Bachelor Thesis Driver`s attention during monotonous driving and
Contents
1. s6 394 38 36 380 42 52 395 3113 391 5099 N A N A s7 423 17 37 454 15 09 488 14 53 512 1556 N A N A s8 511 10 15 482 977 510 1455 469 17 24 N A N A s9 466 9 09 439 10 28 485 8 84 447 17 08 513 19 41 Table 6 2 Results measured from the Cz electrode s Block 1 Block 2 Block 3 Block 4 Block 5 ms uV ms uV ms uV ms uV ms uV sl 473 2077 549 2623 549 262 544 2608 N A N A s2 591 14 06 567 16 49 546 15 74 N A N A s3 498 12 28 486 18 38 502 16 74 578 22 05 545 13 12 s4 373 8 129 389 10 65 450 15 26 374 12 28 N A N A s5 461 13 97 474 15 78 467 16 41 458 1682 N A N A s6 401 33 8 374 39 393 28 22 391 4606 N A N A s7 423 14 13 452 13 42 489 14 34 511 17 6 N A N A s8 513 1065 495 12 33 511 1626 485 1809 N A N A s9 467 15 77 438 13 27 486 1426 450 1957 513 222 The next figures Figure 6 4 Figure 6 5 Figure 6 6 Figure 6 7 show grand averages over all subjects in each block The results from Cz electrode are presented only The results from other electrodes were similar and they are shown in the table with the final grand averages comparison Table 6 4 The fifth grand average is not presented in the final comparison due to a small Table 6 3Results measured on Pz electrode number of measured blocks Cz uv 202. T 100 T T 200 300 T T 400 500 T 600 T T 700 800 ms Figure 6 4 Grand average the first block 43 Figure 6 5 Grand average the second block T T T T T T T T 100 200 300 400 500 600 700 800 ms Figure 6 6 Grand average the third block T T T T T T T T 100 200 300 400 500 600 700 800 ms Figure 6 7 Grand average the fourth block 44 The following table Table 6 4 presents the final results of grand averages of Fz Cz and Pz electrodes Latencies and amplitudes of the P3 component are shown Fz Cz Pz ms uV ms uV ms uV Grand average 1 462 9 635 464 10 01 470 10 56 Grand average 2 476 11 71 476 13 95 476 15 3 Grand average 3 494 9 5 495 11 98 532 14 79 Grand average 4 467 10 51 471 12 37 479 1431 Table 6 4 Grand averages comparison 45 7 CONCLUSION The goal of my bachelor thesis was to understand the basic principles of design an implementation of ERP experiments to get knowledge to assemble my own experiment and to investigate effects of fatigue For better understanding of the principles of the EEG analysis I was also supposed to develop an application for detection of artifacts in recorded data After acquisition of theoretical knowledge my first task was to design appropriate environment for the purposes of the experiment I created a simple and functional track which meets the cond
3. Detection difference level Raw data s Data orientation Soft 30 uv Data Format Medium 25 pV Sampling interval UV 5 Hard 20 py EM D Resolution Output type Marker File 5 Binary File Start 01 Dyrczyk eeg BINARY MULTIPLEXED 1000 0 0488281 Figure A1 Application overview The program has several settings The first is an option to set a difference in level adjustment It is a maximum value of difference between mentioned intervals The second is the choice of the marker file export or binary file export Only positions of artifacts are exported to binary file Detection can take several minutes processing a big file so the progress bars implemented It informs about the state of the task The processing task can be cancelled Figure A2 The message dialog Figure A3 informs about 49 successful finishing of the task Another dialog is shown in the case of the invalid format of the file Figure A4 Outputs are located in the directory of jar launch In the case of the marker export output it is named Artifact detection output vmrk and Artifact detection output bin dat for the binary export Detection i Complete 57 9 CTAX l Only for multiplexed data Format Figure A4 Wrong format of data loaded 50 ATTACHMENT B THE CONTENTS OF DVD The contents of DVD records from measurement complete EEG records eeg header files vhdr
4. The acquisition parameters as well as the impedance check are automatically stored and can be accessed from within the analysis software at any time A complete evoked potential analysis can be performed in real time directly in BrainVision Recorder and the segmented averaged data can be stored together with the raw data The incoming data can be sent out to the network via the TCP IP protocol 4 9 2 BrainVision Analyzer 2 0 1 BrainVision Analyzer 15 software contains numerous modules and calculation methods for the analysis of EEG data The Analyzer includes necessary pre processing functions enhanced time frequency analysis options ICA LORETA MRI correction and a direct interface to MATLAB It is able to read and process the EEG data of numerous EEG amplifiers from various manufacturers Segmentation based on event markers is available to reduce the space required by EEG files Averaging based on event markers is available to form evoked potentials during recording The editable history tree allows users to organize explore and trace evaluation steps 4 9 3 X Motor Racing 1 36 X Motor Racing 16 is aracing simulator with accurate physics simulation It has an advanced graphics engine supporting True HDR rendering and FSAA Full Scene Anti Aliasing XMR SDK Software development kit allows exporting all in game physics data to build a motion platform telemetry system or external controlling of the vehicles The developer releas
5. marker files vmrk Analyzer2 workspace including the complete history tree of transformations with the peaks found XMR Editor 1 32 the editor for the track from the game X Motor Racing Track version for the editor bp 2 wld exported version for the game track trk Artifact Detector sources the build file jar file the testing set of data Documentation for Artifact Detector KeyCounter exe 51
6. the P3a occurs most frequently in frontal area the P3b is located in the temporal or parietal area Potential jV 0 100 200 300 400 500 Time after stimulus ms Figure 2 5 1 A waveform showing particular components 7 2 6 Typical ERP experiment Basically every ERP experiment has a similar structure of the process which can be summarized into several following steps 7 li iii iv Design of scenario for the experiment respecting advantages and disadvantages of the ERP method respecting the general principles rules and strategies quantity and frequency of events stimuli scenario of the experiment is often designed and presented by a software application Preparation of the measured subjects placing of selected scanning electrodes electrodes can be the part of the EEG cap or be independent instructing of the measured subject about his her behaviour during the experiment elimination of artifacts Getting of the EEG record amplification and filtering of the EEG signal during the experiment storage of EEG signal together with markers markers represent single events stimuli in the scenario Data processing identification and elimination of artifacts extraction of the ERP from the EEG record averaging or other methods application of filters isolation of ERP components Statistical analysis examination of ERP components interpretation of results 2 7 EEG laboratory An E
7. the conductive EEG paste The measured subject goes through several procedures First the subject sits down to the car simulator and he adjusts the seat to the comfort position Then he she takes a short test drive He she tries accelerating braking and handling of the car The required task is explained to the subject during the test drive The subject is informed how to react for the stimuli then comes an instruction about his her behaviour during the experiment The subject is supposed to be concentrated during measurement and he she should try to minimize all unnecessary movements especially eye blinking and tongue movements After the test drive it is important to choose ideal projector brightness and contrast settings and set ideal lighting conditions The lights should not be very shiny projector visibility but not too much dark overexpositon of projection At last the subject has to remove any unnecessary electric devices phones music players watches etc and he she is asked about his her condition The subject should not be thirsty and he she should not need to urinate etc When it is all right we can go to the electrode cap placement 4 7 Application of EEG electrodes This part of the experiment has to be handled very carefully Improper placement or poor conduction of the electrodes affects final results For required conductivity there is a need to follow several steps suggested by the neurologist After preparation of ac
