Zoekresultaten

Enhancing Service Interactions with Conversational Agents

In de afgelopen jaren hebben technologische ontwikkelingen de aard van dienstverlening ingrijpend veranderd (Huang & Rust, 2018). Technologie wordt steeds vaker ingezet om menselijke servicemedewerkers te vervangen of te ondersteunen (Larivière et al., 2017; Wirtz et al., 2018). Dit stelt dienstverleners in staat om meer klanten te bedienen met minder werknemers, waardoor de operationele efficiëntie toeneemt (Beatson et al., 2007). Deze operationele efficiëntie leidt weer tot lagere kosten en een groter concurrentievermogen. Ook voor klanten kan de inzet van technologie voordelen hebben, zoals betere toegankelijkheid en consistentie, tijd- en kostenbesparing en (de perceptie van) meer controle over het serviceproces (Curran & Meuter, 2005). Mede vanwege deze beoogde voordelen is de inzet van technologie in service-interacties de afgelopen twee decennia exponentieel gegroeid. De inzet van zogenaamde conversational agents is een van de belangrijkste manieren waarop dienstverleners technologie kunnen inzetten om menselijke servicemedewerkers te ondersteunen of vervangen (Gartner, 2021). Conversational agents zijn geautomatiseerde gesprekspartners die menselijk communicatief gedrag nabootsen (Laranjo et al., 2018; Schuetzler et al., 2018). Er bestaan grofweg drie soorten conversational agents: chatbots, avatars, en robots. Chatbots zijn applicaties die geen virtuele of fysieke belichaming hebben en voornamelijk communiceren via gesproken of geschreven verbale communicatie (Araujo, 2018;Dale, 2016). Avatars hebben een virtuele belichaming, waardoor ze ook non-verbale signalen kunnen gebruiken om te communiceren, zoals glimlachen en knikken (Cassell, 2000). Robots, ten slotte, hebben een fysieke belichaming, waardoor ze ook fysiek contact kunnen hebben met gebruikers (Fink, 2012). Conversational agents onderscheiden zich door hun vermogen om menselijk gedrag te vertonen in service-interacties, maar op de vraag ‘hoe menselijk is wenselijk?’ bestaat nog geen eenduidig antwoord. Conversational agents als sociale actoren Om succesvol te zijn als dienstverlener, is kwalitatief hoogwaardige interactie tussen servicemedewerkers en klanten van cruciaal belang (Palmatier et al., 2006). Dit komt omdat klanten hun percepties van een servicemedewerker (bijv. vriendelijkheid, bekwaamheid) ontlenen aan diens uiterlijk en verbale en non verbale gedrag (Nickson et al., 2005; Specht et al., 2007; Sundaram & Webster, 2000). Deze klantpercepties beïnvloeden belangrijke aspecten van de relatie tussen klanten en dienstverleners, zoals vertrouwen en betrokkenheid, die op hun beurt intentie tot gebruik, mond-tot-mondreclame, loyaliteit en samenwerking beïnvloeden (Hennig-Thurau, 2004; Palmatier et al., 2006).Er is groeiend bewijs dat de uiterlijke kenmerken en communicatieve gedragingen (hierna: menselijke communicatieve gedragingen) die percepties van klanten positief beïnvloeden, ook effectief zijn wanneer ze worden toegepast door conversational agents (B.R. Duffy, 2003; Holtgraves et al., 2007). Het zogenaamde ‘Computers Als Sociale Actoren’ (CASA paradigma vertrekt vanuit de aanname dat mensen de neiging hebben om onbewust sociale regels en gedragingen toe te passen in interacties met computers, ondanks het feit dat ze weten dat deze computers levenloos zijn (Nass et al., 1994). Dit kan verder worden verklaard door het fenomeen antropomorfisme (Epley et al., 2007; Novak & Hoffman, 2019). Antropomorfisme houdt in dat de aanwezigheid van mensachtige kenmerken of gedragingen in niet-menselijke agenten, onbewust cognitieve schema's voor menselijke interactie activeert (Aggarwal & McGill, 2007; M.K. Lee et al., 2010). Door computers te antropomorfiseren komen mensen tegemoet aan hun eigen behoefte aan sociale verbinding en begrip van de sociale omgeving (Epley et al., 2007; Waytz et al., 2010). Dit heeft echter ook tot gevolg dat mensen cognitieve schema’s voor sociale perceptie toepassen op conversational agents.

