![Abbildung 15: Die Abbildung zeigt die mittleren Reaktionszeiten für auditiv sowie visuell präsentierten Aufgaben getrennt nach den drei Schwierigkeiten einfach, mittel und schwer [vertikale Balken zeigen +/-1 Standardabweichung].](https://www.researchgate.net/profile/Maximilian-Schwalm/publication/37368096/figure/fig3/AS:669520410705928@1536637513381/Abbildung-15-Die-Abbildung-zeigt-die-mittleren-Reaktionszeiten-fuer-auditiv-sowie-visuell_Q320.jpg)
![Abbildung 20: Die Abbildung zeigt die mittleren Abweichungen (in Metern) in der Fahraufgabe für die drei experimentellen Bedingungen [vertikale Balken zeigen +/-1 Standardabweichung].](https://www.researchgate.net/profile/Maximilian-Schwalm/publication/37368096/figure/fig4/AS:669520410730517@1536637513685/Abbildung-20-Die-Abbildung-zeigt-die-mittleren-Abweichungen-in-Metern-in-der_Q320.jpg)
![Abbildung 21: Die Abbildung zeigt den mittleren Index of Cognitive Activity (ICA) für die Bedingungen LCT visuell, LCT auditiv sowie LCT plus SURT [vertikale Balken zeigen +/-1 Standardabweichung].](https://www.researchgate.net/profile/Maximilian-Schwalm/publication/37368096/figure/fig5/AS:669520410722314@1536637513740/Abbildung-21-Die-Abbildung-zeigt-den-mittleren-Index-of-Cognitive-Activity-ICA-fuer-die_Q320.jpg)
Content uploaded by Maximilian Schwalm
Author content
All content in this area was uploaded by Maximilian Schwalm on May 16, 2014
Content may be subject to copyright.
A preview of the PDF is not available
Pupillometrie als Methode zur Erfassung mentaler Beanspruchungen im automotiven Kontext
Abstract and Figures
Die vorliegende Arbeit untersucht die Anwendung einer neuen Methode der Pupillometrie zur Erfassung von mentaler Beanspruchung bei der Optimierung von Fahrerinformations- und Fahrerassistenzsystemen im automotiven Kontext. Die in der vorliegenden Arbeit eingesetzte Methode erlaubt es, über charakteristische Veränderungen der Größe der menschlichen Pupille Rückschlüsse auf die aktuelle Beanspruchung eines Fahrers zu ziehen und diese mittels eines Index, dem "Index of Cognitive Activity" (ICA, Marshall, Davis & Knust, 2004) abzubilden. Es werden insgesamt vier Experimente beschrieben. Es zeigt sich, dass der hier beschriebene Index sowohl in einem experimentell streng kontrollierten Kontext, als auch in realistischen, anwendungsnahen Fahr- und Bediensituationen auftretende mentale Beanspruchungen für den Fahrer abbilden kann. Es konnte auch gezeigt werden, dass es mittels der hier beschriebenen Methode der Pupillometrie und deren hoher zeitlicher Auflösung der Messung erstmals möglich ist, den tatsächlichen Verlauf von Beanspruchungen bei der Bearbeitung einer Aufgabe abzubilden. Die hier beschriebenen Ergebnisse zeigen eine Möglichkeit der Optimierung zukünftiger Mensch-Maschine Schnittstellen hinsichtlich der für den Benutzer auftretenden Beanspruchungen mit einer bis lang unerreichten Genauigkeit.
Figures - uploaded by Maximilian Schwalm
Author content
All figure content in this area was uploaded by Maximilian Schwalm
Content may be subject to copyright.
... Additionally, the ICA method has been used in a number of studies and has shown to be a promising method of measuring driver's mental workload. Schwalm was one of the first to show that the ICA may be a reliable method to identify driver's mental demands in both highly controlled driving simulators as well as in real environments [52]. Demberg et al. used the ICA method to estimate mental workload during simulated driving and found increased ICA values resulting from more difficult speech comprehension clauses [53]. ...
... They suggested that this new measure might be particularly useful when rapid changes in the mental workload have to be detected [29]. The index of cognitive activity (ICA; patented by [23]) is a promising method that can be used to identify drivers' mental workload in both highly controlled driving simulators and in real environments due to its ability to account for illumination effects [52][53][54], [57]. This index is calculated through an algorithm that uses pupillometry, but is unfortunately patented and therefore cannot easily be used in research. ...