8. the electrodes A medical syringe with a special blunt needle is used for conductive gel application The first step is slipping the needle through the hole After that we hold the electrode and twist the syringe simultaneously then the conductive gel is injected Twisting of the syringe takes hair out of the electrode it increases general conductivity and its resistance due to movements of the subject uU Z y S X Figure 4 7 2 Subject is prepared for the experiment When the electrodes are prepared Figure 4 7 2 it is the time to check their impedance Values of the impedance are visualised by the software BrainVision Recorder These values should be as small as possible After the first impedance check the conductivity of the electrodes has to be usually slightly improved it means to repeat twisting of the syringe in electrodes with bad conductivity If it is necessary another injection of the conductive gel should help When each electrode has an ideal impedance 0 2 kQ the subject is temporarily disconnected then he she is moved back into the car simulator where the 27 connection is restored again When the subject calms down and the signal is getting stabilized the experiment can be started Figure 4 7 3 Electrodes used during the measurement 4 8 Observation of the experiment The process of the experiment Figure 4 8 1 is observed and controlled on several screens The most important screen is a mon
9. E Y suona Rios hif dle a E AT MEUS 17 4 Realization PEL ceca tact casos Gute naai ase oad na ute BR TAA as 19 Zl CIniroducWoDbus esc tasted utet eu aca Diete qbus bu e roe 19 4 2 Selecting of simulation software erento er rer tin de nerit 19 4 2 1 Software requirements aurei elati a vehe ev Utt IIR 19 422 Tested SO Wares ceinture hdi ten iaa decia ace AH 20 4 2 3 Comparison and resull aieo dettes oon eta ind uei 20 4 3 Constr ctiomof th track aue nis do ee e UO ond seeds Cru ater BON NE 21 44 Editing of vehicle phy sits iesadscanininam acto aire eise ind ines 22 4 5 Scenario of the experiment uneiuissattn dateien taa rU dj Eid 23 4 5 1 Modification of the scetiario ae Deer te toa P e edes 24 4 6 Preparation for the expeblimetitsaunsapecesered rues tovn grec tenere re eius 25 4 7 Application of EEG electrodes eic tovt d TNR p qais 25 4 8 Observation of the experiment iss cn ete S Dette XII bidon dis 28 49 Used aiias Cil casasin easter soueas EAE A EE ai 29 4 9 1 BrainVision Recorder oscuras audeant tinae ine dba an 29 4 9 2 BrainVision Analyzer 2 0 1 Lascnrecno ectpentetvgelemtrebun tetto 30 4 9 3 AXMotor Racing LoG aeeoa decre ended ear prs E d er 30 494 XMR Editor U8 2 n iiien and cb uad oiei P uad Qd rita 31 49 5 V ehiclePhiysics PRO iori bri e cardi seite pal ena Pro daga 31 490 JKeyCounter M cucina e ER DOO Vt ep ne RE 31 4 9 7 Debut Video Capture DoILWate sodes steer edema eese 31 4 10 Developed SOltw
10. EG laboratory has to be equipped with hardware devices important for the presentation of the scenario and EEG record acquiring It is necessary 4 to keep the measured subject strictly focused on the experiment Also it is very important to shield EEG devices from the electromagnetic interference An ideal laboratory contains an electronic shielded chamber This room does not allowing to feel any changes of light ambient sounds or other disturbing elements to the measured subjects Light conditions air conditions and room temperature are set to suitable ideal values due to sensitivity of the EEG signal to subject s blinking perspiration etc 10 3 STATE OF THE ART This chapter presents already performed experiments connected with the EEG driving or attention First there are mentioned related experiments performed at University of West Bohemia during past years followed by experiments associated with human attention performed in other laboratories 3 1 Experiments performed at University of West Bohemia 3 1 1 Analysis of driver s attention and ERP Task An experiment performed by Tom Krej in the year 2009 8 should prove an increase of latency of the P3 component of driver using a car simulator Scenario The primary goal of this experiment was driving without an accident Various stimuli were placed at defined places of the scenario These stimuli were a part of simulation software The primary stimulus wa
11. IFFERENCE Float RESOLUTION ArrayList lt Artefact gt artefacts ArrayList Float small4verage ArrayList Float big verage Loader load Output out Gui qui boolean marker boolean canceled Detector Gui qui String path String nam void run void detection float getAverage ArrayList Float values ArrayList Artefact reducelntervals Arr ArrayList lt Artefact gt getArtefacts o DatalnputStream in o File file Loader String path void loadFile int getFileSize void byteSkip float readvalue void closeFile Figure 4 10 3 Class diagram 34 Graphic user interface Class GUI for better application control is implemented Because the detector Class Detector runs in the separate thread there is no GUI blocking Output sample Brain Vision Data Exchange Marker File Version 1 0 Common Infos Codepage UTF 8 DataFile TEST eeg Marker Infos Each entry Mk Marker number Type Description Position in data points Size in data points Channel number 0 marker is related to all channels Fields are delimited by commas some fields might be omitted empty kl New Segment 1 1 0 20120412095517014141 k2 Stimulus AF 563 0 0 k3 Stimulus AF 566 0 0 k4 Stimulus AF 567 0 0 k5 Stimulus AF 568 0 0 k6 Stimulus AF 569 0 0 k7 Stimulus AF 570 0 0 k8 Stimulus AF 571 0 0 k9 Stimulus AF 572 0 0 4 11 U
12. P task scenario at last a list of used hardware and software is presented A specific software tool was developed specially for purposes of the experiment The part of the scenario which deals with stimulation of the subject was designed in two steps First was created testing scenario using all theoretical knowledge After testing and discussions the scenario was slightly modified The placement of electrodes was consulted with the neurologist MUDr Irena Holeckov to reach the best possible conductivity of the electrodes 4 2 Selecting of simulation software 4 2 1 Software requirements ii iii Graphics Game graphics should be smooth not much pixeled and overall realistic Track customization For purposes of the experiment the used software should offer customization of the environment preferably it should be possible to create a track from the scratch Customization of vehicle physics and vehicle handling The experiment is not a race so vehicle physics of the racing simulator has to be adjustable The car should be slow with good handling to avoid driver s mistakes during measuring process 19 iv Compatibility with used controller The software has to be compatible with the game wheel and other components installed in the cabin of the prepared car simulator v Price The software tool should be cheap distributed for free in the best way 4 2 2 Tested software World Racing II TDK Meadiacti
13. The last block was another dual task Participants were requested to press a button embedded in the steering wheel when they started to take the action of changing This block might help us to calculate the reaction time The whole experiment lasted 3 hours totally Results The results from driving performance analysis indicate that there is no deterioration of performance when the auditory secondary task is added Two possible reasons might account for the stabilized driving performance One is that the LCT occupies the visual cognitive resource whereas the PASAT occupies auditory cognitive resource Another possibility is that the concurrent tasks indeed evoke higher task load compared with the single task condition However the participants is instructed that to maintain the performance of 16 primary task LCT is more important than the secondary task which could account for that why there is a deterioration of PASAT performance but no decline of driving performance 3 3 Summary It is possible to conclude several recommendations from the experiments presented above Some of these facts are applicable for a new experiment If we take the recommendations into consideration together with respecting the typical EEG practices Chapter 1 6 and 1 7 we get the basic options of this new experiment scenario Simulation environment A simple easy track used during the driving experiment can avoid distraction of attention from the ta
