Psycho physiological measurements are primarily used for measurement of four human factors variables:

 The methods are general, but they can be used for assessment purposes as well. They arc particular well suited for evaluation of the effect of human factors training or Crew Resource Management training.

 The disadvantage compared to for example instructor rating, observation of communication and questionnaire-based methods is, that psycho physiological methods require advanced and often expensive equipment. But the cost of the psycho physiological equipment is, when it is compared to the total cost of a full mission simulator, insignificant.

 The most important available techniques are:

　

 See [8] for further details on psycho physiological measurement.

　

8.1 Galvanic Skin Response (GSR)

　

 Galvanic Skin Response is an electrical measure of sweat (perspiration), usually measured with electrodes in the hand. The signals from the electrodes are amplified and the results are recorded electronically. The underlying assumption is that stress will induce activity of sweat glands, and that the sweat response to stress is very immediate. This technique is well known from "lie detectors".

　

 If we assume, that inexperience, poor workload management and poor Crew Resource Management skills are likely to cause stress, a test of stress could be used as assessment technique for Crew Resource Management programmes. We can conclude, that the training has served its purpose if and only if the result obtained by the methodology shows a significant decrease in level of stress from initial to concluding test.

　

8.2 Heart Rate Variability (HRV), electrocardiography (EKG/ECG)

　

 Heart Rate Variability is an electrical measure done with electrodes placed on limbs and/or torso. The signals from the electrodes are amplified and the results are recorded electronically. The so-called 0.1 Hz component or band is especially sensitive to workload (see [9], p. 28) and is often used as a direct measure of workload. One example is the measurement of maritime pilots approaching the harbour of Rotterdam done by Westrenen [9].

　

 Psycho physiological measurements of Heart Rate Variability (the 0.1 Hz band), can alone or together with subjective questionnaire based techniques (like the NASA-TLX methodology) be used for measurement of mental workload, and are therefore well suited as instruments for assessment of the effect of a simulator based Crew Resource Management programme. We can conclude, that the training has served its purpose if and only if the result obtained by the methodology shows a significant decrease in measured mental workload from initial to concluding test due to improved workload management skills.

　

8.3 Ocular activity, blink rate and eye movement detection

　

 Ocular activity, blink rate and eye movement are usually measured with equipment based on video recording of the position of the pupil. The video signals are analysed and interpreted automatically by a computer, and the results are presented through a user interface. This type of measurement can be used for evaluation of fatigue [10], but it can also be used for evaluation and quantification of visual attention.

　

 It is, from the output of eye movement detection equipment possible to directly determine exactly what the subject is gazing at e.g. which instrument or which feature in the surrounding environment outside the wheelhouse. This measure of visual attention can - if it is compared to and analysed and evaluated against a certain required performance standard - be used as an indication of crew performance. Used as an assessment technique an increase in performance indicated by improved visual attention should be observed from the initial to the concluding observation.

　

 If we assume, that inexperience, poor workload management and poor Crew Resource Management skills are likely to cause fatigue, a test of fatigue could be used as assessment technique for Crew Resource Management programmes. Fatigue can be measured with a variety of psycho physiological tests e.g. ocular measurements (as just described) or EEG [10].

　

 We can conclude, that the training has served its purpose if and only if the result obtained by the methodology shows a significant decrease in exposure to fatigue from initial to concluding test. However, the problem about using measurements of fatigue is, that the influence of confounding variables such as time of day, lack of sleep and jetlag is particular large compared to other human factor variables e.g. workload and stress. If the measure is used, one must be sure that the underlying, residual level of fatigue is the same at the beginning of both (initial and concluding) tests, that the time of the day the tests are performed is exactly the same and that the duration of the test is exactly the same. Further, the tests must be at exactly the same level of difficulty. These countermeasures will neutralise the effect of confounding variables. It will, however, always be a problem, that it is impossible to control the sleep quality of the participants. Even though they have gained the same amount of sleep in terms of hours slept, the quality of this sleep could be very different due to a number of influencing factors.

　

 The schematic diagram in figure 9 shows the level of fatigue at the start of a simulator voyage and again in the end of the simulator voyage for both initial and concluding measure. We can conclude, that there is an improvement from initial to concluding test when, and only when the level of fatigue obtained during the voyage is significantly lower in the concluding test than in the initial test.

