointed out that the robot was a good solution for welcoming customers. A third employee was slightly annoyed that they had to make sure Pepper was charged, which costed them extra work.
CONCLUSIONS
Empirical research into service robots and human-robot interaction perceptions in a real-life setting is still underdeveloped. In this study, we examined Pepper in an actual grocery store for a longer period, so we could identify unforeseen effects.
Firstly, we found that customers vary in their thoughts about the added value of Pepper. Recurring topics are entertainment and/or children, and no added value or idea. Since children seem to play a major role, it is important to take this into consideration: it might not be the actual target group of the retailer’s investment. Prior research focused on the effect of humanoids on children with autism , hearing disabilities , but the (unintended) effects for retail stores is undefined.
In line with the above, it is important for both customers and employees to know what Pepper does for the store. Store owners should have a clear goal what they would like to obtain before they invest in a service robot. The importance of instructing employees is highlighted.
Relevance of research
Humanoid service robots (HSR) are on the rise. Designated by scholars and practitioners as the workforce of the future, their introduction into many service contexts and industries is fueled by advancing technological developments in artificial intelligence (AI) and automation (Rust & Huang, 2014). HSR are becoming an integral part of frontline service operations (FSO) to fulfill socially assistive positions across several service sectors such as healthcare (Holland et al., 2021), hospitality (Tuomi et al., 2021), and retail (Amelia et al., 2022). Here, they support or even replace frontline service employees. The increasing relevance of the topic in practice is also reflected in the remarkable increase in research in the field of service robots in recent years (Lu et al., 2020).
Nevertheless, prior studies on the interaction with and acceptance of humanoid service robots were conducted primarily with a focus on a HSR vs. human employee perspective by applying comparative (experimental/quantitative) methods. Thus, the current discussion of HSR in frontline service encounters lacks consideration of the fact that “human-robot interaction analysis requires a different approach that recognizes that humans’ relationship with technologies is multifaceted and context-dependent” (De Keyser & Kunz, 2022, p.68). This goes along with the identified lack of scholarly research on the ethical and moral considerations of de-humanizing service encounters by introducing HSR. These considerations range from privacy concerns and biased consumer outcomes to concerns related to loss of customer autonomy, social isolation and more (Mariani et al., 2022).
Research objectives and research question
Against the background of the blind spots regarding the impact of human-robot service interaction listed above, this research seeks to answer the following research question:
How can humanoid service robots in frontline service encounters put the customer experience and well-being in jeopardy?
To narrow this research question down and make it more approachable and assessable during our empirical design, we further established two guiding questions for our two empirical studies (see “methodology” below) to facilitate the course of our two-study research. The guiding questions helped us collect the relevant insights and knowledge to successively answer the research question above.
These guiding questions were:
(1) What are the ethical concerns and negative outcomes experienced by customers interaction with HSR?
(2) How should service providers manage HSR in frontline service operations to account for customers’ concerns and prevent threats to their well-being?
Research methodology
Our methodological approach acknowledges the lack of conceptual underpinnings regarding the contextual and relational character of human-robot service interaction (HRSI) and its impact on customers’ service experience and well-being. We thus used an exploratory research design with two complimentary studies to (1) understand what constitutes the service experience for customers interacting with HSR and what factors negatively influence the service experience and (2) determine what service providers using HSR in FSO can do to actively address ethical concerns and threats to customer well-being.
The first study was designed as a problem-centered interview study with service customers (n=41, age range 19 to 71 years old) who had had either prior experience with HSR in a service setting themselves or were familiar with the technology through prior research or experience in other contexts. Thus, a purposive sampling approach was applied to account for the novelty and distinctness of the topic.
The second study was conducted as an exploratory study using expert interviews (n=27) with company representatives working for companies that either actively use HSR in their FSO already or manufacture HSR for service application. Service providers from various fields were integrated, among others hospitality, gastronomy, human care, and transportation.
All interviews were transcribed verbatim and checked for correctness and accuracy and then exported to atlas.ti 22, a qualitative data analysis software. We followed a systematic stepwise recursive process in the thematic analysis of the data (Boyatzis, 1998). Transcripts were coded independently by both members of the research team. A code system was established and built inductively, based on the in-depth textual analysis. New codes were created in an iterative fashion to capture the meaning of initial code groups (Thomas & Harden, 2008). Co-occurrence matrices in atlas.ti were applied to hierarchically organize individual codes in the shape of a coding tree. In an iterative process, the data material was merged, and the two members of the research team independently formed the main categories, discussed the content and labeling and, after several rounds, agreed on a final set of themes.