14. University of West Bohemia Faculty of Applied Sciences Department of Computer Science and Engineering Bachelor Thesis Driver s attention during monotonous driving and visual stimulation ERP experiment Pilsen 2012 Jiri Divi ACKNOWLEDGEMENT First of all I would like to thank my bachelor thesis supervisor Ing Roman Moucek Ph D for his kindness and patience with leadership of my work Also I would like to thank Ing Pavel Mautner Ph D I appreciate their valuable advices and comments which brought this work into the final form STATEMENT I hereby declare that this bachelor thesis is completely my own work and that I used only the cited sources Pilsen May 11 2012 Ji Divi ABSTRACT The major contribution of this thesis is to discover if it is possible to predict the driver s attention by measurement of his her brain activity During monotonous driving attention tends to decrease The drop of attention due to fatigue might have serious consequences for the driver and for other traffic participants as well One method to measure human brain activity is called the electroencephalography EEG Together with the EEG signal it is possible to observe specific neuronal responses connected with cognitive stimulation of the subject These responses are known as event related potentials ERP There is a hypothesis claiming that fatigue causes the shift in latency of the defined ERP component An increase of this latency i
15. alculated It means the average value of the same blocks of all subjects The final result is comparison of these grand average values Primary U D1 Dyrczyk Ey 02 Walter 4 TI 03 Hula 4 38 Raw Data 4 10 Filters 4 10 Segmentation 4 10 BaselineCorrection 4 e Artifact Rejection 4 iD Average 5 Peak Detection 4 Xs Average 2 6 Peak Detection 2 4 e Average 3 0 Peak Detection 3 4 e Average 4 OR Peak Detection 4 4 103 Average 5 6 Peak Detection 5 Figure 5 3 1Transformations shown in Analyzer2 history tree 40 6 RESULTS The P3 component was clearly observable in all blocks of each subject Only exception is the second block of the subject s2 The following figures Figure 6 1 and 6 2 present examples of found peaks the last figure Figure 6 3 shows situation when it was not possible to recognize the P3 component Figure 6 1 Peak of the P3 s5 1 block on the Cz electrode Figure 6 2 Peak of the P3 s3 1st block on the Cz electrode 41 Figure 6 3 Example of unrecognized component s2 2 block on the Cz electrode The following tables Table 6 1 Table 6 2 and table 6 3 present averaged values of the latency and voltage of particular P3 peaks The first table shows the results from the Pz electrode Latencies in milliseconds and values of the P3 peak in uV for every measured block of each subject are shown Unrecognisable values of blocks are represented by the question mark The
16. ar eink oisaimweitecnapinn ta reisen rt Rep put Got etta 32 410 1 Arttifact Detector usce Rho dun ad ME NE bas 32 4 11 Used Ward Waves enesmetete pe an onret E Riu dd aeos ts 35 5 JDataapalvsis sod etd e edd teg ato ap a adis 37 5 1 Participants of the experiment uso teenduree teet attentis dens ori iniu 37 D Presentation Ob SUDIBCES sensere nenene enion ai 37 bs Data ProceSSIN geiis a a e id AE Dieu TES 39 6 Results niena betta e aE dede ttp Reit p ERE AE 41 7 RORCIISION qapuithdanitrili mdp nibii emat AE 46 8r Bibliography dues dod ob D EORR AN RD BU Ue ed 47 Attachment A User documentation for Artifact Detector 49 Attachment B The contents of dvd 22 s eeeeeeeeeeeee eene enini 51 1 INTRODUCTION Driver s attention is essential for safe traffic Inattentive drivers are dangerous to their surroundings and there is raising possibility of causing serious accidents Attention tends to decrease especially during long time monotonous driving To minimize this risk it is important to predict fatigue connected with delayed driver s reactions One of the possibilities of predicting inattention is monitoring the driver s brain activity This experiment uses electroencephalography EEG to analyse the brain activity The first step of forming the experiment is to get an appropriate theoretical knowledge The theoretical part of this thesis includes introduction to EEG the method of event related pote
17. arp turns noticeable height differences or unexpected object occurrence It is a ring track the time to Ex Editor 133 bg DD Fite Edt View World Material Lights Operations Camera Took Window Help Common Patch Fast Edi Patch Ad tcnal Fast Edt Patch Standart Object Additional Obiect Addriona Ede M 99e v eoxYZNzaanz Sekcion v x wale s olla ory E Figure 4 3 1 Track editor finish one lap is about 6 7 minutes The scenery is very homogenous so it is hard to detect subject s orientation on the track The track was modelled in the XMR editor Figure 4 3 1 by following added video tutorial instructions This tutorial 16 was made directly by the X Motor Racing developer 21 4 4 Editing of vehicle physics The maximum horsepower was reduced to the value of 45hp maximum of RPMs was lowered to 5000 rpm a maximum value of torque was set to 52 4620 N m rpm Also a brake power was reduced for better vehicle stability The car reaches 125 kph topspeed Transmission was set to an automatic mode The vehicle physics was changed using the Vehicle physics PRO Figure 4 4 1 software which is a standard component of X Motor Racing installation 22 S VehiclePhysics Pro 13 je ims File Options Help Vehicle 1 Dosster EE Common GP Aera QW steering 3 Brakes g Engine a oq re Wheels 8 Tire Model MHRT RESET Eng Po
18. cessories Figure 4 7 1 the subject can sit down to a comfortable chair It is good to keep on mind subject s clothing or his hers bodily needs because after placement of a cap it can be hard to go to 25 the toilet or to take redundant clothes off Then it is time to select an electrode cap in a proper size and cover subject s shoulders with a towel due to possibility of his her clothing pollution After this we can start fitting the electrodes At first it is necessary to clean the subject s forehead the frontal reference and right ear lobe the ground electrode with the abrasive paste to increase skin sensitivity Then the electrode cap is fitted The front edge of the cap should be 2 4 cm above eyebrows the cap has to be centred the central electrode has to be at the top of the subject s head Then we spread an EEG conductive gel to contact the surface of the ground and reference electrodes with the skin Figure 4 7 1 Abrasive paste conductive gel and necessary accessories The ground electrode is placed at right ear lobe the reference electrode is in the middle of the forehead between eyebrows and the cap These electrodes have to match the previously cleaned spots Final EEG data come as a result of brain activity recorded by 16 electrodes placed on the cap Figure 4 7 3 It is important to get proper connection between the skin and the electrodes so 26 alittle drop of the conductive gel is injected through holes in
19. cial software tool This method also allows monitoring of subject s reactions for the stimuli during the experiment So it is possible to check if the subject reacts attentively and correctly The second issue of the testing experiment was an excessive amount of the target stimuli Fifty minutes long continuous stimulation risks subjects overstimulation connected with the decrease of the EEG signal amplitude the target stimuli becomes more predictable Recognition of the ERP components will be more difficult in that case This problem was solved by inserting 5 minutes long pause after every ten minutes of the stimulation The subject simply continues driving trough the scenery during the pause but there is no stimulation The experiment contains four blocks of 10 minutes stimulation intervals Because of the non stimulating block pause between every stimulating block the time of the experiment is newly set to 55 minutes Times of the blocks are approximate because the LED module has manual turning on off If the subject still feels comfortable after all four blocks of the stimulation the amount of blocks can be increased to five approximately 70 minutes long experiment 24 4 6 Preparation for the experiment Before the experiment starts it is necessary to instruct the subject and set ideal conditions for the experiment The car simulator and the laboratory need to be ventilated several minutes before the subject s arrival due to the smell of