Using Lo-Fi prototyping to envision conversational agents in public settings

Speech interactions are often associated with virtual assistants and smart home devices, designed primarily for private contexts. A less developed domain is speech interfaces in public contexts. In a smart city development project, we explored the potential of distributed conversational speech interfaces in lampposts. Deploying a research-through-design method, we created a lo-fi prototype of the speech interface that test subjects could interact with during experiments in a lab setting. Our first exploratory prototype consisted of a loudspeaker that acted as the interface and preconceived dialogues designed to investigate the boundaries of desirable and acceptable experiences regarding issues such as privacy. Experiencing the interaction with this rudimentary prototype helped people envision potential use cases and reflect on privacy issues: the dialogues revealed subjective limits of what kind of (personal) information people were willing to share with the lamppost. They also elicited thoughts on possible consequences in the social context of citizens. LinkedIn: https://www.linkedin.com/in/atjylha/ https://www.linkedin.com/in/josvanleeuwen/

Using Lo-Fi experience prototypes for co-designing conversational speech interactions for public settings

In an on-going smart city development project, we are exploring the potential of distributed conversational speech interfaces in the public context of a city. By using the well-known Wizard of Oz method in combination with a lo-fi prototype, we involve participants in co-design with the focus on potential use cases, social acceptability, and privacy aspects of interacting with a speech interface publicly. The work taps into the gap of design-oriented work in the domain of speech-based HCI. LinkedIn: https://www.linkedin.com/in/atjylha/ https://www.linkedin.com/in/josvanleeuwen/

CUBE: Conversational User-Interface-Based Embodiment: developing a digital humans embodiment for conversational agents: design, implementation, and integration challenges.

Embodied Agents for Obstetric Simulation Training

Post-partum hemorrhaging is a medical emergency that occurs during childbirth and, in extreme cases, can be life-threatening. It is the number one cause of maternal mortality worldwide. High-quality training of medical staff can contribute to early diagnosis and work towards preventing escalation towards more serious cases. Healthcare education uses manikin-based simulators to train obstetricians for various childbirth scenarios before training on real patients. However, these medical simulators lack certain key features portraying important symptoms and are incapable of communicating with the trainees. The authors present a digital embodiment agent that can improve the current state of the art by providing a specification of the requirements as well as an extensive design and development approach. This digital embodiment allows educators to respond and role-play as the patient in real time and can easily be integrated with existing training procedures. This research was performed in collaboration with medical experts, making a new contribution to medical training by bringing digital humans and the representation of affective interfaces to the field of healthcare.

Design of the user interface for “Stappy”, a sensorfeedback system to facilitate walking in people after stroke

Introduction: Sensor-feedback systems can be used to support people after stroke during independent practice of gait. The main aim of the study was to describe the user-centred approach to (re)design the user interface of the sensor feedback system “Stappy” for people after stroke, and share the deliverables and key observations from this process. Methods: The user-centred approach was structured around four phases (the discovery, definition, development and delivery phase) which were fundamental to the design process. Fifteen participants with cognitive and/or physical limitations participated (10 women, 2/3 older than 65). Prototypes were evaluated in multiple test rounds, consisting of 2–7 individual test sessions. Results: Seven deliverables were created: a list of design requirements, a personae, a user flow, a low-, medium- and high-fidelity prototype and the character “Stappy”. The first six deliverables were necessary tools to design the user interface, whereas the character was a solution resulting from this design process. Key observations related to “readability and contrast of visual information”, “understanding and remembering information”, “physical limitations” were confirmed by and “empathy” was additionally derived from the design process. Conclusions: The study offers a structured methodology resulting in deliverables and key observations, which can be used to (re)design meaningful user interfaces for people after stroke. Additionally, the study provides a technique that may promote “empathy” through the creation of the character Stappy. The description may provide guidance for health care professionals, researchers or designers in future user interface design projects in which existing products are redesigned for people after stroke.