A Prototype Toolset for Rapidly Reducing and Reviewing Driving Simulator Data

In our experience, one of the most challenging aspects of conducting research using driving simulators is the postprocessing of simulator data, a process often referred to as data reduction. This step is necessary to produce various numeric measures that in turn are used for statistical analysis. Although commercial software packages can often be used to facilitate the process, such tools are not explicitly aware of the nature of the data and the intricacies of the scenario used to produce it. This knowledge is often necessary to effectively manage the data. These packages also have limited usability for performing simulator-specific calculations (i.e., lane deviation) or linking data based on associations between research participants or treatments. This paper describes an integrated toolset that has been designed explicitly to handle data reduction calculations for driving simulator or similar data. The toolset is tightly integrated with other simulator tools and can intelligently handle binary as well as video, audio, questionnaire, and other data typically collected during simulator-based experiments. The toolset is explicitly aware of the nature of each data file, its contents, and its association with other data files. A prototype of the tool has been constructed and is presented as part of the paper, along with some technical challenges and how they were addressed. INTRODUCTION AND MOTIVATION When developing technology associated with driving simulation applications, a lot of emphasis is placed on the raw computing capacity of the hardware or on the cueing capabilities of a simulator necessary to conduct various research studies. There is significantly less work contributing to the ability of a researcher to process, manipulate, and analyze the data after an experiment has been completed. This post processing step, also referred to as data reduction, involves the computation of quantitative variables that can be used to perform statistical analysis on primary and secondary variables of a research study. The difference between data reduction and statistical analysis is