20. cs with more suitable conditions for the experiment The target stimuli were defined sounds or blinks of coloured LED non target stimuli were represented by another colour of LED or different kind of the sound Duration of the experiment was set about 40 minutes for each tested subject Results The P3 component was recognized but an increase of the peak latency was not confirmed It could be caused by a shorter duration of the experiment a small number of measured objects or bad conditions of the experiment that may caused occurrence of artifacts 12 3 2 Experiments performed outside University of West Bohemia 3 2 1 Effects of mental fatigue on attention An ERP study Abstract The experiment was performed by Maarten A S Boksem Theo F Meijman Monicque M Lorist in Groningen Netherlands 12 Subjects performed a visual attention task for three hours without rest Subjective levels of fatigue performance measures and the EEG were recorded Subjective fatigue ratings as well as theta and lower alpha EEG band power increased suggesting that the three hours of task performance resulted in an increase in fatigue Reaction times misses and false alarms increased with time on task indicating decreased performance efficiency in fatigued subjects Subjects were unable to inhibit automatic shifting of attention to irrelevant stimuli Scenario Seventeen healthy participants were recruited from the university population Each expe
21. directly on a PC from a webcam 31 4 10 Developed software 4 10 1 Artifact Detector Introduction Artifact Detector is a tool developed to search for depreciated spots connected with the EEG signal called artifacts Artifacts are sudden oscillations 100 uV and more in the EEG signal caused by the unwanted subjects s activity Figure 4 10 1 The application marks the places of artifacts occurrence Fz rere c eir RA 00 00 14 00 00 15 00 00 16 00 00 17 Figure 4 10 1 Example of artifacts appeared on the Fz electrode The application works with BrainVision EEG files The record consists of three files the header file marker file and raw data file The header vhdr file describes the EEG This file is an ASCII file The application reads the header file with information about the format of the data file eeg then it analyses raw data measured on the Fz electrode The output is saved to the marker file vmrk or a binary file In the case of saving into the marker file there is a possibility to check the result in the BrainVision Analyzer2 spots with artifacts are shown by markers The format of the header file is based on the Windows INI format It consists of sections of different names containing keynames and assigned values Here is an overview of the most important switches For detailed information see 14 DataFile the name of the file with the EEG record e g TEST eeg 32 MarkerFile t
22. e like background of regular EEG activity in this case the EEG is noise only and there is a need to remove it in a proper way average value For determination of ERP waves the object has to be repetively stimulated with the same stimuli and the stimuli moment has to be precisely synchronized with the EEG record The ERP technique is divided into several types Exogenic response to a physical stimulus it is characterized with a shorter latency time Endogenic is related to a cognitive process a longer latency time 2300 ms The ERP technique is also divided by the type of used stimulation Auditory ERP stimuli are for example a short beep with the defined frequency the response is a set of waves which define how neuro information spreads by nerves of hearing sense Visual ERP stimuli are for example an image of a checkerboard where fields of the board randomly change the colour or blinking image We follow spread of information by sight nerve Somatosensoric a reaction to various stimuli like motion touch temperature etc 2 5 1 Major ERP components This section presents a short description of the ERP components used in cognitive neuroscience research 7 ERP components are usually labelled according to their polarity P positive N negative C polarity is not uniform and it can occur in the positive or in the negative spectrum of different subjects and position of the wavefo
23. erforming in vehicle tasks while driving Paced Auditory Serial Addition Task The PASAT is a measure of cognitive function that assesses auditory information processing speed and flexibility as well as calculation ability The PASAT is usually presented using 15 an audiocassette tape or compact disk to ensure standardization in the rate of stimulus presentation Single digits are presented every three seconds and the participant must add each new digit to the one immediately prior to it The digit was randomly arranged to minimize possible familiarity with the stimulus items when the PASAT is repeated over more than one occasion Overall 30 participants between the ages of 20 34 were assessed The experiment involved four blocks The first block was the primary driving task Participants were requested to perform the LCT under three different speeds 60 km h 80 km h 100 km h which represented three task load levels low moderate and high respectively The second block was PASAT under two paced conditions slow and fast the numbers were presented every 5 and 3 seconds termed p5 and p3 Participants were requested to calculate the numbers and report the results The third block was the combination of the primary and secondary task Participants were requested to do the calculation at two paces while performing the LCT with a fixed speed 80 km h 80 p5 and 80 p3 However they were instructed that the primary task was more important
24. erred hand R right handed L left hander 37 Length length of the experiment the length of each experiment may be little different due to manual switching on off of the LED module Blocks value refers to number of stimulation blocks Count the value refers to the number of counted targets Attentive attention of the subject was checked in random intervals of the experiment Y means that subject was attentive Artifacts occurrence of artifacts N means minimum of blinking or other distractions little blinking refers to increase the amount of blinking medium blinking means more artifacts but measured data is still usable Subject Gender Visual License Active lt rel Subject Blocks IN Attentive PEE a l a a a la a Artifacts little blinking occasional yawing N N medium blinking a hm edm le PR fo us lt ik k IK IK IK IK Table 5 2 1 General information about subjects Astigmatism bad curvature of cornea 38 5 3 Data processing files The most important eeg file contains raw binary dat The header file vhdr provides access to the data file It contains necessary information about the data file sampling frequency data orientation data format resolution etc Marker file vmrk contains information about occurrence of the markers These markers are time stamps
25. es new versions updates and additional content very frequently The physics engine covers all aspects of vehicle dynamics that can be tested immediately in the game Many of the customizable car physics 30 parameters include common mass and aerodynamic value steering brakes engine suspension transmission properties and tyre model There is also allowed reading internal physics information from the physics engine like position speed acceleration forces transform matrixes suspension properties tires data etc 4 9 4 XMR Editor v1 32 XMR Editor 16 is a tool to create edit tracks and vehicles The tool includes 3D Max export utility that allows exporting any object from 3D Max Also it allows creating of a new track from the beginning 4 9 5 VehiclePhysics PRO Vehicle Physics 16 is a special utility that allows the customization of vehicle physics This is the first software of its kind that allows full customization of its tire model Aerodynamics steering system braking system engine suspension transmission damage model and tire model are customizable 4 9 6 KeyCounter1 1 0 KeyCounter is 17 freeware software which counts selected keystrokes For purposes of the experiment the right shift paddle was assigned as key A An assignment was provided by a default Logitech gaming wheel system driver 4 9 7 Debut Video Capture Software Debut is a video capture software 18 which can record video files