The assessment of mental workload could be helpful to road safety especially if developments of vehicle automation will increasingly place drivers into roles of supervisory control. With the rapidly decreasing size and increasing resolution of cameras as well as exponential computational power gains, remote eye measurements are growing in popularity as non-obtrusive and non-distracting tools for assessing driver workload. This review summarizes literature on the relation between eye measurement parameters and drivers’ mental workload. Various eye activity measures including blinks, fixations, and saccades have previously researched and confirmed as useful estimates of a driver's mental workload. Additionally, recent studies in pupillometry have shown promise for real-time prediction and assessment of driver mental workload after effects of illumination are accounted for. Specifically, workload increases were found to be indicated by increases in blink latency, PERCLOS, fixation duration, pupil dilation, and ICA; by decreases in blink duration and gaze variability; and with mixed results regarding blink rate. Given such a range of measures available, we recommend using multiple assessment methods to increase validity and robustness in driver assessment.
... Additionally, the ICA method has been used in a number of studies and has shown to be a promising method of measuring driver's mental workload. Schwalm was one of the first to show that the ICA may be a reliable method to identify driver's mental demands in both highly controlled driving simulators as well as in real environments [52]. Demberg et al. used the ICA method to estimate mental workload during simulated driving and found increased ICA values resulting from more difficult speech comprehension clauses [53]. ...
... They suggested that this new measure might be particularly useful when rapid changes in the mental workload have to be detected [29]. The index of cognitive activity (ICA; patented by [23]) is a promising method that can be used to identify drivers' mental workload in both highly controlled driving simulators and in real environments due to its ability to account for illumination effects525354, [57]. This index is calculated through an algorithm that uses pupillometry, but is unfortunately patented and therefore cannot easily be used in research. ...
The assessment of mental workload could be helpful to road safety especially if developments of vehicle automation will increasingly place drivers into roles of supervisory control. With the rapidly decreasing size and increasing resolution of cameras as well as exponential computational power gains, remote eye measurements are growing in popularity as non-obtrusive and non-distracting tools for assessing driver workload. This review summarizes literature on the relation between eye measurement parameters and drivers’ mental workload. Various eye activity measures including blinks, fixations, and saccades have previously researched and confirmed as useful estimates of a driver’s mental workload. Additionally, recent studies in pupillometry have shown promise for real-time prediction and assessment of driver mental workload after effects of illumination are accounted for. Specifically, workload increases were found to be indicated by increases in blink latency, PERCLOS, fixation duration, pupil dilation, and ICA; by decreases in blink duration and gaze variability; and with mixed results regarding blink rate. Given such a range of measures available, we recommend using multiple assessment methods to increase validity and robustness in driver assessment.
... Here, the temporal resolution is much higher than in other measurements. The ICA has been used and approved in an automotive context with basic driving tasks (Schwalm, Keinath & Zimmer, 2008; Schwalm, 2009) and was chosen as a method for the present study. ...
... Nevertheless, some methodical downsides can be deduced. The prompts for lane changes occur in nearly periodical time spans (7-9 seconds), so that participants can precisely anticipate the next lane change and adapt their behavior to it (Schwalm, 2009). This is not typical for real driving tasks in which demands occur erratically. ...
Drivers usually perform a range of different activities while driving. Following a classical workload approach, additional activities are expected to increase the demand on the driver. Nevertheless, drivers can usually manage even demanding situations successfully. They seem to be able to compensate demands by behavior adaptations, mainly in the following factors: in the driving task itself, in an additional (secondary) task and in their mental workload. It is suggested that by analyzing these three factors in temporal coherence, compensative interactions between them become measurable. Additionally, a reduction of activity in the secondary task is expected to be influenced by the characteristics of this task. To analyze these effects, a driving simulator study with 33 participants was accomplished. It could be shown that if a secondary task can be interrupted without a perceived decline in performance, it is interrupted in demanding driving situations. If an interruption causes a perceived performance loss, efforts are increased, and so the workload is heightened (measured with a high resolution physiological measurement based on pupillometry). Thus, drivers compensate their current demands by behavior adaptations in different factors, depending on the characteristics of a secondary task.
