Eric D. Ragan

Jeremy E. Block, Eric D. Ragan

Many interactive data systems combine visual representations of data with embedded algorithmic support for automation and data exploration. To effectively support transparent and explainable data systems, it is important for researchers and designers to know how users understand the system. We discuss the evaluation of users' mental models of system logic. Mental models are challenging to capture and analyze. While common evaluation methods aim to approximate the user's final mental model after a period of system usage, user understanding continuously evolves as users interact with a system over time. In this paper, we review many common mental model measurement techniques, discuss tradeoffs, and recommend methods for deeper, more meaningful evaluation of mental models when using interactive data analysis and visualization systems. We present guidelines for evaluating mental models over time that reveal the evolution of specific model updates and how they may map to the particular use of interface features and data queries. By asking users to describe what they know and how they know it, researchers can collect structured, time-ordered insight into a user's conceptualization process while also helping guide users to their own discoveries.

Donald R. Honeycutt, Mahsan Nourani, Eric D. Ragan

Mixed-initiative systems allow users to interactively provide feedback to potentially improve system performance. Human feedback can correct model errors and update model parameters to dynamically adapt to changing data. Additionally, many users desire the ability to have a greater level of control and fix perceived flaws in systems they rely on. However, how the ability to provide feedback to autonomous systems influences user trust is a largely unexplored area of research. Our research investigates how the act of providing feedback can affect user understanding of an intelligent system and its accuracy. We present a controlled experiment using a simulated object detection system with image data to study the effects of interactive feedback collection on user impressions. The results show that providing human-in-the-loop feedback lowered both participants' trust in the system and their perception of system accuracy, regardless of whether the system accuracy improved in response to their feedback. These results highlight the importance of considering the effects of allowing end-user feedback on user trust when designing intelligent systems.

Mahsan Nourani, Joanie T. King, Eric D. Ragan

Domain-specific intelligent systems are meant to help system users in their decision-making process. Many systems aim to simultaneously support different users with varying levels of domain expertise, but prior domain knowledge can affect user trust and confidence in detecting system errors. While it is also known that user trust can be influenced by first impressions with intelligent systems, our research explores the relationship between ordering bias and domain expertise when encountering errors in intelligent systems. In this paper, we present a controlled user study to explore the role of domain knowledge in establishing trust and susceptibility to the influence of first impressions on user trust. Participants reviewed an explainable image classifier with a constant accuracy and two different orders of observing system errors (observing errors in the beginning of usage vs. in the end). Our findings indicate that encountering errors early-on can cause negative first impressions for domain experts, negatively impacting their trust over the course of interactions. However, encountering correct outputs early helps more knowledgable users to dynamically adjust their trust based on their observations of system performance. In contrast, novice users suffer from over-reliance due to their lack of proper knowledge to detect errors.

Mahsan Nourani, Chiradeep Roy, Tahrima Rahman et al.

Explainable machine learning and artificial intelligence models have been used to justify a model's decision-making process. This added transparency aims to help improve user performance and understanding of the underlying model. However, in practice, explainable systems face many open questions and challenges. Specifically, designers might reduce the complexity of deep learning models in order to provide interpretability. The explanations generated by these simplified models, however, might not accurately justify and be truthful to the model. This can further add confusion to the users as they might not find the explanations meaningful with respect to the model predictions. Understanding how these explanations affect user behavior is an ongoing challenge. In this paper, we explore how explanation veracity affects user performance and agreement in intelligent systems. Through a controlled user study with an explainable activity recognition system, we compare variations in explanation veracity for a video review and querying task. The results suggest that low veracity explanations significantly decrease user performance and agreement compared to both accurate explanations and a system without explanations. These findings demonstrate the importance of accurate and understandable explanations and caution that poor explanations can sometimes be worse than no explanations with respect to their effect on user performance and reliance on an AI system.

Fan Yang, Shiva K. Pentyala, Sina Mohseni et al.

