Organizational Unit:
College of Sciences
College of Sciences
Permanent Link
Research Organization Registry ID
Description
Previous Names
Parent Organization
Parent Organization
Includes Organization(s)
Organizational Unit
Organizational Unit
Organizational Unit
Organizational Unit
Organizational Unit
ArchiveSpace Name Record
Publication Search Results
Now showing
1 - 5 of 5
-
ItemSupporting Feedback Loop Reasoning in Simulated Systems with Computer-Based Scaffolding(Georgia Institute of Technology, 2023-04-12) Dunbar, TerriFeedback loops are a critical part of systems and a frequent source of misconceptions. These misconceptions are thought to occur because people inappropriately apply their everyday experiences of causality to the types of causal feedback loops present in systems. Feedback loop reasoning can improve with training; however, misconceptions such as failing to close the loop are particularly resistant to change. Two experiments investigated whether factors known to improve positive transfer with other cognitive skills could overcome learners’ misconceptions about feedback loops during simulation training, including learning from multiple examples, similarity to the training context, scaffolding, and desirable difficulties. Results revealed that similarity and potentially cognitive load had the largest impacts on transfer, and the type of scaffolding used or how it was sequenced over training had little effect. Near transfer only occurred for participants who learned from balance systems where the goal is to maintain system equilibrium by counterbalancing relationships, and not with pattern systems where the goal is to determine how spatial patterns emerge from local interactions. There was no evidence of far transfer. Across both experiments, participants also closed the loop more frequently when learning from balance systems. Overall, the current studies suggest that researchers need to carefully consider the type of system used during simulation training because subtle manipulations can lead to different learning experiences. Existing theories of system misconceptions are unable to satisfactorily explain why these performance differences occurred. Instead, the results and their implications are discussed using cognitive load theory.
-
ItemCommunication Pattern Analysis in Human-Autonomy Teaming(Georgia Institute of Technology, 2022-07-20) Zhou, ShiwenCommunication is critical to team coordination and interaction because it provides information flows allowing a team to build team cognition, which contributes to overall team performance. In recent years, autonomous (AI) team members are beginning to be considered as effective substitutes for human teammates. However, research has shown that AI team members may lack the communication skills that are required for effective team performance (McNeese et al., 2018). To better understand which aspects of communication an AI team member performs differently compared to a human team member, and how they impact team performance, the current study analyzes communication features of three-person teams that include all human teams and human-AI teams operating in a remotely piloted aircraft system (RPAS). The current study analyzed communication pattern predictability (communication determinism) and transition probabilities to measure communication flow and Latent Semantic Analysis (LSA) to measure communication content. The current study found that both communication flow and content distinguished communication in all-human teams from communication in human-AI teams and found that these communication flow and content features predicted team performance in all-human versus human-AI teams. In this way, the current study hopes these communication differences can provide feedback and suggestions to future adoption of AI as a teammate in team training and team operations.
-
ItemDeveloping Objective Communication-based Measures of Trust for Human-Autonomy Teams(Georgia Institute of Technology, 2022-07-19) Scalia, Matthew J.As artificial intelligence capabilities have improved, humans have begun teaming with autonomous agents that have the capability to communicate and make intelligent decisions that are adaptable to task situations. Trust is a core component of human-human and human-autonomy team (HAT) interaction. As with all-human teams, the amount of trust held within a HAT will impact the team’s ability to perform effectively and achieve its goals. A recent theoretical framework, distributed dynamic team trust (D2T2; Huang et al., 2021), relates trust, team interaction measures, and team performance in HATs and has called for interaction-based measures of trust that go beyond traditional questionnaire-based approaches to measure the dynamics of trust in real-time. In this research, these relationships are examined by investigating HAT interaction communication-based measures (ICM; amount, frequency, affect, and “pushing” vs. “pulling” of information between team members) as a mechanism for D2T2 and tested for predictive validity using questionnaire-based trust measures as well as team performance in a three-team member remotely-piloted aerial system (RPAS) HAT synthetic task. Results suggest that ICM can be used as a measure for team performance in real-time. Specifically, ICM was a significant predictor of team performance and not trust, and trust was not a significant predictor of team performance. Exploratory factor analyses of the trust questionnaire items discovered clear differences in how human teammates characterize trust in all-human teams and HATs. Specifically for HATs, interpersonal and technical factors associated with trust in autonomous agents were found as a result of dynamic exposure to the autonomous agent by distinct stakeholders through communication. These findings confirmed the underlying theory behind D2T2 as a framework that includes both interpersonal and technical factors related to trust in HAT along a dynamic timeline among different types of stakeholders. The findings provide some insight into the dynamic nature of trust, but continued research to discover interactive measures of trust is necessary.
