How it works: a field study of non-technical users interacting with an intelligent system

By Joe Tullio, Anind K. Dey, Jason Chalecki, James A. Fogarty. 2007

166 citations recorded by the source corpus.

Subfields: Information Systems, Information Systems and Management, Social Psychology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Information Systems and Management

In Social Psychology

Citations in the corpus

Explanation in Human-AI Systems: A Literature Meta-Review, Synopsis of Key Ideas and Publications, and Bibliography for Explainable AI (2019, background): “…Pratt, 2001; Pirolli and Card, 1999, 2005, Prietula et al., 2000, 2000; Roth, Butterworth, and Loftus, 1985; Sebok and Wickens, 2017; Strauch, 2017; Tullio et al., 2007; Vermeulen, Luyten, van den Hoven, and Coninx, 2013; Wallace and Freuder, 2001; Wille, 1997; Wright, Chen, Barnes, and Hancock,…”
'It's Reducing a Human Being to a Percentage': Perceptions of Justice in Algorithmic Decisions (2018, background): “As users form mental models [52] of machine learning systems over time, such long-term exposure may lead to greater understanding [76] and greater trust and reliance [47].”
How People Explain Action (and Autonomous Intelligent Systems Should Too) (2017, background): “Moreover, a long-term study (Tullio et al. 2007) showed that initially established (incorrect) mental models of an intelligent information system remained robust over time, even when details of the system’s implementation were given and initial beliefs were challenged with contradictory evidence.”
How the Experts Do It: Assessing and Explaining Agent Behaviors in Real-Time Strategy Games (2017, background): “Much of the prior research into explaining agent behavior starts at some kind of observable effect and then explains something about that effect or its causes [10, 12, 16, 31].”
What Do We Want From Explainable Artificial Intelligence (XAI)? - A Stakeholder Perspective on XAI and a Conceptual Model Guiding Interdisciplinary XAI Research (2021, background): “Specifically, stakeholders without any prior experience with a given system will likely start with a degree of understanding that corresponds to their (background) knowledge of artificial systems or arose from initial instructions they have received regarding the system [17, 158].”
Interacting meaningfully with machine learning systems: Three experiments (2009, background): “A recent field study of eight non-technical users interacting with an intelligent system that predicted a person’s interruptibility using sensor data (Tullio et al. 2007) is also relevant to our work.”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “The method for explainability tested in [340] listed the two most relevant features and the prediction of the model.”
Toolkit to support intelligibility in context-aware applications (2010, methodology): “[42] explained interruptibility inferred from decision trees and naïve Bayes with What explanations.”
Tell me more?: the effects of mental model soundness on personalizing an intelligent agent (2012, background): “One challenge , however, is that mental models, once built, can be surprisingly hard to shift, even when people are aware of contradictory evidence [28].”
Where Are My Intelligent Assistant's Mistakes? A Systematic Testing Approach (2011, unknown): “, “talking detected”) [31].”
Explanatory Debugging: Supporting End-User Debugging of Machine-Learned Programs (2010, background): “learning new information contradicts findings by [23], whose participants’ mental models were persistent (possibly because counter-evidence was not presented early enough for them to discard already entrenched mental models).”
Too much, too little, or just right? Ways explanations impact end users' mental models (2013, background): “…agent that predicted office workers’ availability, finding that even when faced with contradictory evidence, users often tried to fit it into their existing mental model of the agent [19]; this suggests that it is important to help end users develop correct models as quickly as possible.”
Principles of Explanatory Debugging to Personalize Interactive Machine Learning (2015, background): “Without explanations, however, users struggle to build accurate mental models of such systems [23, 30, 43].”