The Paradoxical Role of Humanness in Aggression Toward Conversational Agents

2023 | Journal of Management Information Systems | Citations: 0

Authors: Brendel, Alfred Benedikt; Hildebrandt, Fabian; Dennis, Alan R.; Riquel, Johannes

Abstract: Conversational Agents (CAs) are becoming part of our everyday lives. About 10 pe ... Expand

Abstract: Conversational Agents (CAs) are becoming part of our everyday lives. About 10 percent of users display aggressive behavior toward CAs, such as swearing at them when they produce errors. We conducted two online experiments to understand user aggression toward CAs better. In the first experiment, 175 participants used either a humanlike CA or a non-humanlike CA. Both CAs worked without errors, and we observed no increased frustration or user aggression. The second experiment (with 201 participants) was the focus of this study; in it, both CAs produce a series of errors. The results show that frustration with errors drives aggression, and users with higher impulsivity are more likely to become aggressive when frustrated. The results also suggest that there are three pathways by which perceived humanness influences users' aggression to CAs. First, perceived humanness directly increases the frustration with the CA when it produces errors. Second, perceived humanness increases service satisfaction which in turn reduces frustration. Third, perceived humanness influences the nature of aggression when users become frustrated (i.e., users are less likely to use highly offensive words with a more humanlike CA). Our research contributes to our theoretical understanding of the role of anthropomorphism in the interaction with machines, showing that designing a CA to be more humanlike is a double-edged sword—both increasing and decreasing the frustration that leads to aggression—and also a means to reduce the most severe aggression. Collapse

Topics: Wikipedia user behavior ease of use electronic mail personality

Methods: experiment theory development statistical hypothesis test chatbot SEM tool

Theories: social response theory

Using machine learning to create and capture value in the business models of small and medium-sized enterprises

2023 | International Journal of Information Management | Citations: 0

Authors: Costa-Climent, Ricardo; Haftor, Darek M.; Staniewski, Marcin W.

Abstract: Start-ups have revolutionised many economic ecosystems, becoming innovation pion ... Expand

Abstract: Start-ups have revolutionised many economic ecosystems, becoming innovation pioneers around the world. Most are based on data-driven business models, particularly relying on machine learning technologies. However, not all start-ups that use machine learning technologies manage to create and capture value. The existing literature on the use value enabled by information technologies does not take into account the unique capabilities of machine learning. The theory of data network effects offers a promising explanation of how to create value using machine learning. However, it does not explicitly describe how to capture value using machine learning. In contrast, business model theory explains how companies use technologies to create and capture value, but not specifically through the use of machine learning technology. Therefore, this study aims to improve the theo­ retical understanding of the key drivers of value creation and capture in start-ups with business models driven by this kind of technology. Statistical techniques are used in a sample of 122 start-ups to explore the theoretical relationships between these two theories. The analysis reveals the link between specific value creation and capture factors of the two theories, such as efficiency, novelty, and performance expectancy. The study also provides evidence of the need to adopt a co-evolutionary perspective of value creation and capture through the use of machine learning. Collapse

Topics: value creation Twitter social media database system website

Methods: descriptive statistic theory development regression analysis method general multivariate regression model linear regression analysis

New Bots – The Influence of a Conversational Agent’s Rookie Personality on Users’ Satisfaction

2023 | International Conference on Information Systems | Citations: 0

Authors: Hildebrandt, Fabian; Brendel, Benedikt; Dennis, Alan; Sachdeva, Agrim

Abstract: Conversational agents (CAs) are not likely to be error-free, and efforts are b ... Expand

Abstract: Conversational agents (CAs) are not likely to be error-free, and efforts are being made by research and practice to mitigate the negative consequences of such errors (e.g., reduced service satisfaction). In this context, our study examines the impact of a CA's rookie personality (i.e., the CA expresses that it is new and still learning) on users. Our findings reveal that the rookie personality is a double-edged sword: while it increases users' perception of humanness, which increases the perception of reliability, it also directly reduces perceived reliability, resulting in less service satisfaction. To explain these seemingly contradictory effects, we turn to the dual processing theory of cognition and propose that the rookie personality influences both automatic and deliberate thinking. Users actively and consciously contemplate the CA's messages, leading them to view the software artifact as "broken" and low-quality. Additionally, users' automatic thinking is influenced by the perception of humanness. Collapse

Topics: personality usability human computer interaction Google+ electronic mail

Methods: chatbot experiment survey statistical power analysis partial least squares regression

Theories: cognitive theory expectation confirmation theory

How Sociotechnical Realignment and Sentiments Concerning Remote Work are Related – Insights from the COVID-19 Pandemic