Mobiles Eye-Tracking ist als Forschungsmethode beliebter denn je und gewinnt in unterschiedlichen Feldern der angewandten Wissenschaften mehr und mehr an Bedeutung. Dieser Beitrag diskutiert, wie die Aufzeichnung und Analyse von Blickbewegungen in der Mobilität, im Usability Engineering, den Sportwissenschaften, der Augmented Reality/Mixed Reality/Virtual Reality und der Medizin bzw. medizinischen Weiterbildung eingesetzt wird. Der Beitrag gliedert sich dabei in drei Teile: in einem ersten Teil werden Grundzüge des Eye-Trackings erläutert; in einem zweiten Teil wird der Einsatz mobilen Eye-Trackings in ausgewählten Feldern der angewandten Wissenschaften veranschaulicht; und in einem abschließenden dritten Teil werden Potentiale und Risiken sowie zukünftige Forschungslinien skizziert, um die Anwendung mobilen Eye-Trackings als digitale Forschungsmethode weiter zu etablieren.
There are a lot of methods and instruments trying to assess mental workload reliable and independent from specific use cases or situations. Each method has its own strengths and weaknesses depending on the situation and the on individual person. Attempting to improve each approach separately would indeed allow a further context of use, but do not lead to a global reliability growth. Within this work a new approach is suggested by combining various ocular parameters adapting the individual person. The results of this study show clearly the advantages in mental workload assessment of an individual adapted set of variables in contrast to one variable trying to fit a collectivity of people. This new preparation considering six physiological ocular variables in two different contexts of use can be one step further to a more reliable and global way to assess mental workload.
This following chapter deals with driver condition detection. After delineating the factors relevant to detecting a driver’s condition and discussing the reasons for addressing the subject in terms of accident risk and the corresponding potentials and challenges (Sect. 1), three potential uses of driver condition detection are examined: detection of inattentiveness (Sect. 2), detection of drowsiness (Sect. 3), and detection of medical emergencies (e.g., a heart attack; Sect. 4). The respective driver conditions are defined, relevant measuring variables and their corresponding measuring procedures present, and selected applications expanded upon. Section 5 addresses driver condition monitoring systems currently available on the market and names the measuring variables and procedures used by those systems, before giving a short overview of the problem of potential false alarms in Sect. 6.
Der Fahrerzustand umfasst die zeitveränderlichen Eigenschaften des Fahrers, die für die Fahraufgabe relevant sein können. Da der Zustand des Fahrers intraindividuellen Schwankungen unterliegt, kann in Abhängigkeit des Veränderungszeitraums zwischen kurzfristig – innerhalb von Minuten oder Sekunden – und mittelfristig – innerhalb von Stunden bzw. Tagen – veränderlichen Faktoren, die den Fahrerzustand beeinflussen, unterschieden werden (in Anlehnung an [1]), z. B.:
Darüber hinaus haben auch die nicht oder nur langfristig veränderbaren Faktoren Auswirkungen auf den Fahrerzustand (beispielsweise die Konstitution oder die Persönlichkeit). Diese werden im Folgenden jedoch nicht weiter betrachtet (s. dazu Kap. 1). In den nachstehenden Kapiteln werden die Themen Müdigkeit, Aufmerksamkeit und medizinische Notfälle näher beschrieben.
While it has long been known that the pupil reacts to cognitive load, pupil size has received little attention in cognitive research because of its long latency and the difficulty of separating effects of cognitive load from the light reflex or effects due to eye movements. A novel measure, the Index of Cognitive Activity (ICA), relates cognitive effort to the frequency of small rapid dilations of the pupil. We report here on a total of seven experiments which test whether the ICA reliably indexes linguistically induced cognitive load: three experiments in reading (a manipulation of grammatical gender match / mismatch, an experiment of semantic fit, and an experiment comparing locally ambiguous subject versus object relative clauses, all in German), three dual-task experiments with simultaneous driving and spoken language comprehension (using the same manipulations as in the single-task reading experiments), and a visual world experiment comparing the processing of causal versus concessive discourse markers. These experiments are the first to investigate the effect and time course of the ICA in language processing. All of our experiments support the idea that the ICA indexes linguistic processing difficulty. The effects of our linguistic manipulations on the ICA are consistent for reading and auditory presentation. Furthermore, our experiments show that the ICA allows for usage within a multi-task paradigm. Its robustness with respect to eye movements means that it is a valid measure of processing difficulty for usage within the visual world paradigm, which will allow researchers to assess both visual attention and processing difficulty at the same time, using an eye-tracker. We argue that the ICA is indicative of activity in the locus caeruleus area of the brain stem, which has recently also been linked to P600 effects observed in psycholinguistic EEG experiments.