Preliminary findings
The preliminary findings (analysis ongoing) decode the major concerns faced by customers interacting with HSR across various service encounters. They reveal customer concerns addressing the personal (e.g., privacy, rapport), the social (e.g., fear of substituting human labor), or the interactive (e.g., service quality) level of the HRSI. As customers interact with a HSR in a specific service encounter, they form expectations about the performance level of the service. Their evaluation of the service experience is based on the perceived performance level in terms of contextual (type of service, i.e., information-processing vs. people-processing), transactional (i.e., task complexity, convenience), and relational (i.e., empathy, emotionality) performance.
Originality of the paper
This research is among the first to openly address critical components of customers’ service experience with HSRs to assess ways in which they can put the customer experience and well-being in jeopardy. It presents negative consequences of unfavorable human-robot service interactions to pinpoint current boundaries of HSR-implementation in service settings.
Scholarly (re)search for determinants and interdependencies of emotionally and psychologically stimulating service experiences with HSR is still in its infancy. This research thus motivates scholars to strive for a better understanding of the ways in which HRSIs can cause negative impacts on customer well-being to inform technological development and contextual implementation in service settings. It reveals interdependencies of personal, technological, and contextual determinants to lead to concerns and threats to customer well-being, thereby creating awareness and leading service providers and managers towards a more customer-oriented design and application of HSR.
Currently, businesses use a variety of artificial intelligence (AI) applications, such as service robots (Wirtz et al., 2018). Aside from the innumerable benefits, their quick and broad deployment has also led to a number of problematic issues (Honig & Oron-Gilad, 2018). For example, several studies focused on how people reacted to failing algorithms (Srinivasan & Sarial-Abi, 2021). Even fewer studies investigated how people react when robots fail (i.e. Choi et al., 2021). Prominent marketing strategies involved depicting resilient and well-engineered robots in states of falling, failing, beating up, and aiming at evoking various feelings (i.e. empathy, warmth, or comfort). Despite it being a significant phenomenon, almost no previous research investigated how consumers react to robots depicted as falling, failing, beaten up, or lost (see Table 1 for a list of popular robot failures and falls).
The most prominently portrayed type of robot failure in popular media is “the fall.” In this research, our goal is to investigate what people think and feel about the phenomenon of “failing robots” in the context of a “fall, as consumers evaluate the same technology (i.e., robots) in somewhat diverse ways (Siino & Hinds, 2005; Gretzel & Murphy, 2019). We present the initial findings of our content analysis to pinpoint the specific concepts consumers focused on when formulating their thoughts and feelings on falling robots.
We started out our in-depth exploratory research by gathering open-ended consumer verbatims from a convenience sample of university students. They responded to a specific robot fall news item, accompanied by a visual, which aided in the discovery of new and relevant issues centered on the “failing robots” theme. Visuals are frequently used in exploratory studies to aid in the probing of meanings and reactions (Christodoulides et al. 2021).
Participants
A convenience sample of eighty-eight (42 female) undergraduates studying business at a major European university took part in the current study in partial fulfillment of their course requirements. We have not prioritized generalizability and scale, which are not key aspects in qualitative sampling (Holloway & Jefferson, 2000). The average age of the participants turned out to be 23.09 (SD = 2.504, ranging between 19-34) and on average they reported medium income.
Procedure
Participants were first shown a piece of news about a robot falling (Appendix 1). The photograph was captioned, “The fall of one of a robotics company’s robots at a trade show.” They were asked to answer a few questions about their evaluation of the robot and the news depiction, as well as some control variables like their involvement in robotics, anthropomorphism level for the robot, and demographics.
Findings
As two researchers, we concurrently open-coded the short-essay reactions to the robot and the news. We used manual coding to gain insight into the characteristics and dimensions of attitudes toward malfunctioning robots. The initial results of consumer verbatim analysis revealed several broad themes of forming an opinion or attitude toward the issue. The following are some of the most notable examples in the three thematic categories: (1) seeing robots as the futuristic technological advancement of humans and getting upset with them falling; (2) seeing robots as yet another simple machine and not minding much about them falling; and (3) in-between: having mixed feelings toward the robots’ falling
Seeing robots as a futuristic technological advancement for humans:
“The sad part is that this shows that we are behind in terms of technology”
“They can develop themselves and improve their technology. Every success comes from after failure.”
“It wasn’t just a falling robot; the ideas and experiences fell, too.”