　

　

8.4 Electric activity of the brain, electroencephalography (EEG)

　

 Measurement of electric activity of the brain, electroencephalogram (EEG), can be used for evaluation of a variety of psycho physiological and psychological variables or parameters. As just mentioned, EEG can be used for measurement of fatigue [10], but it can also be used for measurement of overall level of vigilance or alertness, level of psychological or cognitive activity or level of stress, See [8] and [11] for further details on measurement of EEG.

　

 EEG is measured by means of electrodes placed directly on the skin on the scull after certain predefined principles and standards e.g. the international 10-20 electrode system (see [8], p. 82-84).

　

 The signals from the electrodes are amplified and the results are recorded electronically by means of analog or digital equipment. An example of an amplifier (The Nervus Ul6 Amplifier) is shown in figure 10, and example of recording equipment is shown in figure 11 and 12.

　

 The results are interpreted with respect to the state of-the-art knowledge about how levels of stress and mental activity are reflected in different patters of EEG data (see for example [11]).

　

　

 If we - again - assume, that inexperience, poor workload management and poor Crew Resource Management skills are likely to cause stress, an EEG based test of level of stress could be used as assessment technique for Crew Resource Management programmes. We can conclude, that the training has served its purpose if and only if the result obtained by the methodology shows a decrease in level of stress from initial to concluding test, It is to eliminate the effect of confounding variables important to ensure, that the tests used initially and concluding both are at exactly the same level of difficulty and therefore would be expected to generate exactly the same level of stress if no intervention in the form of training had been given.

　

 It is at the moment of the writing of this paper (April 2003) not possible to find any examples in the literature of application of EEG techniques for measurement of stress levels in maritime crews. However, a number of examples of the application of other psycho physiological methods in the maritime domain e.g. Heart Rate Variability can be found (for example [9]) Examples of use of EEG technology in aviation and flight simulators can be found as well (see for example [12].

　

　

　

 The use of EEG techniques is particular interesting and promising because they represent very objective and accurate means of measurement of internal (not directly observable) mental activity of the crew EEG measurement is therefore likely to produce more valid and reliable measures of certain human factors parameters than the more subjective methodologies based on observations and questionnaires.

　

 It is therefore relevant - on the basis of the great potential of EEG methodologies with respect to assessment techniques in maritime simulator based training - to describe the application of EEG methodologies in the SPIN-HSV project. It is important to notice, that the use of EEG methodologies in the SPIN-HSV project not is with assessment as the purpose. But since the introduction of EEG based measurement in the maritime domain is new, any experience from the application of the methodology will be of relevance and interest to the maritime human factors community. If the technique in itself is mastered, the application as assessment tool is pretty close.

　

 SPIN-HSV is a collaborative Research & Technology Development project under the 5^th Framework Programme of the European Union. The full title of the project is "Shipping quality and Safety of high-speed vessels, terminals and Ports operations. In Nodal points". The objective is assessment of procedures, technologies and existing policies aiming to optimise high speed shipping, and the interface between vessels and terminals and to improve quality shipping.

　

 One point of focus in the project is the experimental quantification of human factors parameters such as mental workload, fatigue and stress levels in normal high-speed vessel operation and the quantification of variations in these parameters during the voyage and time in harbour.

　

 The experiments will be designed as laboratory experiments with use of a maritime full mission simulator as the laboratory and setting. The parameters will be quantified by means of observational and questionnaire based methods (e.g. instructor rating, observation of communication and SAGAT and NASA-TLX like methodologies), and the use of psycho physiological recording methods of brain activity (EEG) will be introduced as a measure of stress levels and overall mental activity of the crew.

　

 The subjects used in the experiment are both inexperienced cadet students and experienced HSV crews. The experienced subjects will sail well-known vessels in well-known environments (routes) and well-known vessels in new environments (routes).

　

 The validity of the methodology will be tested two fold:

　

　

 The use of composite measures will support the internal validity and helps us answer the question: Are we really measuring the level of stress? The use of control measures will support the external validity and helps us answer the question: Can we generalise from our findings to the real world outside the simulator (the laboratory)?

