Kolloquium der Professur Mensch und Technik

Abstract: The proliferation of smart technology in everyday objects provide an opportunity for human-machine collaborations, signaling a shift from typical principal-agent relationships. The current work proposes roboticizing familiar user interfaces to allow objects to communicate with their users without the need of an additional communication modality. We implemented a light switch that is capable of kinetic gestures and report an exploratory study that investigated how naïve users (n=15) would respond to and interact with different gesture designs. A qualitative analysis of recorded interactions and semi-structured interviews reveal that users are surprised by unfamiliar automation and kinetic gestures serve as cues to promote discovery of the system’s mechanisms and mitigate assumed helplessness.

---

Bio: Maximilian Kullmann is a Ph.D. candidate at the Department of Human and Technology (Mensch und Technik).

Abstract: With the ever-growing role of automated technology in our lives, psychological research provides a scientific and humanistic foundation for human-centered design of this technology. An essential psychological prerequisite for efficient, safe and comfortable interaction with these systems is an appropriate level of trust in automation (calibrated trust). At this time, the psychological processes and influencing factors in learning about and understanding automated technology in forming and calibrating one’s trust are not fully understood. Starting from a closer look into the psychological processes and manifold of variables playing a role in trust formation and calibration, this talk presents the Three Stages of Trust framework for the investigation of trust in automated technology. Building on this, in this talk findings from studies applying the framework to the domains of automated driving and human-robot interaction are presented. The studies, amongst others, provide findings on the role of user (e.g., personality and anxiety) and system characteristics (e.g., reliability and human-likeness), available information about the system as well as its behavior after errors occur on the formation and calibration of peoples’ trust in them. Taken together, the studies’ findings support key propositions of the presented framework and extent the understanding of variables playing a role in the formation and calibration of trust in automation. In this, they support the relevance of psychological processes prior and during the interaction with automated systems. The talk concludes with a summary of findings and perspectives for their consideration in research and design of automated systems (e.g. social robots).

Abstract: I present research operationalizing some key concepts of “Embodied Interaction” - an approach in Human Computer Interaction (HCI) which sees interaction as emergent from improvisatory, coordinative processes spanning brain, body, and wider task context. “Embodied” approaches have been influential in HCI research since the 1980s, contributing important insights into how users behave in real contexts, and formalizing these into principles for design. However, these accounts have been resistant to the kinds of formalization which would allow quantitative measurement, hypothesis formation and incorporation into computational systems. This limits the potential to test and refine embodied design theories, and to formalize embodied design principles for incorporation into adaptive technologies. Responding to this, I draw on recent research in multifractality in behaviour, to operationalize the concepts of interactivity, and user-technology coupling, which I suggest underlie many accounts in embodied interaction. I relate this metric to experiential constructs which have been of interest to many HCI researchers — including fatigue, skill, and engagement. I point to future applications in understanding human-technology integration, and in developing readily-interpretable metrics for design. I also point to some of the open questions which need to be answered on the road to such applications.

---

Bio: I am a postdoctoral researcher in HCI, in Elisa Mekler’s group at Aalto University. My research connects the two strands of my academic background - philosophy and computer science, with my interest in cognitive science. I’m interested in ideas from phenomenology, enactivism, and ecological psychology, but also interested in the possibilities of quantitative and computational approaches. My thesis work focused on operationalizing some key ideas of embodiment using methods from ecological psychology and dynamical systems science. I am currently investigating the potential to apply these and similar methods to understand experience of control in computer games. My most recent work has been more theoretical in nature: working to map and clarify understandings of human agency and autonomy in HCI research.

Abstract: In human-computer interaction (HCI), proper use of theories and methods can help explain and predict interactive phenomena, inform design practice and improve scientific rigour. Classic examples are Fitt's law and the theory of distributed cognition applied in interfce design. However, theories and methods can not be simply transferred from one discipline to another. They need to be translated to make them relevant and useful for specific design requirements and context of use - scafolding theories for design. In this talk, I will present some research projects on scafolding the paradigm of crossmodal perception for interaction research, and the methods from psychology experiments for understanding user behaviours in the context of use. In addition, I will also discuss the theory scafolding for game design and research in my future projects.

---

Bio: Feng Feng is an HCI researcher and a designer at Aalto University, school of information technology. She studies multisensory experience and designs multisensory interactive technologies that aim to positively impact people's lives and augment sensory potentials. Before moving to Finland, she was a research associate at the BIG lab at the University of Bristol with a PhD from the Cognitive science research group, Queen Mary University of London.

Abstract: Terrorism is one of the most terrifying threats worldwide, and radicalization is a possible pathway to it. However, there is no common model underlying the complex process of radicalization, and research is usually undermined by a lack of studies based on empirical data. Therefore, a comprehensive model of radicalization was composed and tested empirically by coding biographies of terrorists from public sources and current cases managed by a police authority. In these data, we found specific individual preconditions to play a basic role, while, fueling the actual radicalization process, we observed a gradual increase of individual (e.g., need for significance), group (e.g., polarization) and catalyst processes (e.g., desensitization). Because social exclusion was found to be one important precondition, we systematically investigated its cause-and-effect relationship to radicalism using an experimental approach. By providing empirically founded data on radicalization, findings like these may contribute to the ongoing and urgent debate on the roots of terror and may also suggest how to decrease it.

