Mission Control!
Contents
1. In this project we tried to find out if Emotiv can actually replace mouse Or maybe we can combine it with mouse and keyboard to do normal office or other computer related work When a new technology arises it will take some time for people to learn the new technology Not all people are always open to the new technology they may get skeptical about those new things but if skeptics stop all developer or inventors from pursuing their ideas we will not have the leisure of enjoying the new technologies we are using now It is true that Emotiv is quite new and people may be skeptic about it But looking forward 10 or 15 years from now Emotiv might be fully developed useful and beneficial for the general public Specially for people with disabilities or people who are paralyzed from neck down who still have the ability to use their brain or for the working people like the geologist from Schlumberger who worked most of the time with computers using mouse and keyboard who are vulnerable to wrist or armed injuries due to extensive use of mouse and keyboard This will ease the pain if they have such injuries or prevent them from having the injuries That injury prevention will lead to less people going through sick leave because of pain It will benefit the company employee and government This project satisfies our curiosity of what if we use Emotiv instead of a mouse This project 1s still a working progress that needs more modifications Maybe2. 1 1 0 EA o o0 L1 1 28
3. at the same time which Emotiv register it as a normal detection that causes unnecessary non intended random actions e Atone point one of the participants asked us whether Emotiv emits radiation We don t have the proper facilities to test it 16 We followed the same procedure during testing for Cognitiv suite We manage to test some participants using 4 Cognitiv actions The detailed usability testing plan can be found on the appendices Appendix E and Appendix F The data collection for Cognitiv Suite is listed below Test Results Form Date 2012 10 07 Time 0944 Ri Random Random Operation Combina Right Zoom in Combina tions tions Comments x Participant 1 X X X Participant 8 Participant 9 Participant 10 x lt lt x lt lt lt j lt lt lt lt x lt lt x lt lt x lt Version PICTURE 16 Cognitiv data result As some participants just gave up in the middle of the tests as the tests took quite long time for them the data collected here were compromising result By comparing the success rates between Expressiv Suite PICTURE 15 and Cognitive Suite PICTURE 16 we can say that Expressive Suite have more potential reliability as Expressiv Suite has better results The Expressiv Suite is more reliable Based on the tests carried out above we can draw a conclusion that it is much easier to use
4. facial actions such as mouth movements smirk left smirk right smile laugh and clench and eyebrow movement raise eyebrow and furrow The eye movements do not need training The Emotiv detects the eye movements at once Using the facial expressions on any application doesn t require training but it s up to the user to train all actions if he wants to customize the actions for him her For example if a person have problem doing the clench facial actions he can customized this action by pouting his lips or biting his her lower lip Sometimes some facial expressions have strong detection signals than other facial expressions this makes it harder for a user to control the switching of actions This particular problem can be solved by decreasing the detection on the Sensitivity tab If an action has a high detection there is a high probability that the manipulation of object will be dominated by this action For example on Google earth you want to zoom in using raise eyebrow but instead it keeps on going to the left which is assigned to the facial action smile It is possible whenever you move your eyebrow your mouth move slightly at the same time Although it s just a small movement Emotiv can detect this movement To avoid this kind of problem we can decrease 9 the signal detection of action smile under sensitive panel which will lessen the random movement that action smile caused This also applies to actions with low det
5. lt lt lt lt lt lt lt lt lt lt lt lt lt lt x lt lt lt lt lt lt x lt lt lt x lt lt lt lt lt lt lt lt lt lt lt lt x lt x lt x gt x gt x gt x lt xxix lt x x lt lt lt x lt lt x xjx x x x lt lt x ox x xxx mx x mx PICTURE 15 Expressive data result During the test we encouraged the participants to say what they think and feel while doing the experiments Participant s comments during expressive suite testing e tis easy to control two expressive action e For some participants they feel discomfort after for long usage of the device and for some they feel no discomfort at all e During 4 expressive actions one user commented that it was harder to control One participant in particular said it was hard for her to make the Google earth stay sill for a few seconds It randomly moves e They get distracted doing the expressive actions especially when someone is staring at them We asked one particular participant if we could take a video of him while testing It made him really unable to do simple facial actions e They need to concentrate and try keeping their faces still which they find hard e One participant said it is hard for her to make the Google earth stay sill for a few seconds It randomly moves e When some of the participants use actions such as Look Left and Look Right they unconsciously raised their eyebrows
