Stephan Schlögl

Florian Merkle, David Weber, Pascal Schöttle et al.

Over the last century, deep learning models have become the state-of-the-art for solving complex computer vision problems. These modern computer vision models have millions of parameters, which presents two major challenges: (1) the increased computational requirements hamper the deployment in resource-constrained environments, such as mobile or IoT devices, and (2) explaining the complex decisions of such networks to humans is challenging. Network pruning is a technical approach to reduce the complexity of models, where less important parameters are removed. The work presented in this paper investigates whether this reduction in technical complexity also helps with perceived explainability. To do so, we conducted a pre-study and two human-grounded experiments, assessing the effects of different pruning ratios on explainability. Overall, we evaluate four different compression rates (i.e., 2, 4, 8, and 32) with 37 500 tasks on Mechanical Turk. Results indicate that lower compression rates have a positive influence on explainability, while higher compression rates show negative effects. Furthermore, we were able to identify sweet spots that increase both the perceived explainability and the model's performance.

Luka Pavlič, Reinhard Bernsteiner, Stephan Schlögl et al.

In agile software development, breaking down user stories into actionable tasks is a critical yet time-consuming process. This paper investigates the potential of Generative AI tools to assist in task splitting, aiming to enhance planning efficiency. We conducted a controlled experiment comparing traditional task-splitting methods with AI-assisted approaches using GitLab Duo. Our findings indicate that while current AI tools are not yet mature enough to replace developers, they can aid in generating more granular task lists and ensuring no important tasks are overlooked. Participants favored a hybrid approach, combining AI tools with conventional methods to maintain high accuracy in planning. This study highlights the potential benefits and limitations of integrating Generative AI into agile development processes, suggesting that AI tools can serve as valuable aids in task splitting, provided there is human oversight to filter out irrelevant tasks.

Johanna Schmidhuber, Stephan Schlögl, Christian Ploder

The increasing progress in artificial intelligence and respective machine learning technology has fostered the proliferation of chatbots to the point where today they are being embedded into various human-technology interaction tasks. In enterprise contexts, the use of chatbots seeks to reduce labor costs and consequently increase productivity. For simple, repetitive customer service tasks such already proves beneficial, yet more complex collaborative knowledge work seems to require a better understanding of how the technology may best be integrated. Particularly, the additional mental burden which accompanies the use of these natural language based artificial assistants, often remains overlooked. To this end, cognitive load theory implies that unnecessary use of technology can induce additional extrinsic load and thus may have a contrary effect on users' productivity. The research presented in this paper thus reports on a study assessing cognitive load and productivity implications of human chatbot interaction in a realistic enterprise setting. A/B testing software-only vs. software + chatbot interaction, and the NASA TLX were used to evaluate and compare the cognitive load of two user groups. Results show that chatbot users experienced less cognitive load and were more productive than software-only users. Furthermore, they show lower frustration levels and better overall performance (i.e, task quality) despite their slightly longer average task completion time.

Christian Voigt, Stephan Schlögl, Aleksander Groth

Dark patterns utilize interface elements to trick users into performing unwanted actions. Online shopping websites often employ these manipulative mechanisms so as to increase their potential customer base, to boost their sales, or to optimize their advertising efforts. Although dark patterns are often successful, they clearly inhibit positive user experiences. Particularly, with respect to customers' perceived annoyance and trust put into a given brand, they may have negative effects. To investigate respective connections between the use of dark patterns, users' perceived level of annoyance and their expressed brand trust, we conducted an experiment-based survey. We implemented two versions of a fictitious online shop; i.e. one which used five different types of dark patterns and a similar one without such manipulative user interface elements. A total of $n=204$ participants were then forwarded to one of the two shops (approx. $2/3$ to the shop which used the dark patterns) and asked to buy a specific product. Subsequently, we measured participants' perceived annoyance level, their expressed brand trust and their affinity for technology. Results show a higher level of perceived annoyance with those who used the dark pattern version of the online shop. Also, we found a significant connection between perceived annoyance and participants' expressed brand trust. A connection between participants' affinity for technology and their ability to recognize and consequently counter dark patterns, however, is not supported by our data.

Anna Esposito, Terry Amorese, Marialucia Cuciniello et al.

