Wizard Of Oz Prototyping Research Articles

‘Um, so like, is this how I speak?’: design implications for automated visual feedback systems on speech

ABSTRACT At a time when diverse forms of automated visual feedback systems on speech are introduced to the Human–Computer Interaction (HCI) community, our research investigates the possible design spaces and design implications to consider when creating them. We utilised a Wizard of Oz prototype that recognises fillers during speech and sends out real-time feedback in the form of a flickering screen and a post-report with graphic charts. We let 27 participants use this prototype and interviewed them about their general experience of using such systems. Based on thematic analysis, we propose eight design implications to consider when designing an automated visual feedback system on speech. Additionally, we discuss the possible future for automated visual feedback systems on speech where they collaborate with fields such as personal informatics (quantified self), self-determination theory (SDT) and motivation, and other socio-ethnic areas

Assistant Robot Enhances the Perceived Communication Quality of People With Dementia: A Proof of Concept

Almost all older people with dementia have progressive communication difficulties, which lead to increased social isolation and negative emotions. Thus, providing communication assistance for them is essential. This paper explores the feasibility of using social robots to assist older people with dementia in their face-to-face communication with others. We designed the behavior of a humanoid Pepper robot and made a Wizard of Oz prototype that the robot can serve as a personal memory assistant. The robot stores personal information for older people and assists in their communication through voice and screen display. In a video-based study with 88 participants, we investigated the effects of this assistive robot from a third-person observer perspective. Data were collected and analyzed using both three-way MANCOVAs for quantitative analysis and conventional content analysis for qualitative data. The results revealed that, by providing memory support, the robot significantly improved the observer's perceptions of an older person with dementia, including her perceived communication ability and performance, and personal image. Meanwhile, the communication is perceived to be significantly more effective when the robot assisted an older person. The willingness of others to communicate with more senior people also increased accordingly. Based on these findings, we present guidelines that may inform the design and development of communication assistant robots for older people with dementia.

Development of a Mobile Application for Smart Clinical Trial Subject Data Collection and Management

Wearable devices and digital health technologies have enabled the exchange of urgent clinical trial information. We developed an application to improve the functioning of decentralized clinical trials and performed a heuristic evaluation to reflect the user demands of existing clinical trial workers. The waterfall model of the software life cycle was used to guide the development. Focus group interviews (N = 7) were conducted to reflect the needs of clinical research professionals, and Wizard of Oz prototyping was performed to ensure high usability and completeness. Unit tests and heuristic evaluation (N = 11) were used. Thematic analysis was performed using the focus group interview data. Based on this analysis, the main menu was designed to include health management, laboratory test results, medications, concomitant medications, adverse reactions, questionnaires, meals, and My Alarm. Through role-playing, the functions and configuration of the prototype were adjusted and enhanced, and a heuristic evaluation was performed. None of the heuristic evaluation items indicated critical usability errors, suggesting that the revised prototype application can be practically applied to clinical trials. The application is expected to increase the efficiency of clinical trial management, and the development process introduced in this study will be helpful for researchers developing similar applications in the future.

How Players Speak to an Intelligent Game Character Using Natural Language Messages

AI-driven characters that learn directly from human input are rare in digital games, but recent advances in several fields of machine learning suggests that they may soon be much more feasible to create. This study explores the design space for interacting with such a character through natural language text dialogue. We conducted an observational study with 18 high school students, who played Minecraft alongside a Wizard of Oz prototype of a companion AI character that learned from their actions and inputs. In this paper, we report on an analysis of the 186 natural language messages that players sent to the character, and review key variations in syntax, function and writing style. We find that players’ behaviour and language was differentiated by the extent to which they expressed an anthropomorphic view of the AI character and the level of interest that they showed in interacting with it.

Evaluating Early Prototypes in Context: Trade-offs, Challenges, and Successes

We would like to say that experience prototyping and Wizard of Oz prototyping, taken together, constitute a suitable alternative to true contextual evaluation for early prototypes, one gaining realism at the expense of impartiality, the other impartiality at the expense of realism. As our experiences show, however, each pervasive application design poses unique challenges that you must consider when applying these approaches. Regardless, it's often a good trade-off to sacrifice some measure of realism to evaluate early prototypes. Evaluation of early pervasive computing prototypes in context is a pragmatic exercise, but one that is nonetheless informed by general approaches that reflect and adapt to the challenges of pervasive application development.

Embarking on Spoken-Language NL Interface Design

Natural language (NL) user interfaces are growing in popularity. Unfortunately, the complexity of NL interaction makes these interfaces difficult to design. For NL interfaces to become successful, universal tools are needed to help support the NL design process. What work practice should these tools explicitly support? Interviews with NL designers and our own experiments have identified a specific work practice that designers should consider as they begin to incorporate NL into user interface designs. The work practice study highlights the value of using Wizard of Oz prototyping in NL design. We describe a tool that we have built, called SUEDE, to explicitly support the first stage of NL design for spoken-language user interfaces. Our tools and tools like it will help make NL in human-computer interaction (HCI) more commonplace.