26. field N A means not available It represents blocks which were not performed Block 1 Block 2 Block 3 Block 4 Block 5 ms uV ms uV ms uV ms uV ms uV s1 483 21 89 517 21 69 549 161 527 1863 N A N A s2 565 10 39 531 1175 545 124 N A N A s3 475 6779 482 13 16 489 1039 592 13 84 561 8 22 s4 399 6 9995 377 6 143 445 11 62 384 1001 N A N A s5 447 1273 472 10 08 469 106 459 1143 N A N A s6 419 2748 386 32 94 392 1997 391 3555 N A N A s7 423 1764 454 1479 487 12 67 512 1409 N A N A s8 510 1001 475 8 232 509 9362 469 1407 N A N A s9 466 11 85 450 14 72 485 1254 478 17 87 512 15 82 Fz Table 6 1 Results measured from the Fz electrode ye Block 1 Block 2 Block 3 Block 4 Block 5 ms uV ms uV ms uV ms uV ms uV s1 473 18 76 548 2441 549 23 19 544 23 42 N A N A s2 592 11 66 566 1284 545 14 02 N A N A s3 486 9 694 486 16 82 497 1409 579 1885 563 11 11 s4 398 6 98 386 8 963 447 1037 377 11 N A N A s5 462 1627 473 15 86 468 1639 458 1708 N A N A 42
27. he name of the marker file e g TEST vmrk DataFormat the format of data possible values BINARY or ACII DataOrientation orientation of stored data possible values VECTORIZED first the file contains all data points for the first channel followed by all data points for the second channel etc MULTIPLEXED all channels for every data point follow on from each other directly The data structure is multiplexed SamplingInteroal value of the sampling interval in us BinaryFormat possible values Possible values IEEE FLOAT 32 INT 16 UINT 16 The application works for default settings of the recorder used in the laboratory at the University of West Bohemia Multiplexed binary data with floating point format are used Algorithm of detection For detection is used comparison between averages of two floating subsequent intervals Figure A2 The sizes of the intervals are fixed The values are read one by one Every newly read value goes into the first interval then its last value goes into second interval Then the last value of the second interval is removed Averages of both intervals are computed If there is a specific difference between both averages this value is marked as an artifact The value of difference is from 20uV to 30 uV Interval 1 has the capacity of 15 values interval 2 has the capacity of 110 values Figure 4 10 2 shows less values due to simplification This method works reliably but it
28. hese diodes are situated in driver s visual field on the dashboard of the simulator Figure 4 5 1 Figure 4 5 1 Placement of LED diodes The blink interval of diodes is 1500 ms This delay was also recommended by the neurologist One blink takes 500 ms It includes three colours of blinking diodes red green and yellow The red one represents a non target stimulus the green represents a target stimulus and finally the yellow is defined like arare stimulus The subject is focussed on blinking of the target green diode and he she counts up these blinks The subject counts silently for 23 himself herself The required P3 component origins in the unpredictability of the target stimulus Diodes blink randomly probability of occurrence are 75 red 20 green and 5 yellow The experiment takes 50 minutes 4 5 1 Modification of the scenario Testing of the designed scenario showed several imperfections The reaction to the primary stimuli counting up diode blinks took the subject out of concentration Trying to remember which number comes next was very distracting for the subject Also there was an occurrence of facial expressions like squinting tongue moves etc connected with the counting task Pressing pushing the right shift paddle button under the steering wheel replaced a quiet counting Each time when the primary stimulus appears the subject simply pushes the paddle button The button keystrokes are counted by aspe
29. ific neural responses in the signal are connected with sensory motor or cognitive events stimuli they can be extracted from the EEG signal with many techniques and then they can be processed These specific responses are called event related potentials 2 4 Artifacts Artifacts 7 are signals affecting the EEG record They are not cerebral in origin and basically can be divided into physiological and non physiological Physiological artifacts are generated from the patient itself include cardiac glossokinetic muscle eye movement respiratory and pulse artifact among many others The EEG recording can be contaminated by numerous non physiological artifacts generated from the immediate patient surroundings Common non physiological artifacts include those generated by monitoring devices infusion pumps and suctioning devices though electrical devices like mobile phones may also contaminate the EEG record 2 5 Event related potentials ERP The ERP technique 7 is an electric response of the brain or brainstem to different types of stimulation visual audio or sensory stimulation These potentials are recorded in the same way like the EEG electrodes placed on the head of the subject They have shape of short time waves with very low amplitude its shape latency or duration time depends on strength of stimuli and the mental state of the measured object In comparison with the EEG ERP waves are comparatively lower These waves com
30. is component is larger for stimuli containing target features and this effect is enhanced when the targets are relatively infrequent in this sense the anterior P2 wave is similar to the P3 wave However the anterior P2 effects occur only when the target is defined by fairly simple stimulus features whereas P3 effects can occur for arbitrarily complex target categories N2 it has been identified a lot of clearly different components in this time range These components occurs usually about 200 ms post stimulus The main N2 component is a response to repeated non target stimuli The N2 7 component can be observed as several subcomponents The N2b subcomponent is larger for less frequent target stimuli and it is thought to be a sign of the stimulus categorization process The N2pc is a subcomponent which reflects the focusing of spatial attention onto the target location P3 the third positive wave of the ERP record with latency approximately 300ms post stimulus Latency range is the value between 300 500 ms it depends on specific experiment Maximum of amplitude the P3 component reaches at the Pz electrode There are several theories about origin of the P3 e Memory updating the wave appears when is time to update working memory P3 occurrence caused by surprise when less frequent stimuli has appeared e Occurrence of the awaited stimuli The P3 wave should be composed of the two subcomponents P3a and P3b
31. itions of the experiment The next task was to come up with the scenario of stimulation in order to observe single components of the evoked potentials Because this analysis is related to attention of drivers the stimulation by visual cognitive events was used After thorough preparation the experiment might have been started One by one I measured and analysed the brain activity of ten participating subjects The first success was that in most cases it was able to observe the requested P3 component and to determinate the time of its occurrence After further examination and comparison it unfortunately failed to prove the expected hypothesis which predicts an increase of the P3 component latency as a result of the driver s fatigue Comparing the calculated averages the latency successively grew but at the end of each measurement it slightly decreased again Subjectively I think the result was unsatisfactory due to the characteristics of the EEG ERP method In our conditions it is not possible to perform EEG recordings for a long time Most of the participating subjects complained about discomfort and several participants got headache at the end of the experiment Although it was possible to observe the results the subjects were obviously disturbed and he she could not focus on the task 46 8 BIBLIOGRAPHY 1 Hartl Pavel Hartlov Helena Psychologicky slovn k Praha Portal 2000 2 Thorne Glenda Ph D Thomas Alice What is a