This paper reports on a study conducted as part of the EU EMPATHIC project, whose goal is to develop an empathic virtual coach capable of enhancing seniors' well-being, focusing on user requirements and expectations with respect to participants' age and technology experiences (i.e. participants' familiarity with technological devices such as smartphones, laptops, and tablets). The data shows that seniors' favorite technological device is the smartphone, and this device was also the one that scored the highest in terms of easiness to use. We found statistically significant differences on the preferences expressed by seniors toward the gender of the agents. Seniors (independently from their gender) prefer to interact with female humanoid agents on both the pragmatic and hedonic dimensions of an interactive system and are more in favor to commit themselves in a long-lasting interaction with them. In addition, we found statistically significant effects of the seniors' technology savviness on the hedonic qualities of the proposed interactive systems. Seniors with technological experience felt less motivated and judged the proposed agents less captivating, exciting, and appealing.

Luisa Brinkschulte, Natascha Mariacher, Stephan Schlögl et al.

This paper outlines the EMPATHIC Research & Innovation project, which aims to research, innovate, explore and validate new interaction paradigms and plat-forms for future generations of Personalized Virtual Coaches to assist elderly people living independently at and around their home. Innovative multimodal face analytics, adaptive spoken dialogue systems, and natural language inter-faces are part of what the project investigates and innovates, aiming to help dependent aging persons and their carers. It will uses remote, non-intrusive technologies to extract physiological markers of emotional states and adapt respective coach responses. In doing so, it aims to develop causal models for emotionally believable coach-user interactions, which shall engage elders and thus keep off loneliness, sustain health, enhance quality of life, and simplify access to future telecare services. Through measurable end-user validations performed in Spain, Norway and France (and complementary user evaluations in Italy), the proposed methods and solutions will have to demonstrate useful-ness, reliability, flexibility and robustness.

Natascha Mariacher, Stephan Schlögl, Alexander Monz

With the ongoing penetration of conversational user interfaces, a better understanding of social and emotional characteristic inherent to dialogue is required. Chatbots in particular face the challenge of conveying human-like behaviour while being restricted to one channel of interaction, i.e., text. The goal of the presented work is thus to investigate whether characteristics of social intelligence embedded in human-chatbot interactions are perceivable by human interlocutors and if yes, whether such influences the experienced interaction quality. Focusing on the social intelligence dimensions Authenticity, Clarity and Empathy, we first used a questionnaire survey evaluating the level of perception in text utterances, and then conducted a Wizard of Oz study to investigate the effects of these utterances in a more interactive setting. Results show that people have great difficulties perceiving elements of social intelligence in text. While on the one hand they find anthropomorphic behaviour pleasant and positive for the naturalness of a dialogue, they may also perceive it as frightening and unsuitable when expressed by an artificial agent in the wrong way or at the wrong time.

Markus Thaler, Stephan Schlögl, Aleksander Groth

Visual appearance is an important aspect influencing the perception and consequent acceptance of Embodied Conversational Agents (ECA). To this end, the Uncanny Valley theory contradicts the common assumption that increased humanization of characters leads to better acceptance. Rather, it shows that anthropomorphic behavior may trigger feelings of eeriness and rejection in people. The work presented in this paper explores whether four different autonomous ECAs, specifically build for a European research project, are affected by this effect, and how they compare to two slightly more realistically looking human-controlled, i.e. face-tracked, ECAs with respect to perceived humanness, eeriness, and attractiveness. Short videos of the ECAs in combination with a validated questionnaire were used to investigate potential differences. Results support existing theories highlighting that increased perceived humanness correlates with increased perceived eeriness. Furthermore, it was found, that neither the gender of survey participants, their age, nor the sex of the ECA influences this effect, and that female ECAs are perceived to be significantly more attractive than their male counterparts.

Leigh Clark, Phillip Doyle, Diego Garaialde et al.

Speech interfaces are growing in popularity. Through a review of 68 research papers this work maps the trends, themes, findings and methods of empirical research on speech interfaces in HCI. We find that most studies are usability/theory-focused or explore wider system experiences, evaluating Wizard of Oz, prototypes, or developed systems by using self-report questionnaires to measure concepts like usability and user attitudes. A thematic analysis of the research found that speech HCI work focuses on nine key topics: system speech production, modality comparison, user speech production, assistive technology \& accessibility, design insight, experiences with interactive voice response (IVR) systems, using speech technology for development, people's experiences with intelligent personal assistants (IPAs) and how user memory affects speech interface interaction. From these insights we identify gaps and challenges in speech research, notably the need to develop theories of speech interface interaction, grow critical mass in this domain, increase design work, and expand research from single to multiple user interaction contexts so as to reflect current use contexts. We also highlight the need to improve measure reliability, validity and consistency, in the wild deployment and reduce barriers to building fully functional speech interfaces for research.