“I am not interested in how and why it fell off the stage at all, but I am curious about what is planned for the future of this robot in terms of AI improvements.”
Seeing robots as yet another simple machine:
“I feel nothing about the robot’s fall”
“I have no emotion about the robot’s fall”
“Since it’s a machine, a fault could appear at any time, which, on the other hand, shows that we cannot rely on robots 100% and that human interaction is gonna be always needed.”
“Since that was a robot, I did not feel anything.”
Having mixed feelings:
“This news is partly fun and partly sad”
“They’re exciting, but a little scary”
The first group got depressed with the robot’s fall and took it as a sign that the technological advancement and efforts set forth for such were wasted, or at best, they would like to understand what the key learning was to make sure that desired progress can be achieved in future technologies. The second group did not take the robot’s fall as something to be of importance as it was yet another machine that surrounds modern daily life; however, they still felt a bit disappointed that humans were needed to compensate for the robot’s falls. Lastly, the third group had mixed feelings and reported confusion when they saw a robot fall.
Conclusion
Consumers may experience unexpected and mixed emotions after witnessing robotic failures, and these emotions may then influence their attitudes and related behavioral intentions. For example, previous research demonstrated that using service robots attenuated consumers’ embarrassment (Pitardi et al. 2021). It is yet unknown which emotions and mechanisms are in place for evaluating failing robots in a favorable or unfavorable light by the consumers. The apparent polarity reflected in the verbatims points out that using robot fall strategies is a double-edged sword for robotic service providers, producers, and even retailers, and the involved mechanisms need to be analyzed in-depth in order to avoid unanticipated and unintended consequences.
The deployment of AI chatbots is driving significant changes in the way that service is delivered and experienced (Belanche, Casaló and Flavián, 2021; Ostrom et al., 2021). The co-creation literature emphasises actor-to-actor networks that create value for each other (Vargo and Lusch, 2004); however, the introduction of AI chatbots in the frontline adds an uncharted dimension to the study of co-creation, as AI developments promise value co-creation that changes as the AI adapts to other actors (such as customers), and these actors then adapt to the AI. AI chatbots take an active part in the service encounter, and as pseudo frontline employees, become important actors in the value co-creation process (Verhagen et al., 2014).
However, chatbots often fall short of customer expectations, resulting in service failure (Sheehan, Jin and Gottlieb, 2020). In this manner, the failure can also be perceived as co-created, since customers invest considerable time and energy (resources) in order to interact with the chatbot and co-create the interaction (Heidenreich et al., 2015). The negative effects of such service failures may be further compounded if customers are not allowed to choose their preferred customer representative: a human agent or a chatbot; or if customers perceive to be interacting with a human, but would be interacting with a chatbot instead (Robinson et al., 2020). Although previous literature extensively examined the determinants of consumer use and adoption of novel technologies, extant research has largely been conducted in voluntary co-creation contexts, and has neglected the investigation of less harmonious forms of co-creation, such as forced co-creation contexts.
Different co-creation contexts are also likely to be influential in forming customer expectations regarding the forthcoming co-creation experience, the chatbot’s performance, as well as possible chatbot limitations (Crolic et al., 2022). The nature of such expectations can have a significant impact in shaping customer evaluations about their participation in co-creation (Nijssen, Schepers and Belanche, 2016) and eventually in assigning attributions for failure (Belanche et al., 2020). However, the literature only offers contradictory findings regarding the possible impact of different co-creation contexts on customer expectations, and responsibility attributions.
Against this background, the aim of this study is to investigate not only the customer perception of chatbot interactions in different co-creation contexts, but also to understand how such contexts may influence customer expectations and the resulting responsibility attributions when chatbot failure occurs.
Guided by a Pragmatism research philosophy, a mixed methods approach was adopted in order to be able to effectively address the research aim. Specifically, an exploratory sequential design was employed, consisting of the collection and analysis of qualitative data in the first stage, followed by a quantitative phase which built on the findings obtained from the qualitative study.
The qualitative study comprised a total of 39 semi-structured interviews conducted in a face-to-face format. This study demonstrated how customers perceive three distinct co-creation contexts when interacting with chatbots. One context, volitional co-creation, is more aligned with the concept of voluntary participation and collaboration that is prevalent in the co-creation literature. Yet, the other two contexts, coercive and deceptive co-creation, are novel to the co-creation literature and as such, are worthy of further investigation. As a result, the qualitative study advanced a research framework and accompanying hypotheses regarding the impact of distinct co-creation contexts on expectations and attribution of responsibility.
