Social robots are a lesser known technology with uncertain but seemingly very powerful potential, which for decades has been portrayed in cultural artifacts as threats to human primacy. Research on people’s relationships to non-robotic technology, however, indicates that people will treat robots socially and assimilate them into their lives in ways that may disrupt existing norms but still fulfill a fundamental human need. Through the theoretical lenses of media equation and apparatgiest, this dissertation examines facets of robot humanization, defined as how people think of robots as social and human-like entities through perceptions of liking, human-likeness, and rights’ entitlement. In a 2 (gender) x 2 (physical humanness) x 3 (status) between-subjects online experiment, this dissertation explores the influence of fixed technological traits (the robot’s gender, physical humanness, and described status) and participants’ individual differences on humanization perceptions. Findings show that the robots’ features mattered less than participants’ individual traits, which explained the most variance in humanizing perceptions of social robots. Of those, participants’ prior robot exposure (both in real life and mediated) and efficacy traits were the strongest predictors of robot liking, perceived human-likeness, and perceptions of rights entitlement. Specifically, those with more real-life exposure and who perceived themselves as more technologically competent were more likely to humanize robots, while those with higher internal loci of control and negative mediated views of robots were less inclined to humanize robots. Theoretically, this study’s findings suggest that technological affordances matter less than the ontological understanding that social robots as a category may have in people’s humanizing perceptions. Looking forward, these findings indicate that there is an opportunity in the design of social robots to set precedents now that are prosocial and reflective of the world people strive for and want to inhabit in the future.

Unsupervised image clustering is a challenging and often ill-posed problem. Existing image descriptors fail to capture the clustering criterion well, and more importantly, the criterion itself may depend on (unknown) user preferences. Semi-supervised approaches such as distance metric learning and constrained clustering thus leverage user-provided annotations indicating which pairs of images belong to the same cluster (must-link) and which ones do not (cannot-link). These approaches require many such constraints before achieving good clustering performance because each constraint only provides weak cues about the desired clustering. In this work, we propose to use image attributes as a modality for the user to provide more informative cues. In particular, the clustering algorithm iteratively and actively queries a user with an image pair. Instead of the user simply providing a must-link/cannot-link constraint for the pair, the user also provides an attribute-based reasoning e.g. "these two images are similar because both are natural and have still water'' or "these two people are dissimilar because one is way older than the other''. Under the guidance of this explanation, and equipped with attribute predictors, many additional constraints are automatically generated. We demonstrate the effectiveness of our approach by incorporating the proposed attribute-based explanations in three standard semi-supervised clustering algorithms: Constrained K-Means, MPCK-Means, and Spectral Clustering, on three domains: scenes, shoes, and faces, using both binary and relative attributes. / Master of Science

Human-Machine Communication and fields like journalism studies have been discussing new technological developments in journalism, especially automation technologies like automated writing software. However, existing literature has terminological problems: Terms are not distinctly defined and delimited, different aspects can be referred to with the same term, while different, often misleading, terms exist for the same aspect. As a result, it is often unclear which concept is being referred to. To gain a better understanding and modeling of automation in journalism as well as a theoretical foundation, this paper first describes current problems with terms used in scientific literature and argues that existing automation taxonomies are not fully transferrable to journalism, making a new theoretical basis necessary. Subsequently, Rammert and Schulz-Schaeffer’s concept of distributed and gradualised action is described and proposed as such a theoretical basis for the unification of terminology and conceptual foundations, providing the opportunity to empirically and normatively describe automation as well as delivering necessary theoretical underpinnings. Lastly, the concept is applied to automation in journalism, resulting in a proposed automation concept, suggestions for terminology, and further implications for Human-Machine Communication theory.