RESUMEN
OBJECTIVES: The purpose of this study was to refine a commercial car driving simulation for occupational research. As the effects of ethanol on driving behavior are well established, we choose alcohol as a test compound to investigate the performance of subjects during simulation. MATERIALS AND METHODS: We programmed a night driving scenario consisting of monotonous highway and a rural road on a Foerst F10-P driving simulator. Twenty healthy men, 19-30 years, participated in a pilot study. Subjects were screened for simulator sickness, followed by training on the simulator one hour in total. Experiments were performed in the morning on a separate day. Participants were randomized into either an alcoholized or a control group. All subjects drove two courses consisting of highway and rural road and were sober for the first course. During a 1 h break the ethanol group drank an alcoholic beverage to yield 0.06% blood alcohol concentration (BAC). Generalized linear mixed models were used to analyze the influence of alcohol on driving performance. Among others, independent variables were Simulator Sickness Questionnaire scores and subjective sleepiness. RESULTS: Subjects did not experience simulator sickness during the experiments. Mean BAC before the second test drive was 0.065% in the mildly intoxicated group. There was no clear-cut difference in the number of crashes between 2 groups. BAC of 0.1% would deteriorate mean braking reaction time by 237 ms (SE = 112, p < 0.05). Ethanol slightly impaired the tracking in the right-hand curves (p = 0.058). Braking reaction time improved by 86 ms (SE = 36, p < 0.05) for the second test drive, indicating a learning effect. CONCLUSIONS: In sum, a clear ethanol effect was observed in the driving simulation. This simulation seems suitable for occupational research and produces little simulator sickness. Controlling for possible learning effects is recommended in driving simulation studies.
Asunto(s)
Intoxicación Alcohólica/psicología , Conducción de Automóvil/psicología , Investigación Biomédica/instrumentación , Simulación por Computador , Medicina del Trabajo , Análisis y Desempeño de Tareas , Adulto , Humanos , Aprendizaje , Masculino , Proyectos Piloto , Tiempo de Reacción , Adulto JovenRESUMEN
This article describes a general framework for detecting sleepiness states on the basis of prosody, articulation, and speech-quality-related speech characteristics. The advantages of this automatic real-time approach are that obtaining speech data is nonobstrusive and is free from sensor application and calibration efforts. Different types of acoustic features derived from speech, speaker, and emotion recognition were employed (frame-level-based speech features). Combing these features with high-level contour descriptors, which capture the temporal information of frame-level descriptor contours, results in 45,088 features per speech sample. In general, the measurement process follows the speech-adapted steps of pattern recognition: (1) recording speech, (2) preprocessing, (3) feature computation (using perceptual and signal-processing-related features such as, e.g., fundamental frequency, intensity, pause patterns, formants, and cepstral coefficients), (4) dimensionality reduction, (5) classification, and (6) evaluation. After a correlation-filter-based feature subset selection employed on the feature space in order to find most relevant features, different classification models were trained. The best model-namely, the support-vector machine-achieved 86.1% classification accuracy in predicting sleepiness in a sleep deprivation study (two-class problem, N=12; 01.00-08.00 a.m.).