Focusing on the investigation of coercive co-creation contexts, the hypotheses were tested through two experimental research studies conducted in a customer service setting. Data was collected from US participants (N = 315; N = 324), who, as part of the experimental research, were asked to interact with a chatbot that was specifically programmed for this study. Half of the participants were forced to interact, whereas the other half were given the illusion of choice between interacting with a chatbot and an alternative customer service channel. Structural Equation Modelling (SEM) was used to evaluate the experimental data.
The experimental studies found that in situations which result in service failure, customers who were forced to interact with chatbots, attribute more negative responsibility towards the company, than customers who were given a choice among several contact options. In such cases, customers blame the company directly for the failure. When compared to those customers who were offered a choice, customers who were forced to interact with the chatbot showed stronger controllability and stability attributions towards the company; in other words, they perceived the company as having had the ability to prevent the failure, whilst also perceiving the failure as more permanent and likely to happen again. These results were replicated within the two different service settings that were included in the experimental research. These results show that despite all the benefits associated with chatbots, it is important to understand why customers may feel forced into interacting with a chatbot, and the resulting customer evaluations following such interactions. Although the co-creation literature has shown a strong reliance on the notion of co-creation as a harmonious, voluntary collaborative activity, this study challenged this view and contributed to the co-creation literature by demonstrating the possibility of discordant forms of co-creation, which may especially arise in human-to-non-human interactions. Additionally, this study also contributes to the service failure literature by assessing the impact of service failures in an AI context and uncovering the underlying psychological processes following a co-created service failure between a customer and an AI chatbot.
The experimental research also found support for the mediating effect of disconfirmation of expectations. Forcing customers to interact with chatbots results in a more negative disconfirmation of expectations (the experience rated as worse than expected) than those customers who were given a choice. This, in turn, leads to stronger attributions of controllability, stability and causality towards the company. It is likely that forced involvement represents a more comprehensive level of mental involvement for customers (Reinders, Dabholkar and Frambach, 2008), in the process expecting more from the interaction, and ultimately being more disappointed when the service results in failure. This result confirms the role of customer expectations as a reference point when judging the actual service (Oliver, 2015) and contributes to the co-creation literature by establishing disconfirmation of expectations as an important link between the perceived freedom of choice when co-creating and the resulting attributions of responsibility.
The findings also suggest a number of strategic implications for managers who are considering the partial or complete replacement of human staff with chatbots in customer service settings.
The service sector is at an inflection point, with potential industrialisation arising from robotics creating new opportunities for innovation and the potential for a service revolution (Wirtz et al., 2018). A service revolution through technology may address issues including labour shortages (Beesley, 2021). In Ireland, for example, although tourism generates €9.2 billion annually and employs 265,000 people, there is a skills crisis with approximately 40,000 vacant positions (McGowan, 2021). Inflation in wages means businesses struggle to operate below 40% labour cost (McGowan, 2021), and 90% of businesses highlight difficulties in recruiting staff (Finn, 2021). In addition, workers in the gig economy face erratic working hours and uncertainty about pay (Martyn, 2021), with human costs of the gig economy including isolation without supportive colleagues or mentors, affecting perceptions of human dignity at work (Lillington, 2019).
Cognisant of these challenges, we recently conducted research based on a sample of 805 employees in the Irish hospitality sector (Wallace and Coughlan, 2022). Drawing on Conservation of Resources theory (COR), we proposed affective commitment and perceived leader support (specifically LMX) as resources against burnout. We found that the emotional exhaustion component of burnout was associated with counterproductive workplace behaviour, and affective commitment and LMX are effective resources against burnout, and against CWB. We also found that those who perceived they were on zero-hour contracts (or in the ‘gig economy’) were less able to draw on affective commitment or LMX as resources when experiencing burnout, but they were also less likely to ‘act out’ when they experienced burnout. We cautioned that zero-hour workers in the hospitality sector may internalise their stress and cope with the challenges of work in other ways.
One solution to these challenges would be a greater reliance on service robots; to support staffing, reduce uncertainty and work, and allow hospitality to deliver a consistent, efficient offering. Yet extant literature has focused less on the implications of frontline employees working with robotics (De Keyser et al., 2020). For example, if a service encounter is positive experience for the customer, the service robot may receive praise while the human employee remains unacknowledged, negatively impacting employee commitment (Robinson et al., 2020). Yet when technology such as service robots are involved in repetitive work, this may free up time to allow frontline employees to engage in more exciting and varied work (Wirtz et al., 2018). Also, if employees are no longer dealing with trivial requests, this may allow them to deal with higher-level tasks (Robinson et al., 2020).