Eine vielversprechende Maßnahme dem deutlich erhöhten Verkehrsunfallrisiko, das Fahranfänger aufweisen (vgl. z.B. Maycock, Lockwood & Lester, 1991; Schade, 2001), entgegenzuwirken, besteht in einer Verbesserung der theoretischen Fahrerlaubnisprüfung. Insbesondere die Optimierung ihrer Validität erscheint relevant, da diese mit einer verbesserten Identifikation derjenigen Personen, die noch sicherheitsrelevante Defizite aufweisen, einhergeht. Das Ziel der vorliegenden Arbeit bestand darin, auf einer theoretisch und empirisch fundierten Grundlage Aufgabenformate zu entwickeln, die eine valide Trennung von Könnern und Nicht-Könnern ermöglichen. An solche validen Aufgaben werden die Anforderungen gestellt, dass sie zum einen sicherheitsrelevante Inhalte abprüfen und zum anderen eine adäquate Operationalisierung der gewählten Merkmale und Fähigkeiten darstellen. Die Gefahrenwahrnehmung hat sich in der Verkehrssicherheitsforschung als ein Aspekt der Fahrkompetenz erwiesen, der als relevant für das sichere Autofahren gelten kann (z.B. McKenna & Crick, 1994b). Wie man diese sicherheitskritische Fähigkeit angemessen überprüfen kann, muss empirisch untersucht werden. Um Vorhersagen darüber zu treffen, worin eine optimale Operationalisierung der Gefahrenwahrnehmung besteht, können aufgrund des Fehlens spezieller, empirisch gesicherter Modelle zur Fahrexpertise allgemeine Modelle zu Expertise und Expertiseerwerb herangezogen werden. Insbesondere das Konzept der Domänenspezifität erscheint hier zielführend. Expertise zeigt sich deutlicher, je mehr die zu bewältigende Aufgabe den natürlichen Anforderungen, die in einer Domäne bestehen, entspricht (Glaser & Chi, 1988).
In der vorliegenden Arbeit wurden verschiedene Aufgabenformate entwickelt, deren Eignung zur Erfassung von Fahrexpertise durch Experten-Novizen-Vergleiche im Quer- und Längsschnittdesign überprüft wurde. Diejenigen Aufgaben, bei denen deutliche Leistungsunterschiede zugunsten der Experten festgestellt wurden, können als krieriumsvalide Messungen von Fahrexpertise betrachtete werden. In zwei Vorstudien (N = 268) wurden mithilfe von 2x2-faktoriellen Versuchsdesigns Aufgabenformate getestet, die unterschiedliche Anforderungen an die Teilnehmer stellten (z.B. Mehrfachwahlaufgaben zum verkehrsspezifischen Faktenwissen, Reaktionszeitaufgaben). Neben dem Faktor Expertise (Fahrschüler vs. erfahrene Fahrer) wurde das Präsentationsformat des vorgegebenen Bildmaterials variiert (statisch vs. dynamisch). Um festzustellen, inwieweit die Formate, die sich in den Vorstudien bewährt hatten, auch die Leistungsentwicklung von Fahrschülern abbilden können, wurde eine Längsschnittuntersuchung mit drei Messzeitpunkten (N = 152) durchgeführt.
