Julia Krüger

Desiderata for the Design of Companion Systems

We report about evaluations of the LAST MINUTE corpus which comprises multimodal recordings (audio, video, biopsychological data, verbatim transcripts, psychological questionnaires, in-depth user interviews) of Wizard of Oz simulated naturalistic human companion interactions in German. Based on the results of these analyses we discuss consequences for the design of future companion systems.

Users’ Relational Ascriptions in User-Companion Interaction

In a qualitative study in the field of user-companion interaction (UCI), we figured out that users of a system, which was meant to represent a preliminary step towards future Companion-systems, tend to individually ascribe (mostly human-like) characteristics to the system in order to turn it into a potential relational partner. Users’ intrinsic motivation to establish and maintain a relationship with these individualized systems was found throughout the analyses and led us to the development of a concept called ‘users’ relational ascriptions’. In this paper, we present the empirical background of this concept and describe defining characteristics of relational ascriptions, reasons for their formation, factors that influence their content and quality as well as factors, which are influenced by the ascriptions. We sum up with a definition of relational ascriptions and discuss practical implications and future work.

LAST MINUTE: An Empirical Experiment in User-Companion Interaction and Its Evaluation

The LAST MINUTE Corpus (LMC) is a unique resource for research on issues of Companion-technology. LMC not only comprises 57.5 h of multimodal recordings (audio, video, psycho-biological data) from interactions between users—133 subjects in sum, balanced in age and gender—and a WoZ-simulated speech-based interactive dialogue system. LMC also includes full verbatim transcripts of all these dialogues, sociodemographic and psychometric data of all subjects as well as material from 73 in-depth user interviews focusing the user’s individual experience of the interaction. In this chapter the experimental design and data collection of the LMC are shortly introduced. On this basis, exemplifying results from semantic analyses of the dialogue transcripts as well as from qualitative analyses of the interview material are presented. These illustrate LMC’s potential for investigations from numerous research perspectives.

Users’ Sense-Making of an Affective Intervention in Human-Computer Interaction

This qualitative interview study builds on an empirical experiment in which an affective intervention was given to users in a critical dialog situation of human-computer interaction (HCI). The applied intervention addressed users on a personal level by asking for their thoughts and feelings. Since this is still an unusual behavior for a technical system, the aim of the present study was to investigate how users reason about this. Three different kinds of individual sense-making processes regarding the intervention were worked out. These clarify that a personal level of interaction between system and user is only appropriate for some users, whereas it also can have adverse effects on others. By explicating users’ experiences and conceptions, this study contributes to research on affective interventions in HCI that in the past was mainly focused on measurements of effectiveness rather than on understanding users‘inner processes regarding such interventions.

Discourse Particles in Human-Human and Human-Computer Interaction --- Analysis and Evaluation

Discourse particles are verifiably used in both human-human interaction HHI and human-computer interaction HCI. In both types of interaction form-function-relations could be confirmed. Also correlations with specific subject characteristics, personality traits and the use of these particles could be uncovered. But these investigations are performed on separated datasets containing either HHI or HCI. Moreover, the subjects analyzed in both interaction types are not the same and thus, direct connections could not be made. In our contribution, we report about analyses of discourse particles in both HHI and HCI with the same subjects. This enables us to draw conclusions of the communication partners influence in relation to subject characteristics and personality traits. This will prospectively help to better understand the use of discourse particles. By using this knowledge, future technical systems can react to known subjects more individually.

Speech Matters – Psychological Aspects of Artificial versus Anthropomorphic System Voices in User-Companion Interaction

The design of this forthcoming study was created to investigate the influences of different system-voices on users while they interact with a simulated Companion-system. By using a Wizard of Oz experiment, we want to find out what kind of voice output (artificial vs. anthropomorphic) is better suited for keeping up users’ cooperation with a system while solving a task. The goal of this study is to gain a deeper understanding of influences of the speech-output in User-Companion Interaction. Users’ perceived trustworthiness towards the system, their experienced affective states and individual user characteristics as important mediators are the main focus of the present study.

NaLMC: A Database on Non-acted and Acted Emotional Sequences in HCI

We report on the investigation on acted and non-acted emotional speech and the resulting Non-/acted LAST MINUTE corpus (NaLMC) database. The database consists of newly recorded acted emotional speech samples which were designed to allow the direct comparison of acted and non-acted emotional speech. The non-acted samples are taken from the LAST MINUTE corpus (LMC) [1]. Furthermore, emotional labels were added to selected passages of the LMC and a self-rating of the LMC recordings was performed. Although the main objective of the NaLMC database is to allow the comparative analysis of acted and non-acted emotional speech, both audio and video signals were recorded to allow multimodal investigations.