6. many geologists and be able to observe them We were lucky and came in contact with one professional geologist who works for a petroleum consultant company and also needs to use Petrel every day This made us able to get a first hand observation on a real working situation regarding petroleum related tasks We observed the geologist interaction with mouse keyboard actions and classified them into 6 types Click Move up Move Down Zoom In Zoom Out and Keyboard Input We observed the geologist in four different sessions which lasted about 20 minutes each This is of course a small sample size but as we already stated we were not able to observe more than one All actions have been recorded and calculated according to percentage so as to find out which action is most likely used during the whole process Keyboard Input PICTURE 13 Seismic Interpretation The data shown above clearly demonstrates to us that most of geologists works are connected with the mouse actions We had an interview with geologist and got those key information listed below e Average Project Time for Geologist 3 to 6 months e Average mouse control actions At least 70 for the whole project as far as seismic interpretation project is concerned Based on that information we could draw a tentative conclusion if we could help them to reduce the workload by using Emotiv it will be great relief for petroleum industry Usability Testing Ex
7. near the rear of the skull Next side the headset forward unsi the sensors dosest to the headset pvot ponts are located drecty above your ears PICTURE 3 Emotiv Status Panel Furthermore it displays Expressiv suite Affectiv suite and Cognitiv suite which we will all explain respectively but first we ll go through the top part of the window PICTURE 4 Top of Control Panel This is the top part of the Control Panel called EmoEngine Status Pane which will provide real time feedback regarding the system status wireless signal battery power and so on This is also where you can manage user profiles The head displayed at the far right indicates how good the signal is from each of the sensors Black No signal Red Very poor signal Orange Poor signal Yellow Fair signal and Green Good signal This top will as you can see on the next few pictures be displayed at all times Expressiv Suite The Expressiv suite has to the left an avatar mimicking the facial expressions of the user and in the middle a series of graphs indication the expression detection To the right there is sensitivity adjustment for the facial detection Zi Emotiv Control Panel 0800 HeadsetSenp Exressv Sute AffectvSute Cogniby Sute Status OK PICTURE 5 Expressiv Suite Affectiv Suite This will the area focused by the other Emotiv group Cognitiv suite The Cognitive suit us
8. neutral in between actions Task 3 Move the google earth using cognitive action move right for 3 times stopping neutral in between actions Task 4 Move the google earth using cognitive action scroll up 3 times stopping neutral in between actions Task 5 Move the google earth using cognitive action scroll down for 3 times Task 6 randomly move the google earth left right up and down following the sequence the facilitator instructs For 3 times 6 cognitive actions left right up down zoom in and zoom out Task 1 the participant shall train 2 more cognitive action With the help of the facilitator the trained actions will be assign to left right move up move down zoom in and zoom out Task 2 Move the google earth using cognitive action move leftfor a 3 times Task 3 Move the google earth using cognitive action move right for 3 times Task 4 Move the google earth using cognitive action scroll up for 3 times Task 5 Move the google earth using cognitive action scroll down for 3 times Task 6 Move the google earth using cognitive action Zoom in for 3 times Task 7 Move the google earth using cognitive action zoom out for 3 times Task 8 Randomly move the google earth left right up down zoom in and zoom out following the sequence the facilitator instructs For 3 times Appendix D Raw Data for Cognitiv and Expressiv Suits LO ations Facial Expression 3 1 i uM EG 1 EN Em
9. responsible for the bad signal Due to the time constraints and the unavailability of the users we decide to adopt Within Group mode Ideally it will be better to use two separate groups one in the Gadget Noise Place and another group as controlled group in the Noise Free Place 10 Gadget Noise Test amp Data Collection Phase Before the formal tests start we need to have data regarding those elements which could affect our tests results such as the noise on the signal different environment the shape of the head as well as the gender influences We have recruited 10 participants for our Gadget Noise Test Each participant will test Emotiv product in different environment one in Noise Place another in Noise free place It means that we will have 20 sets of data collected from those 10 participants The ideal participants should be those geologists from petroleum industry but we were not able to get that many Instead we used informatics students to help us to do the tests Below are some explanations on Noise Place and Noise free Place 1 Noise Place We intentionally put many electronic products around us and we selected Sonen as the room location because it has many computers as well as digital products The purpose is to establish an environment with as many digital products as possible 2 Noise free Place We will choose a place without or with as few as possible electronic products so as to reduce the influences on the