---

Bio: Prof. Dr. Michaela Pfundmair is Professor Psychology in the Faculty of Intelligence at the Federal University of Administrative Sciences (Hochschule des Bundes) in Berlin, Germany. 
After studying Psychology in Salzburg, she went to LMU-Munich and received her PhD in Social Psychology in 2013 and her Habilitation in Psychology in 2015. Before she took her current position, she taught at the University of the Armed Forces in Munich and worked as Visiting Scholar at DePaul University Chicago. She also worked as interim professor of Social Psychology, inter alia, at LMU-Munich and the University of Klagenfurt. Furthermore, she is a forensic psychological expert assessing the credibility of witnesses and leading a nationwide team of forensic psychologists. Currently, she is the head of Legal Psychology at the German Association of Psychologists (BDP).
Her main line of research is about consequences of social exclusion and the psychology of radicalization on which she is leading several third-party-funded projects.

Abstract: Stereoscopic 3D (S3D) dashboards have been widely used in the field of autonomous driving. However, the design space and possibilities of S3D dashboards are not well understood. To investigate and apply such dashboards, we developed an integrated prototyping platform based on a virtual reality environment simulation and a car mock-up featuring a spatial augmented reality dashboard. In this talk, I will provide insights in various domains - driver distraction, trust, and take-over process - but also discuss new application domains and future research possibilities

---

Bio: Florian Weidner is a postdoctoral researcher at the Virtual Worlds and Digital Games group at Technische Universität Ilmenau, Germany. In 2021, he received his Ph.D. for his research on spatial augmented reality for virtual dashboards in vehicles. Before that, he studied media computer science at Technische Universität Dresden, Germany. His research interests include the high-level applications and fundamental processes of perception and cognition of mediated reality.

Abstract: My presentation will consist of three parts. In the first part, I will focus on human eye movements which are essential for visual perception. During fixations, information is extracted from the environment and internally processed. Since highest visual acuity is limited to the small foveal region, fast saccadic movements are required to redirect the foveal region from one fixation point to another. Analyzing fixation durations and saccade amplitudes during everyday activities allows understanding what details of the environment receive attention. With a combined analysis of fixations and saccades it can be determined how such details were processed. In the second part, I will show how knowledge about the mechanisms of eye movement control can be related to various areas of application. Since these findings are still very much based to research in the lab, the final part will be about our recent research efforts in relation to the perspectives of autonomous driving.

Abstract: Neuroscience, psychology, and many other fields, such as anthropology or philosophy, try to understand our cognition and cognitive processes. However, as time passed, new views on cognition emerged. One of the latest views on cognition, known as 4E cognition, refers to embedded, embodied, extended, and enacted cognition. Alternatively, to put it in simpler terms, our cognition and cognitive processes emerge from us by being in our environment, interacting with our environment, and enacting our actions within our environment. Although the need to study human cognition from a higher perspective led to the emergence of cognitive sciences, despite these advancements, our experimental methods have stayed relatively unchanged for the past centuries. The recent trends in cognitive science and related fields lean toward real-world experimentation. The main argument for real-world experimentation is the ecological validity of our experimentation and finding. However, despite all the positive voices advertising real-life experimentation, there are also significant concerns and voices against such a movement. Real-world is full of dynamics and sources of noises and events that have yet to be studied in detail. As alluring as the idea of moving out of the lab and doing experiments in real life is, the challenges of real-life experimentation should not be neglected, at least with our current methods and tool kits. However, one can partially abandon the control of the lab environment to get closer to real-life experimentation. Immersive virtual reality experiences can offer a close to real-life and interactive foundation for conducting cognitive science experiments. Virtual reality experiments can offer the same level of control over the conditions and precision in measurements as laboratory-based experimentation yet enable a realistic, immersive environment to simulate real-life situations.

Bio: Dr. Farbod Nosrat Nezami graduated from his Ph.D. program in September 2022 in the field of cognitive science at the University of Osnabrück, where he also did his master-level studies at the same faculty. During his Ph.D. he focused mainly on the aspect of acceptance and human-machine interaction with regards to Autonomous vehicles as well as virtual reality and experimental methods based on VR for conducting behavioral research. He is currently a post-doctorate research assistant at the University of Osnabrück with the focus on neural dynamics, modeling, and neuromorphic ( brain-inspired ) hardware design.

Abstract: Jeanette Falk will talk about her recent research on game jams and hackathons as examples on accelerated design processes. She uses these design formats as cases to reflect on the role of time and temporality in design processes and creativity. In her recent work she explores and discusses how short-term and long-term design processes shape design decisions and how, for example, creativity support tools may be designed to support different temporalities of design processes, focusing on fast design thinking.

Bio: Jeanette Falk got her Ph.D from Aarhus University, Denmark, and is currently a postdoc researcher at the Department of Artificial Intelligence and Human Interfaces, University of Salzburg. She has researched accelerated and time-limited formats, such as hackathons and game jams, and how these formats f.ex. shape participants’ creativity and design processes. Her recent research focus is how temporality plays a role in design and how we may design tools for supporting fast or slow design thinking, including how AI can be used as a design material as part of these kinds of tools.