Using modern advanced driving simulator for human factors research has many advantages such as experimental control, expense, safety, and ease of data collection. However, the literature describes some possible disadvantages, i.e. simulator sickness, inaccurate replication of physical sensations, and most importantly, validity, which is the extent to which human behavior observed in simulation conditions, can be generalized to real situations. In other words, whereas physical validity is improving, psychological and behavioral validity remains a difficult problem. Assessing the absolute validity of driving simulation would require the comparison of results obtained from studies conducted in a real situation and in a virtual environment. However, this comparison is expensive (instrumentation) and complex, if possible (strict control of all the events occurring in a real situation). One way to deal with this problem is to focus on relative behavioral validity, trying to evaluate to what extent human behavior observed in simulation conditions is qualitatively similar to a "real world" behavior. A correlative approach is to try to measure, by manipulating experimental conditions, to what extent this behavior is modified, with reference to a "real world" behavior. This is typically what the concept of presence, introduced in virtual reality research to describe psychological and behavioral effects of immersive virtual environments, tries to achieve.

This pilot study aims to find a way to measure ‘presence’ as a proxy for ecological validity in driving simulators. The underlying assumption is that a person experiencing a strong sense of presence in the virtual environment will react as if it were real. We measure ‘presence’ through the ‘attention’ given to the driving task. We hypothesize that the greater the attention given to the primary driving task, the more the subject will experience spatial presence. ‘Attention’ was varied by adding a second task and oncoming traffic; we then analyzed behavioral measures of driving performance and subjective ‘presence’. The main result is a lack of congruence between subjective and behavioral measures. Although behavioral differences were observed between the various experimental conditions, there was no significant difference in subjective measures of presence. One explanation for this result could be that in all experimental conditions the driving activity did not require high-level cognitive processes, and was instead based on bottom-up attentional processes. Many of the processes involved in driving seem to be automatic, and this study argues for the concomitant use of subjective measures (such as questionnaires) and objective measures to assess presence in driving simulators. Furthermore, the development of a sensitive measure of presence seems to require more challenging scenarios in terms of controlled attention, cognitive involvement and more specifically, the emotions induced by the media. Participants are clearly aware that they are not exposed to any physical danger when using the simulator and the problem of their motivation must be taken into consideration. Another major problem is to establish the extent to which they are absorbed in the simulated driving task. A significant challenge for future research is the emotional validity of driving.

In this paper, our ambition is to find a way of measuring “presence” to use it as a measure for ecological validity in driving simulators. The underlying assumption is that a person experiencing a strong sense of presence in the virtual environment will react in this environment as if it would be a real one. We propose to measure "presence" by measuring “attention” toward the driving task". Our objective is to demonstrate that the higher the subject's attention required by the primary driving task will be, the more the spatial presence will be felt. In the experiment we tried to vary "attention" by adding a dual task and by adding traffic and measure driving performance and subjective "presence". The main result is a lack of congruence between subjective and behavioral measures.

The main contribution of this paper to the Cognitive Ergonomics field is to propose a new approach of the behavioral validity's assessment of driving simulators. In this paper, our ambition is to find a way of measuring “presence” to use it as a measure for ecological validity in driving simulators. In this way, the purpose of this study is especially to investigate the relationship between subjective ratings of presence and physiological responses in virtual driving environments. Six driving scenarios were, thus, created, resulting from crossing two independent variables, i.e., the visual realism of the virtual environment (three virtual environments were created ranging from very realistic to not realistic) and the degree of visibility on the road for each virtual environment (two conditions were proposed : very good visibility and very bad visibility). Whatever the scenario, the driving task was always to follow a red car without losing sight of its location. To measure subjective presence, attention and emotional involvement, Presence, and Mood questionnaires were used. As measures of physiological response, heart rate and skin conductance level were acquired and more specifically the heart rate variability (HRV) was calculated from the acquired ECG data. Driving performance (standard deviations of speed and lateral position) was also analyzed. Fourteen participants (25.4±4.5 years of age) were placed in a fixed base driving simulator and performed consecutively the six driving scenarios in a different order. The results show correlations between Presence Questionnaire scores and both HRV parameters (time and frequency domains) and skin conductance level. In the study, HRV parameters and skin conductance level appeared to indicate the participant's degree of presence.

The development of direct-injection (DI) engines opened t he way to even more fuel-efficient and cleaner engines. However, their optimization requires precise unde rstanding of phenomena involved in the injection and atomization processes. The micronozzle internal structure plays a key role by favouring cavitation inside the micronozzle, which highly influe nce the fuel spray development. In the present study, X-ray microtomography was used to determine the 3D geometry of two micronozzles, one being cavitant and the other non-cavitant. Experiments were performed using a microfocus X-ray source coupled to a CMOS flat-panel detector. Two-dimensional images were acquired at different 360 angles and reconstructed using a 2D fan-beam reconstruction algori thm. The micronozzle surface was extracted from reconstructed images after segmentation. The obtaine d spatial resolution allowed to estimate micronozzle diameter values with a standard deviation o f 5 µm, but was too poor was not to estimate accurately the curvature radius of the orifices. Micro metric imaging techniques, such as propagationbased phase-enhanced imaging or local tomography, should be investigated.