In this demo paper, we present the XFake system, an explainable fake news detector that assists end-users to identify news credibility. To effectively detect and interpret the fakeness of news items, we jointly consider both attributes (e.g., speaker) and statements. Specifically, MIMIC, ATTN and PERT frameworks are designed, where MIMIC is built for attribute analysis, ATTN is for statement semantic analysis and PERT is for statement linguistic analysis. Beyond the explanations extracted from the designed frameworks, relevant supporting examples as well as visualization are further provided to facilitate the interpretation. Our implemented system is demonstrated on a real-world dataset crawled from PolitiFact, where thousands of verified political news have been collected.

Sina Mohseni, Niloofar Zarei, Eric D. Ragan

The need for interpretable and accountable intelligent systems grows along with the prevalence of artificial intelligence applications used in everyday life. Explainable intelligent systems are designed to self-explain the reasoning behind system decisions and predictions, and researchers from different disciplines work together to define, design, and evaluate interpretable systems. However, scholars from different disciplines focus on different objectives and fairly independent topics of interpretable machine learning research, which poses challenges for identifying appropriate design and evaluation methodology and consolidating knowledge across efforts. To this end, this paper presents a survey and framework intended to share knowledge and experiences of XAI design and evaluation methods across multiple disciplines. Aiming to support diverse design goals and evaluation methods in XAI research, after a thorough review of XAI related papers in the fields of machine learning, visualization, and human-computer interaction, we present a categorization of interpretable machine learning design goals and evaluation methods to show a mapping between design goals for different XAI user groups and their evaluation methods. From our findings, we develop a framework with step-by-step design guidelines paired with evaluation methods to close the iterative design and evaluation cycles in multidisciplinary XAI teams. Further, we provide summarized ready-to-use tables of evaluation methods and recommendations for different goals in XAI research.

Sina Mohseni, Alyssa Pena, Eric D. Ragan

Our work aims to generate visualizations to enable meta-analysis of analytic provenance and aid better understanding of analysts' strategies during exploratory text analysis. We introduce ProvThreads, a visual analytics approach that incorporates interactive topic modeling outcomes to illustrate relationships between user interactions and the data topics under investigation. ProvThreads uses a series of continuous analysis paths called topic threads to demonstrate both topic coverage and the progression of an investigation over time. As an analyst interacts with different pieces of data during the analysis, interactions are logged and used to track user interests in topics over time. A line chart shows different amounts of interest in multiple topics over the duration of the analysis. We discuss how different configurations of ProvThreads can be used to reveal changes in focus throughout an analysis.

Sina Mohseni, Andrew Pachuilo, Ehsanul Haque Nirjhar et al.

We present an analytic provenance data repository that can be used to study human analysis activity, thought processes, and software interaction with visual analysis tools during exploratory data analysis. We conducted a series of user studies involving exploratory data analysis scenario with textual and cyber security data. Interactions logs, think-alouds, videos and all coded data in this study are available online for research purposes. Analysis sessions are segmented in multiple sub-task steps based on user think-alouds, video and audios captured during the studies. These analytic provenance datasets can be used for research involving tools and techniques for analyzing interaction logs and analysis history. By providing high-quality coded data along with interaction logs, it is possible to compare algorithmic data processing techniques to the ground-truth records of analysis history.

Sina Mohseni, Jeremy E. Block, Eric D. Ragan

Research in interpretable machine learning proposes different computational and human subject approaches to evaluate model saliency explanations. These approaches measure different qualities of explanations to achieve diverse goals in designing interpretable machine learning systems. In this paper, we propose a human attention benchmark for image and text domains using multi-layer human attention masks aggregated from multiple human annotators. We then present an evaluation study to evaluate model saliency explanations obtained using Grad-cam and LIME techniques. We demonstrate our benchmark's utility for quantitative evaluation of model explanations by comparing it with human subjective ratings and ground-truth single-layer segmentation masks evaluations. Our study results show that our threshold agnostic evaluation method with the human attention baseline is more effective than single-layer object segmentation masks to ground truth. Our experiments also reveal user biases in the subjective rating of model saliency explanations.