　

 As described in the beginning of this paper, it is possible to enhance the overall quality of the assessment - especially the validity - by composite use of methodologies. The methods described in this paper covers a variety of methodologies from subjective measures based on observations and questionnaires to objective measures based on psycho physiological techniques such as for example Heart Rate Variability or electric activity of the brain (EEG). Corresponding to that, the human factors parameters suggested as indicators of improved performance (improved knowledge and skills) also covers a wide selection, for example mental workload, situation awareness, stress and fatigue.

　

 The principle of composite use of methods could therefore emerge to be a very powerful assessment tool for evaluation of the effect of maritime simulator based training. If - for example - the effect of Crew Resource Management training is examined, more than one parameter should be used as indicator (e.g. mental workload and stress), and each of these parameters should be measured by means of more than one method or technique (e.g. mental workload measured by NASA-TLX questionnaire and Heart Rate Variability and stress measured by Galvanic Skin Response and EEG).

　

[1] Rabjerg. C.T. & Smyth, R. (2003). A UNIQUE CONCEPT FOR SIMULATOR BASED CADET TRAINING. Submitted for: MARSIM'03, Kanazawa, Japan, August 25-28, 2003.

　

[2] Andersen. P.B. (1998). Analysls of Maritime Operations II, Analysis. Center for Human-Machine Interaction. Report CHMI-2-1998, p. 64 (http://www/cs.aue.dk/~pba/Reports/MarOp2.pdf)

　

[3] Endsley, M. R. (2000a). Theoretical Underpinning of Situation Awareness: A Critical Review. In M.R. Endsley & D.J. Garland (Ed.), Situation Awareness Analysis and Measurement (pp. 3-32), Mahwah, New Jersey: Lawrence Erlbaum Associates.

　

[4] Grech, M. & Horberry. T. (2002). Human Error in Maritime Operations: Situation Awareness and Accident Reports. 5th International Workshop on Human Error, Safety and Systems Development 17 - 18 June 2002, Noah's On The Beach, Newcastle, Australia.

　

[5] Koester. T. (2003). Situation Awareness and Situation Dependent Behaviour Adjustment in the Maritime Work Domain. Submitted for the 10^th International Conference on Human - Computer Interaction (HCI International 2003). Crete, Greece, June 22-27, 2003.

　

[6] Endsley. M.R. (2000b). Direct Measurement of Situation Awareness: Validity and Use of SAGAT. In M.R. Endsley & D.J. Garland (Ed.), Situation Awareness Analysis and Measurement (pp. 147- 173), Mahwah, New Jersey: Lawrence Erlbaum Associates.

　

[7] Hilburn. B. & Jorna. P.G.A.M. (2001), Workload and Air Traffic Control. In Hancock. P.A. & Desmund, P.A. (Ed,), Stress, Workload and Fatigue (pp. 384-394), Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers.

　

[8] Stem. R.M., Ray. W.J. & Quigley. K.S. (2001), Psychophysiological Recording (2.ed.). Oxford: Oxford University Press.

　

[9] Van Westrenen. F. (1999). The Maritime Pilot at Work, Delft: Uitgeverij Eburon.

　

[10] Rau. P.S. (2001). A Heavy Vehicle Drowsy Driver Detection and Warning System: Scientific Issues and Technical Challenges. In Hancock. P.A. & Desmund, P.A. (Ed.). Stress, Workload and Fatigue (pp. 503-512) Mahwah, New Jersey: Lawrence Erlbaum Associates. Publishers.

　

[11] Fisch, B.J. (1999). Fisch & Spehlmann's EEG Primer (3. rev. ed.). Amsterdam: Elsevier.

　

[12] NASA Langley Research Center. March 11, 2003. Task 3. Physiological Factors (548-50-21).

　

 Document published and available on the Internet

　

 Thomas Koester, psychologist, MA. 2¹/2 years of experience as human factors expert (within the maritime domain, offshore industry, power plants and the medical domain) at FORCE Technology, Division for Maritime Industry (DMI).

　

 Peter K. Sorensen, shipmaster, B.Sc.Eng. 10 years of seagoing experience and 9 years seniority at FORCE Technology, Division for Maritime Industry (DMI) as instructor, navigation advisor and manager. Peter K. Sorensen is also chairman of the Danish Human Factors Center which is a cooperation between FORCE Technology and the National Danish Research Laboratory RISOE.