Building on our earlier research, the first objective of our current study is to examine how service robots could be integrated in services by considering managers’ and employees’ views about these technologies. One could suggest that adding service robots would help to alleviate workload issues leading to burnout, support zero-hour workers in providing more assurance regarding hours worked, and support employees to allow them to engage with customers in a more genuine way. Yet while robots are predicted to have a profound impact on the sector (Lu et al., 2020), they have some weaknesses relative to service employees (Huang and Rust, 2018; Wirtz et al., 2018). For example, although robots offer an advantage of homogeneity in the delivery of repetitive services, customisation may be required to meet specific customer needs, and a heterogeneous delivery may be more appropriate (Palmer, 2011). Additionally, robots’ feigned emotion may be easily distinguished as not genuine, especially over longer- or high-involvement service encounters (Wirtz et al., 2018). Furthermore, customers may be relationship-motivated and expect social relationships with frontline employees, and this form of rapport, rich communication and emotional expression may engender customer trust and satisfaction, which could be lost in inhuman interactions (Robinson et al., 2020).
Rust (2020 p.18) highlighted that we are in a transformation, where artificial intelligence may compete with human intelligence, and this could dramatically change the skillset that humans need to remain relevant in the workplace. Huang et al. (2019) assert that human intelligence must emphasise empathy – a ‘feeling economy’ – where the empathetic dimensions of work are emphasised as mechanical and analytical tasks are replaced by AI and Robotics, especially in the services sector where interpersonal relationships are critical.
A consideration of the ‘feeling economy’ is particularly relevant in researching the implementation of service robots in Ireland’s hospitality sector. In Ireland, frontline service employees are a ‘secret ingredient’ in the sector. The Irish hospitality and tourism industry has built campaigns around ‘Ireland of the welcomes’ (McGrath, 2018). McGrath (2018) cites Niall Tracey of Failte Ireland, who explains “…what visitors constantly come back to is the people they engaged with,” he says. “When we ask visitors what it is about the people, it’s the smile. Irish people will smile at you, and without a word, that smile uniquely says, ‘I’m only here to help, how are you getting on?’”. He adds: “it’s very relaxed. It’s not manufactured friendliness, it’s a really authentic connection…it really makes Ireland quite magical compared to other destinations.” Yet when robots takes over a task, these human skills are displaced (Rust, 2020). How can, and should, the service be best delivered without losing that ‘magic’ that is unique and special within the sector?
The second objective of this study therefore is to explore how to best engage robots to alleviate challenges in the hospitality sector, while retaining the ‘magic’ of its service delivery and build an advantage in the ‘feeling economy’. Drawing on the Irish context provides unique insights into employee response to service robots, where frontline employees are the point of difference. Utilising survey, experimental design and in-depth discussions with managers and employees in hotels in particular, findings elicit their attitudes about service robots. We investigate the relationship between burnout, employee’s’ resources, and their response to service robots working alongside them. We also investigate managers’ views about integrating service robots in a sector famous for its friendly, human face.
Metabolism of pathogens in infectious diseases is important for their survival, virulence and pathogenesis. Mycobacterial pathogens successfully scavenge multiple host nutrient sources in the intracellular niche. It is therefore important to identify the intracellular nutrient sources and their metabolic fates in these pathogens. Metabolic phenotype of an organism is defined by metabolic fluxes. We quantified in vivo fluxes of the pathogens and probed host-bacterial metabolic cross talks in tuberculosis (TB) and leprosy using systems-based strategies and techniques of isotopic labelling, metabolic modelling and metabolic flux analysis (MFA). We show that the TB pathogen metabolizes a number of carbon and nitrogen sources in human macrophages and identified vulnerable nodes such as glutamine and serine biosynthesis as potential drug targets. Mycobacterium leprae, the leprosy causing pathogen, uses host cell glucose in infected schwann cells and the enzyme, phoenolpyruvate carboxylase is a potential drug target. Our research provides an understanding of the intracellular diets and metabolism of these important human pathogens and identified vulnerable metabolic nodes that can be used for developing innovative chemotherapies in TB and leprosy.
