| Several recent events, such as 9/11 and hurricane Katrina, have placed an increased emphasis on supporting decision-makers during disaster recovery missions. To this end advanced information technologies have been developed that provide real-time feedback to practitioners who have traditionally been forced to operate in a feed-forward control mode due to a lack of timely information about the system they are controlling. In particular, a number of new remote sensing technologies have been introduced that present remotely located decision-makers with real-time imaging of the area of interest. This new data channel promises several benefits to decision-makers at all levels of command and control. It can aid in assessing and comprehending the overall situation while also allowing planners to consider challenges faced by local operators (e.g., restricted access, potential dangers) when they are developing their response strategy. As the mission evolves, reviewing earlier and the most recent imaging data can serve as a tool for comparative analysis and for assessing the effectiveness of the current strategy. It can raise the decision-maker's awareness of constraints and possibilities, aid them in making informed decisions, and serve as an evaluation and re-planning tool. While potentially useful, past experience has shown that introducing technologies that change the type, amount, and nature of information available to a decision-maker can also have unintended and undesirable consequences, such as new data management tasks, new forms of errors, and a narrowing of data search activities. In some cases, data gathering activities that were traditionally knowledge, or expectation-driven, become primarily data-driven; the operator no longer actively frames the exploration but instead is guided by what the system provides. The goal of the present research was to determine the impact of real-time, image-based feedback on data gathering, integration, and analysis activities and the decision making process. It aimed to identify benefits as well as unintended and undesirable side effects that need to be addressed through design, training, and procedures to avoid breakdowns in system performance. Field observations and interviews were used to gain an understanding of the tasks, information needs, and challenges for the Incident Commander (IC) and other decision-makers in this domain. The knowledge gained from these activities was used to design a staged-world experiment. Eight experienced ICs took part in a 90-minute disaster scenario, which was based on an actual chemical release accident to study several important research questions. First, we examined how, when, and why the ICs accessed and used the new imaging data for situation assessment and planning purposes to determine if using this new data source adversely affected the use, interpretation, and integration of other currently available sources of data? Another important question concerned the risk for high-level decision makers to get lost in low-level, highly detailed information and use it to directly control activities they should be supervising. Results indicate that using this new data source adversely affected the search for, and analysis of, other sources of data. All but one of the ICs failed to detect important changes in the situation that were not captured in the imaging data and failed to corroborate it with other available data sources. It appears that the ICs placed an inappropriate level of trust in the information, which resulted in a narrowing of data search activities. For decision-makers with many years of experience "on site", the literal nature of the imaging was more familiar, and therefore garnered greater trust, than the pre-processed, filtered, and abstracted data sources normally provided. The timeliness of the feedback also appears to have contributed to the perception that the information was highly valuable. These results suggest that the observed breakdo... |