32. itor with a real time image of recorded EEG data There is possibility to check proper functionality of all electrodes to observe frequency of artifacts and to get a general overview about the data If there is a suspicious output we can check a monitor which provides a webcam stream directly from the cabin of the simulator It shows information about subject S movements and his her overall condition The third monitor displays a number of counted targets and it also allows us to check how 28 attentively the subject reacts to the stimuli There is also a need to check the projection if the subject drives continuously It is very important to watch the time because the LED module has to be manually switched on off to keep defined blocks stimulation non stimulation of the experiment Figure 4 8 1 Observation of the experiment 4 9 Used software 4 9 1 BrainVision Recorder BrainVision Recorder is a multifunctional EEG recording software 14 designed to provide versatile and easy to use platform for recording setup and execution A convenient menu structure simplifies these steps guiding the user through the entire hardware setup and hardware software filter configuration Even selecting the hardware filters on a channel by channel basis is very easy 29 and fast with Recorder It provides the channel by channel electrode impedance check Each electrode is placed at the topographic position and its impedance value is displayed
33. nment game is projected in front of the car simulator for more realistic feelings during the experiment Webcam Each subject is recorded by the HD webcam placed into car interior This record can prove or falsify occurrence of artifacts during experiment It also allows monitoring the subject s behaviour during the experiment 36 5 DATA ANALYSIS 5 1 Participants of the experiment Ten volunteers were individually invited take a part in the experiment One female and nine men were 18 24 years old all were students and all were healthy Nine individual usable EEG records were successfully taken One of the participants felt sick during the measurement so this experiment had to be terminated prematurely Two participants completed five stimulation blocks 10 minutes for each block The rest of the group took the standard of four blocks 5 2 Presentation of subjects The following table Table 5 2 1 presents summary of information about each subject and about a single experiment The legend for contents of the table is Subject identification of the subject Gender M is for male F is for female Age age of the subject in years Visual presence of visual impairment N No Y Yes value refers to the number of dioptres Hearing presence of hearing impairment Y N License driving license Y N Active active driving Y refers to active everyday driver N is for occasional driver Hand pref
34. ntials and the general measurement process The next chapter presents already performed experiments related to the topic This can provide helpful information and cornerstones for a new experiment This experiment is focused on extraction and analysis of the visual event related potentials ERP specifically the P3 component The realization part describes particular steps in creating the experiment The first step is to create a track with monotonous characteristics Then it is necessary to design a scenario of the cognitive stimulation The stimulation is essential for observation and analysis of ERP Subsequently performing of the experiment begins A group of volunteers was invited to drive a car simulator and EEG ERP of each participant was recorded The last section of the thesis presents the analysis process of measured data and evaluation of the results A software application detecting deprecated sections in the EEG signal was also developed 2 THEORETICAL PART 2 1 Human attention Attention can be defined 1 like concentrating of physics activity on a defined object or process e g focused listening instead of simple hearing Attention depends on subject and current environment attention of subject is attracted by originality eccentricity non expectancy or diversity of sensed object Otherwise attraction is generally weakened by fatigue There are two basic forms of attention 2 Passive attention refers to the involun
35. of stimuli appearance The BrainVision Analyzer 2 software Figure 5 3 1 is used for analysis of the data The P3 component was analyzed on Pz Cz Fz electrodes but it was also very often clearly observable on other electrodes mostly P3 P4 The process of data analyzing Figure 5 3 1 can be The BrainVision Recorder saves the measured signal into three types of summarized into several steps i ii iii iv Create a new workspace import the data file Apply IIR filter e Set high cut off frequency to 15 35 Hz as needed e Set high cut off slope to 12 db Oct Segmentation of data based on the marker position e Select the marker of the target stimulus e Set the time interval from 100 ms to 900 ms Baseline correction e Seta range for mean value calculation o Start 100 ms o End0ms Artifacts rejection e Manually mark segments with artifacts 39 vi Averaging e Set interval for the first block e Set interval for the second block e Set interval for the third block e Set interval for the fourth block e Set interval for the fifth block vii Result evaluation e Semi automatic peak detection for each averaged block vii Peak export After passing all mentioned steps is time to evaluate the results Average value of the latency of the P3 component in all blocks of each subject was calculated These averages were compared and the increase of the latency was looked for After comparison grand average values are c
36. rgets Stimulus letters were randomly chosen from the alphabet excluding the letters g i o q u x and y 14 Results Subjects developed more aversion against continuation of task performance with increasing time of task Subjects on average slowed down and missed more targets with increasing time on task 3 2 2 Driver s Mental Workload Assessment Using EEG Data in a Dual Task Paradigm Abstract The following experiment was performed by Shengguang Lei Sebastian Welke Matthias Petting Berlin Germany 13 The main goal of this study was representation of mental workload using EEG data A simulated driving task the Lane Change Task LCT combined with a secondary auditory task the Paced Auditory Addition Serial Task PASAT was adopted to simulate the situation of in vehicle conversations Participants were requested to perform the lane change task under three task conditions primary LCT LCT with a slow PASAT and LCT with a fast PASAT The analysis of event related potentials ERP revealed that LCT evoked cognitive responses such as P2 N2 P3b CNV and the amplitudes of P3b decreased with the task load A crucial benefit of these findings is that the increase or decrease of amplitudes of ERP components can be directly used for representing driver s mental workload Scenario The Lane Change Task LCT was an easy to implement low cost and standardized methodology for the evaluation of the attention associated p
37. rimental block began with the presentation of a fixation cross which remained on screen throughout a block of trials and was followed by the presentation of a memory set of two letters 2000 ms Next a cue frame was presented for 2000 ms to indicate which display positions left up or right up diagonal were relevant Thereafter participants were randomly presented a series of 160 stimulus displays constituting 1 block each 50 ms in duration as illustrated in Figure 32 1 13 Interstimulus intervals varied randomly between 1000 and 1500 ms Subjects received a new memory set after every odd block and a new diagonal cue after every even block The following restrictions applied memory set letters for one block could not be memory set letters for the next seven blocks and the cued diagonal could not be the same for more than four subsequent blocks 1 Block 160 Trials Memory set Diagonal cue Relevant Relevant Irrelevant Irrelevant target nontarget target nontarget Figure 3 2 1 Presentation of stimuli 12 The stimulus display contained two letters which were randomly presented on either the left up 5076 or the right up 5076 diagonal positions In 25 of the trials a memory set item appeared at a relevant diagonal position relevant target in 25 of the trials a memory set item appeared at an irrelevant diagonal position irrelevant target and in the remaining trials the display contained no memory set items nonta