Insgesamt hat sich gezeigt, dass Gefahrenwahrnehmungsaufgaben im Reaktionszeitaufgabenfornat eine deutliche Trennung von Experten und Novizen ermöglichen, während Aufgaben zum verkehrsrelevanten Faktenwissen im Mehrfachwahlaufgabenformat nicht geeignet waren, um zwischen erfahrenen und unerfahrenen Fahrern zu differenzieren. Im Verlauf ihrer Fahrausbildung erreichten die Novizen bezüglich ihres verkehrsspezifischen Wissens das Leistungsniveau der Experten. In den Aufgaben, die im Reaktionszeitaufgabenformat vorgegeben wurden, konnten sich die Teilnehmer über die Zeit nicht steigern. Das Präsentationsformat hatte auf die Validität der Aufgaben keinen generellen Einfluss. Dynamische Präsentationen von Verkehrsszenarien haben die Novizen bei der Bearbeitung von Aufgaben zum deklarativen Faktenwissen jedoch unterstützt.
Vorhersagen, die auf der Grundlage des Konzeptes der Domänenspezifität aus der allgemeinen Expertiseforschung getroffen wurden, konnten bestätigt werden: Wissen und Fähigkeiten werden am besten auf einem anwendungsbezogenen Niveau überprüft. Für die Fahrerlaubnisprüfung bedeutet dies, dass durch den Einsatz handlungsnaher Aufgabenformate eine Validitätssteigerung ermöglicht wird.
The optimization of the theoretical driving test can be considered as a promising measure to reduce the high accident liability of novice drivers. Particularly, the improvement of the test’s validity can be regarded as attractive, because it should lead to a better identification of those applicants, who still lack important competencies. Therefore, the aim of the present studies was to create tasks that allow for a criterion valid separation of driving experts and driving novices. These tasks were supposed to include contents, which are relevant for driving safety and, moreover, represent an appropriate operationalization of the chosen important skills. Hazard Perception can be considered as one such skill (e.g., McKenna & Crick, 1994). In order to identify appropriate features of test items assessing this skill, empirical evidence is needed. Since empirically proven models of driving expertise are not yet provided, considerations about some findings of general expertise research serve as a basis for hypothesizing what kind of improvement can be expected from different hazard perception tasks. In particular, the concept of domain specificity seems relevant for this purpose; experts outperform novices more clearly with increasing domain specificity of the given task–it’s similarity to the typical demands of a certain domain (Glaser & Chi, 1988).
Within the present studies, different kinds of tasks were developed and tested with regard to their criterion validity in cross-sectional and longitudinal designs via expert-novice comparisons. In two cross-sectional preliminary studies (N = 268), various types of exercises (e.g., knowledge tasks, reaction tasks) were tested in 2 by 2 factorial designs, involving the two factors driving expertise (experts vs. novices) and presentation mode of the applied visual materials (static vs. animated). In order to determine to which extend the tasks, that had turned out to be criterion valid in the preliminary studies, were able to depict the learner drivers increase in performance during their instruction, a longitudinal study (N = 152) with three measurements was conducted.
Overall results show, that reaction time measures of hazard perception skills support the separation of expert and novice drivers. In contrast, those tasks, assessing driving specific knowledge via multiple choice items, did not lead to performance differences between experienced drivers and learner drivers. The presentation mode of the presented driving scenarios did not have any impact on the criterion validity of the tasks: none of the two modes was better qualified to differentiate between experts and novices. A supplantation effect of the dynamic materials could be found for the novices: they benefitted from animations. The longitudinal study revealed that during their instruction, learner drivers managed to increase their performance considerably regarding driving specific knowledge. However, they did not improve their accuracy and reaction time in identifying potential hazards within driving scenarios.
Predictions made on the basis of the concept of domain specificity were confirmed: skills and knowledge should be assessed with tasks, whose demands are similar to the characteristic demands of a defined domain. The practical implication for driving assessment is the following: the test’s validity can be enhanced by the introduction of tasks (e.g., reaction time measures) that are close to real driving.
reviews major categories of empirical workload measurement techniques and provides guidelines for the choice of appropriate assessment procedures for particular applications
sensitivity / diagnosticity
rating scales / psychometric techniques (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Spearman, in proposing his influential two-factor theory of intelligence, adopted an implicit biological model which was untestable in his time. Spearman suggested both a general factor of intelligence (g), which corresponded to the amount of “general mental energy” available to an individual for information processing, and a set of specific ability factors, which were brain systems or “mental engines” drawing upon the general energy pool (Spearman, 1904).