2023 | Business & Information Systems Engineering | Citations: 0

Authors: Kohn, Vanessa; Frank, Muriel; Holten, Roland

Abstract: The COVID-19 pandemic forced sociotechnical systems (STS) to highly integrate re ... Expand

Abstract: The COVID-19 pandemic forced sociotechnical systems (STS) to highly integrate remote work. Large-scale analyses show that the positivity of tweets about work from home decreased until COVID-19 was declared a pandemic by the WHO and re-increased in the weeks that followed. Nevertheless, it is unclear if this reaction is due to personal and organizational developments or if it mirrors the realignment of entire STS. The present study uses Q methodology to identify differences in how STS realign to the externally enforced integration of remote work. Only STS that reach a state of high alignment to remote work conditions by successfully shifting communication and procedures to digital spheres can be considered resilient. The results show that employees describe their personal experiences with remote work as more positive the higher their level of sociotechnical realignment. Furthermore, personal digital resilience is correlated to successful STS realignment as well. The results confirm the importance of realigning not only the technical and social components of STS but above all their sociotechnical interaction. Negative sentiments relate in particular to the low realization of humanistic objectives in STS. Collapse

Topics: remote work IT supported collaboration pandemic productivity privacy

Methods: sentiment analysis Q methodology qualitative interview linear regression analysis Mann Whitney U test

Theories: sociotechnical systems theory socio technical theory

AI-based Technologies for Everyone: How and Why to Adapt Voice Assistants’ Complexity to Older Adults

2023 | International Conference on Information Systems | Citations: 0

Authors: Rzepka, Christine; Berger, Benedikt; Tams, Stefan

Abstract: Technological advancements in the area of artificial intelligence have rapidly ... Expand

Abstract: Technological advancements in the area of artificial intelligence have rapidly improved the performance of speech recognition and natural language processing. These improvements have facilitated the proliferation of voice assistants (VAs), which can understand human speech and provide spoken answers to assist in various tasks. More and more individuals and organizations adopt VAs because they value the naturalness of speech interaction. However, speech interaction is of ephemeral nature and processed in sequential order, which puts cognitive load on the user. Therefore, we investigate the relationship between the complexity of speech interaction and the interaction outcomes enjoyment, satisfaction, and intention to explore. Our results show that this relationship has an inverted U-shape for people with above-median information processing speed (i.e., younger adults) but is negatively linear otherwise. The results contribute to the literature on interface complexity and on the use of IT systems by the elder. Collapse

Topics: decision support system website information system use password Google+

Methods: survey experiment regression analysis method statistical hypothesis test experimental group

Theories: complexity theory compensatory adaptation theory

Trusting the trust machine: Evaluating trust signals of blockchain applications

2023 | International Journal of Information Management | Citations: 0

Authors: Völter, Fabiane; Urbach, Nils; Padget, Julian

Abstract: Information systems research emphasizes that blockchain requires trust in the te ... Expand

Abstract: Information systems research emphasizes that blockchain requires trust in the technology itself. However, we lack knowledge on the applicability of established trust cues to blockchain technology. Thus, this paper’s objective is to empirically evaluate the effectiveness of several established IS trust formation factors on end user trust. We do so by conducting a between-groups experiment. While we can validate the applicability of previous IS trust research for blockchain technology to some extent, we find that trust signals emphasizing the technol­ ogy’s underlying trust-building characteristics are most effective. Hence, we highlight the need for contextual­ ization of trust research on blockchain technology. We provide both researchers and practitioners with insights for building trustworthy blockchain applications that enable trust-less interactions not only in theory but in practice. Collapse

Topics: blockchain human computer interaction usability IT work Bitcoin

Methods: experiment experimental group design artifact parametric test survey

Exploring firm strategy using financial reports: performance impact of inward and outward relatedness with digitisation

2022 | European Journal of Information Systems | Citations: 0

Authors: Jabr, Wael; Zheng, Zhiqiang (Eric)

Abstract: A firm’s success critically hinges on its strategies in selecting its portfolio ... Expand

Abstract: A firm’s success critically hinges on its strategies in selecting its portfolio of products and services. In this paper, we study how differentiation and market alignment at the offering level impact firm performance. To that end, we mine firms’ 10-K filings to characterise the portfolio of offerings through the lens of outward relatedness, inward relatedness, and digitisation. We define outward relatedness as a measure of alignment of firm offerings within its market space, inward relatedness as a measure of differentiation of firm offerings with its own past offerings, and digitisation as a measure of the firm’s focus on IT. We find that markets react positively to firms that operate with high levels of outward relatedness, low levels of inward relatedness and high levels of digitisation. However, we find that highly digitised firms do not have to conform to peers’ offerings. Digitisation enables these firms to differentiate by internally diversifying their offerings. Interestingly, our results show that only firms already highly digitised benefit from further digitisation. Collapse