Mycobacterium ulcerans is the causative agent of the chronic skin infection Buruli ulcer. In contrast to most mycobacteria, M. ulcerans is predominantly found in the extracellular milieu in patient lesions. This is attributed to the production of a polyketide lactone, mycolactone, which inhibits innate immune responses and is cytotoxic to macrophages. Nevertheless, early in infection, intracellular bacteria are readily detectable and genetic evidence suggests that macrophages may play a role in controlling disease progression. In particular, polymorphisms in components of the autophagy pathway are reported to influence the severity of Buruli ulcer. However, little is known about the interaction between M. ulcerans and macrophages. We are currently investigating the cellular response to M. ulcerans by the autophagy pathway in macrophages. Early findings indicate that autophagy markers are upregulated in infected THP-1 macrophage-like cells, but these rarely colocalise with the bacteria.
The assessment of dietary advice delivered by a personalised mobile application to improve glucose control for adults with type 2 diabetes
Introduction
Finding an optimal diet that suits millions of people has proven to be challenging for nutrition research. Advances in technology, specifically AI, have enabled us to gather and process vast amounts of information as a tool to create a higher level of personalization and subsequently to boost diet adherence. While the number of nutrition apps is infinite, few provide adaptive evidence-based nutrition advice for patients with type 2 diabetes (T2D) residing in Germany. Therefore, as part of the PROTEIN project we aim to trial a dynamic personalized nutrition application to improve dietary compliance in T2D.
Objectives
We aim to customize meal plans according to data from the PROTEIN health expert team, the subjects’ preferences and their continuous glucose monitor (CGM) data. Our primary objective is to increase the time in range (TIR) of the study participants by 5%.
Methods
PROTEIN, a RCT, has a 3-month intervention period. Here, the subjects will use our application, a CGM and an activity tracker to collect essential data to provide personalization. The AI advisor is responsible for providing the meal plans and consists of a food and activity recommender system (FARS) and a reasoning-based decision support system (RDSS). The researchers on site will upload recent CGM data (7-14 days) that will be used to provide customized plans according to postprandial glucose excursions during the intervention period. If the TIR is below 70%, the goal „Decrease carbohydrates“ will be activated for next week’s recommendations. Foods, apart from vegetables and pulses, that caused high glucose levels (≥140 mg/dL over 2-4 hours) will activate a push notification directly to the user. If a food causes high glucose levels on two or more separate occasions, it will be excluded from the meal plans the following week.
Results
The app is being tested in virtual users to assure the appropriateness of the meal plans created by the AI advisor for each user. 300 subjects are planned to be tested.
Conclusions
The PROTEIN system creates adequate meal plans for virtual users with high glucose levels suggesting that an improvement of TIR can be achieved in real patients.
Funding
European Union’s Horizon 2020 research and innovation programme under grant agreement No 817732
The DarT/G toxin-antitoxin system encodes a pair of enzymes that mediate the addition of an ADP-ribose moiety onto thymidine in ssDNA in a reversible, sequence specific manner. Although originally characterised in Thermus aquaticus, the system is present in a number of important pathogens including all members of the mammal adapted M. tuberculosis complex, notably including human and bovine TB. Utilizing CRISPRi technology to silence DarG antitoxin expression, we have shown that DarT performs ADP-ribosylation of gDNA in cellulo in M. bovis BCG, leading to a rapid arrest of DNA replication and cell division, and that is ultimately toxic to the bacterium. In MTBC, darT and darG are transcriptionally linked to the dnaB gene, which encodes the replicative helicase that interacts with ssDNA at the chromosome origin (OriC) to initiate then drive DNA branch migration during replication. We demonstrate in vitro and in cellulo that MTBC DarT heavily ADP-ribosylates TTTW motifs in the AT-rich DnaB-loading region of OriC, suggesting that the DarTG system may work as a reversible regulator of replication. Furthermore, unregulated ADP-riboslyation by DarT induces the DNA damage SOS response, including the ImuA’ImuB/DnaE2 mutasome which has been implicated in DNA damage-induced mutagenesis and acquisition of resistance to antibiotics.
Immunopurification and NGS sequencing of ADP-riboslyated gDNA fragments has given further insight into the role of ADP-riboslyation in M. tuberculosis physiology, confirming ADP-ribosylation of OriC and demonstrating ADP-ribosylation at additional genomic loci, prominently including genes involved in the SOS response, DNA metabolism, and ribosomal proteins. This identifies the potential for ADP-ribosylation to act as a genome-wide epigenetic and cell signalling factor.
We aim to further understand the role of DarTG in bacterial physiology including DNA replication, the DNA damage response, persistence and drug-resistance in Mycobacterium tuberculosis.