-
ItemCHARACTERISTIC LAG & THE INTERMANUAL SPEED ADVANTAGE(Georgia Institute of Technology, 2022-01-24) Harrison, JuliaPrevious research has found evidence for the intermanual speed advantage, wherein novice actors perform a visually-guided, two-handed task faster with one hand from each member of a dyad (i.e., intermanually) compared to when one actor completes the task with their own two hands (i.e., bimanually). The intermanual speed advantage is reversed or erased, however, after the task has been well-practiced by both actors bimanually. Furthermore, visuomotor coupling (i.e., coupling between eye and hand movements) has been found to underlie the presence of the intermanual speed advantage in novices and its erasure in experienced actors. This is due to a reduced reliance on visual input as the execution of the manual task becomes more fluent. Using secondary data, the present study seeks to further investigate how visuomotor coupling changes as a function of previous bimanual practice. This is done through a characteristic lag analysis, a dynamical systems metric that assesses how close in time and space the gaze and hands are while actors complete a simulated laparoscopic cutting task. Results suggest that the individual visuomotor coordination of the component actors impacts the execution of the task by the dyad in the intermanual condition, and that this change in coordination depends on previous bimanual practice. Specifically, findings show that the lag between the gaze and the hands of novice actors entrains to the partner with the longer lag (i.e., the less coupled partner) when acting in the intermanual trials. However, in experienced actors with previous bimanual practice, the dyad entrains to the actor with the shorter lag (i.e., the more coupled partner) imposing a ceiling on the dyad in intermanual trails and preventing them from uncoupling further. This pattern of results demonstrates how changes in visuomotor coupling lag help account for the erasure of the intermanual speed advantage after previous bimanual practice.
-
ItemDynamical Analysis and Modeling of Team Resilience in Human-Autonomy Teams(Georgia Institute of Technology, 2020-12-07) Grimm, David A.P.A resilient team would be proficient at overcoming sudden, unexpected changes by displaying a rapid, adaptive response to maintain effectiveness. To quantify resilience, I analyzed data from two different experiments examining performance of human-autonomy teams (HATs) operating in a remotely piloted aircraft system (RPAS). Across both experiments, the HATs experienced a variety of automation and autonomy failure perturbations using a Wizard of Oz paradigm. Team performance was measured by the successful completion of simulated reconnaissance missions, a mission level team performance score, a coordination-based target processing efficiency (TPE) score to quantify team efficiency, and a ground truth resilience score (GTRS) to measure how teams performed during and following a failure. Different layers, composed of vehicle, operator controls, communication, and overall system layers, of sociotechnical elements of the system were measured across RPAS missions. To measure resilience, I used entropy and a root mean squared error (RMSE) metric across all system layers. I used these measures to examine the time taken to achieve extreme values of reorganization during a failure and the novelty of the reorganization, respectively, to quantify resilience. I hypothesized that faster times to achieve extreme values of reorganization during a failure would be correlated with all performance measures. Across both experiments, I found negative correlations of time taken to achieve extreme values of reorganization and novelty of reorganization with team performance measured using TPE, and positive correlations while using GTRS. Additionally, I found that teams displayed more reorganization in response to failures, but this was not pronounced for effective teams. In Experiment 2, I also found differential effects of training in the communication and control layers. These results can help inform the measurement and training of resilience in HATs through targeted team training, feedback, and real-time analysis applications.