Ryan M. Schuetzler

Establishing a foundation for automated human credibility screening

Automated human credibility screening is an emerging research area that has potential for high impact in fields as diverse as homeland security and accounting fraud detection. Systems that conduct interviews and make credibility judgments can provide objectivity, improved accuracy, and greater reliability to credibility assessment practices, need to be built. This study establishes a foundation for developing automated systems for human credibility screening.

Robustness of Multiple Indicators in Automated Screening Systems for Deception Detection

Abstract This study investigates the effectiveness of an automatic system for detection of deception by individuals with the use of multiple indicators of such potential deception. Deception detection research in the information systems discipline has postulated increased accuracy through a new class of screening systems that automatically conduct interviews and track multiple indicators of deception simultaneously. Understanding the robustness of this new class of systems and the limitations of its theoretical improved performance is important for refinement of the conceptual design. The design science proof-of-concept study presented here implemented and evaluated the robustness of these systems for automated screening for deception detection. A large experiment was used to evaluate the effectiveness of a constructed multiple-indicator system, both under normal conditions and with the presence of common types of countermeasures (mental and physical). The results shed light on the relative strength and robustness of various types of deception indicators within this new context. The findings further suggest the possibility of increased accuracy through the measurement of multiple indicators if classification algorithms can compensate for human attempts to counter effectiveness.

Examining the learning effects of live streaming video game instruction over Twitch

Abstract Technology facilitates advances in learning and drives learning paradigms. One recent innovation is Twitch™, an online streaming platform often used for video game tutorials but also enables amateur online instruction ( Hamilton, Garretson, & Kerne, 2014 )). Twitch represents a unique learning paradigm that is not perfectly represented in previous technologies because of its “ground-up” evolution and the opportunity for novice instructors to educate mass audiences in real-time over the Internet while enabling interaction between teachers and learners and among learners. The purpose of this research is to empirically examine the efficacy of Twitch as a learning platform by manipulating each of the key characteristics of Twitch and to understand the conditions in which novice instructors may be beneficial. Drawing from Cognitive Load Theory, we demonstrate the worked-example effect in the Twitch environment by manipulating teacher-learner-learner interactions, live versus recorded streaming, and expert-versus novice-based instruction. Based on a laboratory experiment involving 350 participants, we found that learning performance under novice instructors was at least as good as that of experts. However, an exploratory analysis of learner personalities revealed that extroverts benefit only when learner-learner interaction is enabled. Surprisingly, those who are highly agreeable and less neurotic benefited more from novice instructors.

The Value of Distrust in Computer-Based Decision-Making Groups

During crises, relief agency commanders have to make decisions in a complex and uncertain environment, requiring them to continuously adapt to unforeseen environmental changes. In the process of adaptation, the commanders depend on information management systems for information. Yet there are still numerous reports of situations in which commanders had to make decisions based on incomplete, outdated or incorrect information, indicating poor information quality. In many of these situations, poor information quality can be attributed to the information management process incapable of adapting to external (environmental) changes and internal (team) information needs. Using dynamic capability theory and the findings of a case study, this paper presents four principles for information management adaptability: (1) maintain and update team memory, (2) dedicate resources for environmental scanning, (3) maximize the number of alternative information sources and (4) integrate forecasting and back casting methods in the information management process.

Man vs. machine

Deception is an inevitable component of human interaction. Researchers and practitioners are developing information systems to aid in the detection of deceptive communication. Information systems are typically adopted by end users to aid in completing a goal or objective (e.g., increasing the efficiency of a business process). However, end-user interactions with deception detection systems (adversarial systems) are unique because the goals of the system and the user are orthogonal. Prior work investigating systems-based deception detection has focused on the identification of reliable deception indicators. This research extends extant work by looking at how users of deception detection systems alter their behavior in response to the presence of guilty knowledge, relevant stimuli, and system knowledge. An analysis of data collected during two laboratory experiments reveals that guilty knowledge, relevant stimuli, and system knowledge all lead to increased use of countermeasures. The implications and limitations of this research are discussed and avenues for future research are outlined. We present adversarial systems as a novel/growing area of IS research.Knowledge of a deception systems operations increases countermeasure use.Presenting deceivers with relevant stimuli increases countermeasure use.Truth tellers use countermeasures when aware of the systems functionality.An extensive set of novel countermeasures is identified.

The influence of conversational agent embodiment and conversational relevance on socially desirable responding

Abstract Conversational agents (CAs) are becoming an increasingly common component in a wide range of information systems. A great deal of research to date has focused on enhancing traits that make CAs more humanlike. However, few studies have examined the influence such traits have on information disclosure. This research builds on self-disclosure, social desirability, and social presence theories to explain how CA anthropomorphism affects disclosure of personally sensitive information. Taken together, these theories suggest that as CAs become more humanlike, the social desirability of user responses will increase. In this study, we use a laboratory experiment to examine the influence of two elements of CA design—conversational relevance and embodiment—on the answers people give in response to sensitive and non-sensitive questions. We compare the responses given to various CAs to those given in a face-to-face interview and an online survey. The results show that for sensitive questions, CAs with better conversational abilities elicit more socially desirable responses from participants, with a less significant effect found for embodiment. These results suggest that for applications where eliciting honest answers to sensitive questions is important, CAs that are “better” in terms of humanlike realism may not be better for eliciting truthful responses to sensitive questions.

Mitigation of cognitive bias with a serious game: Two experiments testing feedback timing and source

Oneof thebenefitsofusingdigitalgames foreducation is thatgamescanprovide feedback for learnerstoassesstheirsituationandcorrecttheirmistakes.Weconductedtwostudiestoexaminethe effectivenessofdifferentfeedbackdesign(timing,duration,repeats,andfeedbacksource)inaserious gamedesignedtoteachlearnersaboutcognitivebiases.Wealsocomparedthedigitalgame-based learningconditiontoaprofessionaltrainingvideo.Overall,thedigitalgamewassignificantlymore effectivethanthevideocondition.Longerdurationsandrepeatsimprovetheeffectsonbias-mitigation. Surprisingly,therewasnosignificantdifferencebetweenjust-in-timefeedbackanddelayedfeedback, andcomputer-generatedfeedbackwasmoreeffectivethanfeedbackfromotherplayers.