Working memory is a key executive function for flying an aircraft. This function is particularly critical when pilots have to recall series of air traffic control instructions. However, working memory limitations may jeopardize flight safety. Since the functional near-infrared spectroscopy (fNIRS) method seems promising for assessing working memory load, our objective is to implement an on-line fNIRS-based inference system that integrates two complementary estimators. The first estimator is a real-time state estimation MACD-based algorithm dedicated to identifying the pilot’s instantaneous mental state (not-on-task vs. on-task). It does not require a calibration process to perform its estimation. The second estimator is an on-line SVM-based classifier that is able to discriminate task difficulty (low working memory load vs. high working memory load). These two estimators were tested with 19 pilots who were placed in a realistic flight simulator and were asked to recall air traffic control instructions. We found that the estimated pilot’s mental state matched significantly better than chance with the pilot’s real state (62% global accuracy, 58% specificity, and 72% sensitivity). The second estimator, dedicated to assessing single trial working memory loads, led to 80% classification accuracy, 72% specificity, and 89% sensitivity. These two estimators establish reusable blocks for further fNIRS-based passive brain computer interface development.
A B S T R A C TWe investigated the possibility of developing a decision support system (DSS) that integrates eye-fixation measurements to better adapt its suggestions. Indeed, eye fixation give insight into human decision-making: Individuals tend to pay more attention to key information in line with their upcoming selection. Thus, eyefixation measures can help the DSS to better capture the context that determines user decisions. Twenty-two participants performed a simplified Air Traffic Control (ATC) simulation in which they had to decide to accept or to modify route suggestions according to specific parameter values displayed on the screen. Decisions and fixation times on each parameter were recorded. The user fixation times were used by an algorithm to estimate the utility of each parameter for its decision. Immediately after this training phase, the algorithm generated new route suggestions under two conditions: 1) Taking into account the participant's decisions, 2) Taking into account the participant's decisions plus their visual behavior using the measurements of dwell times on displayed parameters. Results showed that system suggestions were more accurate than the base system when taking into account the participant's decisions, and even more accurate using their dwell times. Capturing the crucial information for the decision using the eye tracker accelerated the DSS learning phase, and thus helped to further enhance the accuracy of consecutive suggestions. Moreover, exploratory eye-tracking analysis reflected two different stages of the decision-making process, with longer dwell times on relevant parameters (i.e. involved in a rule) during the entire decision time course, and frequency of fixations on these relevant parameters that increased, especially during the last fixations prior to the decision. Consequently, future DSS integrating eyetracking data should pay specific care to the final fixations prior to the decision. In general, our results emphasize the potential interest of eye-tracking to enhance and accelerate system adaptation to user preference, knowledge, and expertise.
In this paper, we describe the outcomes of the challenge organized and run by Airbus and partners in 2018. The challenge consisted of two tasks applied to Air Traffic Control (ATC) speech in English: 1) automatic speech-to-text transcription, 2) call sign detection (CSD). The registered participants were provided with 40 hours of speech along with manual transcriptions. Twenty-two teams submitted predictions on a five hour evaluation set. ATC speech processing is challenging for several reasons: high speech rate, foreign-accented speech with a great diversity of accents, noisy communication channels. The best ranked team achieved a 7.62% Word Error Rate and a 82.41% CSD F1-score. Transcribing pilots' speech was found to be twice as harder as controllers' speech. Remaining issues towards solving ATC ASR are also discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.