38. rm Figure 2 5 1 which is time in milliseconds from the onset of the stimulus C1 is the first major visual ERP this wave appears to be generated in area of primary visual cortex Polarity of this component depends on the 6 location of stimulus in the visual field The C1 wave typically onsets 40 60 ms post stimulus and peaks 80 100 ms post stimulus and it is highly sensitive to stimulus parameters such as contrast and spatial frequency P1 the C1 wave is followed by the P1 wave which is largest at lateral occipital electrode sites and typically onsets 60 90 ms post stimulus with a peak between 100 130 ms However that P1 onset time is difficult to assess accurately due to overlap with the C1 wave P1 latency will vary substantially depending on stimulus contrast The P1 wave is also sensitive to the direction of spatial attention and to the subject s state of arousal N1 the P1 wave is followed by the N1 wave There are several visual N1 subcomponents The earliest subcomponent peaks 100 150 ms post stimulus at anterior electrode sites and there appear to be at least two posterior N1 components that typically peak 150 200 ms post stimulus one arising from parietal cortex and another arising from lateral occipital cortex N1 subcomponents appear to be larger when subjects are performing discrimination tasks than when they are performing detection tasks P2 follows the N1 wave at anterior and central scalp sites Th
39. rs with eyes closed It is blocked with eyes opened Delta activity frequency 4Hz 0 1 3 Hz e amplitude 10 300 uV occurrence symmetrical usually electrodes F3 C3 regular activity up to the age of 1 year Delta activity of the adult brain appears in the fourth phase of sleep It signalizes attention dysfunction in wide awake state of the adult brain Theta activity e frequency 4 7 Hz e amplitude lt 30 uV e measurable usually at the place of the temporal lobe amplitude may be higher over the left hemisphere e itisa common part of a regular record e it should not exceed values of alpha activity more than about 50 It is activity occurring between sleep and wide awake state It should appear by meditation or praying it is connected with human creativity intuition daydreaming fantasy memories Beta activity e frequency 14 40 Hz e amplitude 10 20 uV sometimes 20 30 uV e it should not be synchronic over both hemispheres e amplitude should be higher in the state of sleepiness it should occur in female brain more often It occurs mostly over the frontal lobe 25 30 Hz the parietal lobe 14 22 Hz or the occipital lobe The major disadvantage 6 of this method is troubles with the final processing of the EEG signal In the raw form the EEG is a cluster of different sources which originate from neural brain activity and so it is very difficult to isolate the specific neurocognitive process Spec
40. runs for a long time The complexity is O N when N is the number of analysed values It is resistant to baseline deflections 33 step 1 average2 step 2 average2 vales e e e e e e e e e step 3 average2 values e e e e e e e e average vales e e le e e e e e e e e o o e e e e average average Figure 4 10 2 Floating windows of averaged intervals Implementation and architecture The application was developed in Java SE Eclipse Galileo environment was used Figure 4 10 3 shows the UML class diagram 9 ad Output FileWriter out DataOutputStream out2 ArrayList lt Artefact gt artefacts String outputPath Output ArrayList lt Artefact gt artefacts Strin void outputToTerminal void outputToMakerFile void outputToBinFile Parser parser String absPath ProgressMonitor progressMonitor Gui void detect boolean updateProgressMonitor Float progress void placeRightComponents void placeLeftComponents int getDifference void evaluateDialog void showFinalDialog boolean checkVlaues void setGuiEnabled boolean enabled pbuonee e lrP oo o o on o oc oa void setPosition int position Parser File file m void parse String getInfotint i 9 ad Detector 9 ad Artefact Artefact int position 5 Scanner sc e o ArrayList String info int getPosition int SMALL AVARAGE NO oF int BIG AVARAGE NO int VOLTAGE D
41. s associated with a fade of attention The goal is to design and perform the experiment which verifies the hypothesis Another goal of the thesis is to design and develop an application for artifact detection CONTENTS 1 BOO OUO acea etit ode e QU de oa tvi ios ac ass dir Maca Den de 1 2 Jheoteticdl parti iori rari pas nda atus Hat pedi aiu i mata rade 2 2d Hyman QUOIHODbu oed eec ae Ae OUAIS NE ARIA AN ARE RR RUN Re HA 2 2 Fate CMe eset sg eau MD anpra d ad uaa idu d ub d HAN o URP SPINE 2 2 3 Electroencephalography EEG eite niet redeem rmt tt dana sh tes 2 Z4 Artifacts a ese siat Uie Reha be d he ute a PR ERE LN 5 2 5 Eyenterelated potentials BRI itecto mettra Ute teede atre 5 294 MajJorERP compohetils o iaceo ee naga trn aed Data 6 2 6 Typical ERP experimieritus dept ti uei vitet p iN nd 9 2 FEG laborato enengena tidem due estem in te uat 9 3 State OF tHe SEL vo vb a as Ovi ne Depp Ur ob n ep 11 3 1 Experiments performed at University of West Bohemia 11 3 11 Analysis of driver s attention and ERD 11 3 1 2 Driver s attention audio and visual stimulation 12 3 2 Experiments performed outside University of West Bohemia 13 3 2 1 Effects of mental fatigue on attention An ERP study 13 3 2 2 Driver s Mental Workload Assessment Using EEG Data in a Dual Task Paradiptisuossesetiten erede dat ite teca tutelle tese dix dad ian e d 15 Dos SOUP
42. s represented by an obstacle on the scenario track a vehicle crossing the track driver was supposed to avoid collision The secondary stimulus was to count roadside objects billboards placed in the scenario The scenario was designed as a monotonous boring track to decrease object s attention which should increase a probability of causing an accident The maximum length of the experiment was 60 minutes and the scenario was repeated several times 11 Results The Scenario was too difficult Therefore it was not possible to recognize the P3 component related to primary stimulus 3 1 2 Driver s attention audio and visual stimulation Task In the year 2011 three students of bachelor studies L Jan k D Gorschenek and J Reficha 9 10 11 investigated an impact of fatigue on driver s attention and prove an increase of the P3 latency These experiments had different types of stimulation There were used visual stimuli audio stimuli and combination of these stimuli The stimuli were taken away from the simulated environment They were not a part of the simulation software Visual stimulation was replaced by an external LED module audio stimuli sounds were performed by the special software This allowed better time synchronization of stimuli occurrence with the EEG signal Scenario It was possible to use a wider scale of simulation environments by outsourcing stimuli from simulation itself The selected game had better graphi
43. sed hardware This experiment was performed at the laboratory at University of West Bohemia Figure 4 11 1 which is fully equipped for measurement of ERP experiments presentation of stimuli recording of EEG ERP data data postprocessing etc The following laboratory equipment was used for implementation of the experiment High end computers One computer is used for recording and storing the EEG signal the second computer is used for the presentation of simulated environment game Another one is used for monitoring the subject by a webcam LED Module It is used for stimulation during EEG experiments It includes a PC synchronizing output an output for a panel with LEDs and a control unit which is placed in the box with the LCD display Blinking 35 of diodes is random with the option to setup individual blink intervals Diodes are placed in the cockpit of the car in front of the measured subject Devices for recording It includes an EEG electrode cap an EEG amplifier Figure 4 11 1 and a synchronising adapter Amplified signals from the cap are joined together with stimuli markers from the LED module in synchronising box and then they are transmitted to the recording PC Figure 4 11 1 BrainVision V amp amplifier Car simulator It is a trimmed Skoda Octavia II body with a Logitech G27 gaming steering wheel pedals and a gearbox installed into the original car interior Projector Simulated enviro
