Processed Data

<p dir="ltr">Anonymized processed data from the <b>Non-Emergency Notification Timing in Autonomous Vehicles</b> study. This dataset depository is part of the <b>Non-Emergency Notification Timing in Autonomous Vehicles</b> artifact collection: <a href="https://doi.org/10.1184/R1/c.7894613" rel="noreferrer" target="_blank">10.1184/R1/c.7894613</a>.</p><p dir="ltr">The goal of this study is to investigate the best moment to send AV occupants in non-emergency notifications when they are engaged in non-driving-related tasks (NDRTs).</p><p dir="ltr">This archive contains data collected through a series of user study sessions in an autonomous vehicle simulator. On four different days, the participants experienced a simulated AV commute and completed four different non-driving-related tasks (NDRTs): Task of Choice (denoted as Scenario 1), Gaming (S2), Video Watching (S3), and Reading (S4). For the latter three scenarios, the tasks were completed on an Android tablet. The order of the latter three scenarios was also randomized.</p><p dir="ltr">While completing NDRTs, participants were instructed to respond to an audio signal with gradually increasing volume. Participants would say "yes" or "good" if they thought they were available for non-emergency notifications upon hearing the signal; they would say "no" or "bad" otherwise. In this way, the participants provided labels for the data stream around the moment when the signal went off.</p><p dir="ltr"><b>For more details about the content of this dataset, please refer to </b><code><strong>README.txt</strong></code><b>. </b>A video demonstration of the procedure can be found at <a href="https://doi.org/10.1184/R1/29396957" target="_blank">10.1184/R1/29396957</a>. The video also presents multiple samples of our complete video stream data. Supplementary figures showcasing the data collection setup can be found at <a href="https://doi.org/10.1184/R1/29372027" rel="noreferrer" target="_blank">10.1184/R1/29372027</a>.</p><p>---</p><p dir="ltr">Due to certain terms in our consent form, at this moment, we are unable to publicly share video data containing identifiable information of the participants -- <b>video streams containing participants' faces and their own devices are excluded from this archive</b>. The video demonstration mentioned above contains sample video data from all available camera angles.</p><p dir="ltr"><br></p><p dir="ltr">This public dataset contains the following for each participant in each session:</p><p dir="ltr">(note: files with names in brackets [*] are not accessible to the public; many files have "before_" or "after_" suffixes, which means the file spans [t-20s, t] or [t, t+20s], respectively; t is the time of each signal onset)</p><p><br></p><ul><li><code>all_event_log.xlsx</code> combined event logs recording labels and detection, response, reaction, and decision time measurements (formatted, reorganized, and plotted)</li><li><code>all_event_log.csv</code> combined event logs recording labels and time measurements</li><li><code>sessions_order.csv</code> order of scenarios each participant experience</li><li><code>event_log.csv</code> event logs recording labels and time measurements of the session</li></ul><ul><li><code>[*_composite_*_*.m4v]</code> a video composite of all video feeds in each signal_* folder</li><li><code>_good</code> (or <code>_bad</code>) an empty file whose name denotes the label provided by the participant</li><li><code>*_car.csv</code> simulated vehicle data stream</li><li><code>*_gaze.csv</code> participant gaze data stream</li><li><code>*_mems.csv</code> participant head movement data stream (from eye tracker)</li><li><code>[*_cams.m4v]</code> side and rear camera video streams</li><li><code>[*_rgb.m4v]</code> front-facing camera video stream</li><li><code>*_disp.m4v</code> simulation screen recording video stream</li><li><code>*_gaze.m4v</code> eye tracker first-person-view video stream (This file may not be accessible for S1 for privacy reasons.)</li><li><code>*_gaze_p.m4v</code> eye tracker first-person-view video stream with gaze position overlay (This file may not be accessible for S1 for privacy reasons.)</li><li><code>*_tab_touch.csv</code> tablet touch data stream (Not in S1)</li><li><code>*_tab_accel.csv</code> tablet accelerometer data stream (Not in S1)</li></ul><p><br></p>