Topics: organizational productivity innovation management research and development database system digital innovation

Methods: natural language processing Compustat descriptive statistic longitudinal research case study

Voice Assistant vs. Chatbot – Examining the Fit Between Conversational Agents’ Interaction Modalities and Information Search Tasks

2022 | Information Systems Frontiers | Citations: 0

Authors: Rzepka, Christine; Berger, Benedikt; Hess, Thomas

Abstract: Owing to technological advancements in artificial intelligence, voice assistants ... Expand

Abstract: Owing to technological advancements in artificial intelligence, voice assistants (VAs) offer speech as a new interaction modality. Compared to text-based interaction, speech is natural and intuitive, which is why companies use VAs in customer service. However, we do not yet know for which kinds of tasks speech is beneficial. Drawing on task-technology fit theory, we present a research model to examine the applicability of VAs to different tasks. To test this model, we conducted a laboratory experiment with 116 participants who had to complete an information search task with a VA or a chatbot. The results show that speech exhibits higher perceived efficiency, lower cognitive effort, higher enjoyment, and higher service satisfaction than text-based interaction. We also find that these effects depend on the task’s goal-directedness. These findings extend task-technology fit theory to customers’ choice of interaction modalities and inform practitioners about the use of VAs for information search tasks. Collapse

Topics: information storage and retrieval system task characteristic Google+ human computer interaction smart speaker

Methods: experiment laboratory experiment survey factor analysis data transformation

Theories: cognitive fit theory theory of task technology fit cognitive load theory

Privacy Risks in Digital Markets: The Impact of Ambiguity Attitudes on Transparency Choices

2022 | International Conference on Information Systems | Citations: 0

Authors: Sachs, Nikolai; Schnurr, Daniel

Abstract: Transparency is viewed as an essential prerequisite for consumers to make infor ... Expand

Abstract: Transparency is viewed as an essential prerequisite for consumers to make informed privacy decisions in digital markets. However, it remains an open research question whether and when individuals actually prefer transparency about privacy risks when given a chance to avoid it. We investigate this question with a randomized controlled online experiment based on an Ellsberg-type design, where subjects repeatedly choose between risk and ambiguity while facing the threat of an actual disclosure of their personal data. We find empirical support for ambiguity attitudes as a novel behavioral mechanism underlying people's transparency choices in privacy contexts. In particular, we find that most individuals avoid ambiguity and prefer transparency for low likelihood privacy losses. However, this pattern reverses for high likelihood losses and when subjects perceive data disclosure as a gain. Most notably, a significant share of people seek ambiguity and thus prefer to avoid transparency when facing high likelihood privacy risks. Collapse

Topics: privacy video conference decision making mobile system internet of things

Methods: experiment archival research chi squared test exact binomial test experimental design

Theories: behavioral economics transaction cost economics

Executive Functions and Information Systems Learning

2022 | Management Information Systems Quarterly | Citations: 0

Authors: Sullivan, Yulia; Davis, Fred; Koh, Chang

Abstract: Information systems (IS) are complex and effortful, placing ever-greater demands ... Expand

Abstract: Information systems (IS) are complex and effortful, placing ever-greater demands on humans’ executive functions. Executive functions, general-purpose control processes that regulate one’s thoughts and behaviors, are the subject of growing investigation in cognitive psychology. The present research examines the relationship between individuals’ executive functions and IS learning. Using neuropsychological methods from cognitive psychology, we measured three key dimensions of executive functions: working memory, shifting, and inhibition. Two empirical studies were conducted. Study 1 tested the relationship between executive functions and IS learning in a self-paced offline learning environment. Study 2 replicated Study 1 and extended it to include a comparison of two self-paced online learning methods: behavior modeling and text-based learning. Both studies found significant effects of executive functions on IS learning after controlling for known IS learning determinants. Study 2 also showed that declarative knowledge was higher for behavior modeling than for text-based learning. Overall, our research highlights the influence of executive functions on IS learning. This research advances knowledge about determinants of IS learning and opens important research avenues for gaining deeper insights into cognitive mechanisms underlying effective IS learning. Collapse

Topics: online learning system self efficacy database system perceived ease of use perceived usefulness

Methods: partial least squares regression survey descriptive statistic experiment self reported survey

Theories: social cognitive theory skill acquisition theory Baddeley's model of working memory flow theory resource allocation theory