44. sk Scenario difficulty The performed task cannot be too difficult It is important to find a compromise between a realistic task and a task which produces the results The task stimuli should be clear simple easy to recognize and it should be easy to detect exact time of subject s reaction Time of measurement It is important to perform an experiment as long as possible but the subject still must feel comfortable otherwise the data should be pointlessly degenerated by artifacts Type of stimuli Both types of stimuli visual or auditory have some advantages and disadvantages The visual stimulation seems to be less annoying than audio stimuli also there is no need to wear headphones which can affect the measurement results otherwise the auditory stimulation shows faster reaction times of the subject Number of stimuli It is necessary to repeat stimulation of the subject to keep the ERP components active Also there is a need to define the correct interstimulus intervals 17 Sufficient number of measured subjects Data of some subjects can be inconclusive Comfort of measured subjects There is a need to set a comfort lighting seating temperature etc to eliminate artifacts 18 4 REALIZATION PART 4 1 Introduction This chapter describes design and implementation of the experiment in detail At first design of scenery of the experiment is presented It continues with a description of the ER
45. tary process directed by external events that stand out from their environment such as a bright flash a strong odour or a sudden loud noise Active attention is voluntary is guided by alertness concentration interest and needs such as curiosity and hunger Active attention also involves effort 2 2 Fatigue Fatigue is defined 1 as a decrease of the human activity which comes as a result of previous effort Physical fatigue is the inability to continue operating at level of one s normal abilities it usually becomes noticeable during heavy exercise Mental fatigue manifests in drowsiness 2 3 Electroencephalography EEG EEG is one of the methods for recording activity of the human brain This record comes as a result of activity caused by thalami and cortex neurons cerebral cortex It is represented as a change of electric potentials measured by the electrodes placed on the object s head surface Figure 2 3 1 Figure 2 3 1 Example of the electrode cap 4 Recorded brainwaves are usually rhythmic and they have a sine shape Brainwaves are divided into the following categories 5 Alpha activity e measurable mostly at the back part of the head e frequency 8 13 Hz e amplitude 30 80 uV e produced by the brain which is o healthy Alpha changes or looses with tumour trauma encephalitis etc o vigilant It disappears in the state of the coma or sleep o full blown It is regular since the age of 8 e appea
46. ttention Online visited 20 11 2011 lt http www cdl org resource library articles attention2 php gt 3 Brain Products GmbH Official patent approval for the actiCAP EEG cap approaches Online visited 20 11 2011 lt http www press releases online com Official patent approval for the acti CAP EEG cap approaches 231 html gt 4 Moucek Roman Mautner Pavel Pozornost idi e pri dvoj z t i EEG ERP experiment Online visited 12 12 2011 lt http dai fmph uniba sk events kuz2009 prispevky pdf moucek pdf gt 5 Mautner P Mou cek R Neuroinformatika metoda evokovanych potenci l ZCU Plzen Intern materi ly 2008 6 Sethi N K Sethi P K Torgovnick J Arsura E Physiological and non physiological EEG artifacts Online visited 8 1 2012 lt http www ispub com journal the internet journal of neuromonitoring volume 5 number 2 physiological and non physiological eeg artifacts html gt 7 Luck Steven J An Introduction to the Event Related Potential Technique Cambridge The MIT Press 2005 8 Krej Tomas Diploma thesis Anal za pozornosti idi e a ERP Plze 2009 9 Janak Ladislav Bachelor thesis Driver s attentionandvisual stimulation ERP experiment Pilsen 2011 10 Gorschenek David Bachelor thesis Driver s attention audio and visual stimulation ERP experiment Pilsen 2011 11 Reficha Jan Bachelor thesis Driver s Attention and Auditory Stim
47. ulation ERP Experiment Pilsen 2011 12 Lei Shengguang Welke Sebastian Roetting Matthias Driver s Mental Workload Assessment Using EEG Data in a Dual Task Paradigm Online visited 12 12 2011 lt http www nrd nhtsa dot gov pdf esv esv21 Track 2027 20Written pdf gt 47 13 Boksem Maarten A S Meijman Theo F Lorist Monicque M Effects of mental fatigue on attention An ERP study Online visited 12 12 2011 lt http www sciencedirect com science article pii S0926641005001187 gt 14 Brain Products GmbH Brain Vision Recorder User Manual Munich 2005 15 Brain Products GmbH BrainVision Analyzer 2 0 1 User Manual Munich 2009 16 Youtube Track Creating for X Motor Racing 1 Online visited 26 10 2011 lt http www youtube com watch v hxs2OP_HJuw gt 17 Exotypos X Motor Racing Online visited 29 10 2011 lt http www xmotorracing com gt 18 Skwire KeyCounter Online visited 16 3 2012 lt http skwire dcmembers com wb pages software keycounter php gt 19 NCH Software Debut Video Capture Software Online visited 16 3 2012 lt http www nchsoftware com capture index html gt 48 ATTACHMENT A USER DOCUMENTATION FOR ARTIFACT DETECTOR The application Figure A1 is started from the included Artifact Detector jar file The input file is loaded by clicking on the Load button then the standard file choosing dialog appears Artefact Detection File info
48. ve Release date August 17 2006 License Commercial Live for Speed Scawen Roberts Eric Bailey and Victor van Vlaardingen Release date July 13 2003 License Commercial X Motor racing Exotypos Release date October 14 2011 License Commercial 4 2 3 Comparison and results Live for Speed was eliminated as first It allows users to adjust only several vehicle attributes but in connection with relatively good graphics and excellent car handling Live for Speed offers best driving feel of all tested simulators However there was not any required track editor World Racing II was used for similar experiments before but it is a quite old game so the graphics environment is poor Many independent software utilities were developed for this game including the track editor object editor vehicle physics editor but they did not work properly they have also complicated controlling The best choice seems to be X Motor Racing It was chosen like simulation environment meeting most points of the requirements There is an XMR editor for 3D creating and editing the track and VehiclePhysics PRO editor for setting car physics and behaviour The game offers very good 20 graphics At first vehicle handling was not comfortable but after editing attributes the handling has improved 4 3 Construction of the track For purposes of the experiment a monotonous track with minimum of disturbing elements was created There are not any sh
49. wer Torque RPM Javia Max Engine Torque 321 N m Terquecequ Power 98 hp Palo Roster 1 avi i TaviTT Min RPM Er t i 8n ii ud s Viperi E i i Max RPM 6000 i 78 i 5 Redline RPM 4800 gr 1 1 led Ta b t 58 Seas arc off phs z FF To Engine 1 em z COPY BEL Een la E UJ PEDALS Shutdown Engine 380 RPM z 7 USAmperial 2 Tent famat update profite 4 Forced Table Update Min Value 0 00 20 Inertia ka m 2 0 013 hz ze TOT ire 0 Medus 1134620 N ryREM MaxPower 78 5140 hp RPM P 4000 8000 Ms 0 200 4000 5000 8000 10000 12000 n x 18000 RPM Dissipation Dig Don t change these fields when the game is e running with the current profile j E 1 V Engine Braking 2 RPM Dissipation Fun fap M Torque I Power o RPM Limit Losd Taie Torque vs Throttle Breking True Nmt 2 use TableData Watts hp Rey Limiter Figure 4 4 1 Vehicle physics editor 4 5 Scenario of the experiment Environment the driving task is projected by the data projector on the wall right in front of the car simulator The subject drives the scenery in highway speed about 130 kph in one lane while his her EEG is recorded Subject s behaviour is also monitored by a webcam It allows preventing some sort of problems during the experiment Electrode cap failure release of the electrodes subjects condition monitoring artifacts etc Stimulation is provided by a LED module with blinking diodes T
Download Pdf Manuals
Related Search