10. rn Emil Stensby jorn emil stensby gmail com 22 Appendix B Usability Testing Expressiv Suite facial expressions 23 Usability testing plan In this test we would like to find out how many facial expressions we can combine applying to a simple application like Google Earth Participant A group of male and female with ages ranging from 15 to 50 Participants should have computer skills preferably expert Task before testing 5 6 Screen recruit participants suitable for testing Check if the Emotive batteries are fully charged Check if all the sensors are moist Create profile for the user Prepare the questionnaire consent form Prepare the room where the testing will takes place Facilitator s task on testing day 9 Briefing session with the user Ask the user to sign the consent form Make sure the user is seated comfortably in front of the computer Help put on the Emotiv headset on users Turn on the device and adjust the headset until the signal is detected Ask the user to train facial actions Assign the actions to different mouse function using Emokey Upload the application Google Earth the user will use Implement the test 10 Help the user take off the Emotiv 11 Interview and debriefing session 24 Usability Testing Expressiv Suite facial expressions Goal In this test we would like to find out how many facial expressions can we use to integrate to an ap
11. signals We will use the same group of people for both Noise Place test and non Noise Place in order to eliminate the influences from individual differences Hypothesis for Tests HO0 There is no difference between Noise Place and Noise Free Place on the signal strength H1 There is a difference between Noise Place and Noise Free Place on the signal strength We will determine the signal strength according to the number of the different lights on the products panel as listed below 4 Emetiv Control Panel 1 0 0 5 c c E Application Connect Help PICTURE 9 Sample picture of signal strength As the picture shown above we need to test the signal strength according to the different lights appearing on the panel The signal strength can be classified into several levels from Green Yellow Orange Red and Black no signal at all 11 In order to get qualified data for the tests we will check signal strength after the participants have had the helmet on for around 5 minutes The reason the tests are quite short is to avoid fatigue among our participants Not to mention that the students we use as users are often busy and can only give us a few minutes of their time In order to clearly show differences between Noise Place and Noise free Place we created the excel data collection table Below are two examples for two different participants Emotiv Noise Tests Data No Noise Location zene
12. the google earth using expressive action move right for 3 times stopping in between actions Task4 Alternately move the google earth move left and move right for 3 times stopping in between actions 4 expressive actions left right up and down task 1 the participant shall train 2 additional expressive action With the help of the facilitator the trained actions will be assign to move left move right move up and move down Task 2 Move the googie earth using expressive action move left for 3 times stopping in between actions Task 3 Move the google earth using expressive action move right for 3 times stopping in between actions Task 4 Move the google earth using expressive action scroll up 3 timesstopping in between actions Task 5 Move the googie earth using expressive action scroll down for 3 times stopping in between actions Task 6 randomly move the google earth left right up and down following the sequence the facilitator instructs For 3 times stopping in between actions 6 expressive actions left right up down zoom in and zoom out task 1 the participant shall train 2 more expressive action With the help of the facilitator the trained actions will be assign to left right move up move down zoom in and zoom out Task 2 Move the google earth using expressive action move left for 3 times stopping in between actions Task 3 Move the googie earth using expressive action move right for 3 time st
13. when the new modified version of Emotiv EPOC we can find greater use of this device During the whole process we have learned how to gather the data how to leverage the resources from different places and we also learn some experiences from the industry company regarding the project control At the same time the project is carried out among three students from different cultural background so we have learned to cooperate with each other to have common understanding The tests did not go smoothly as we had expected due to specific unforeseen situations during the whole process We have to balance our time and resources to achieve the expected the target Creativity is not the finding of a thing but Making something out of it after it is found James Russel Lowell 1819 1891 20 References Lazar Feng og Hochheiser Research Methods in HCI 2010 Wiley ISBN 978 0 470 72337 1 Sommerville Ian Software Engineering 9th ed 2010 Pearson Education Limited ISBN 978 0 13 705346 9 Web sites http www emotiv com 21 Appendices Appendix A Informed Consent Form Consent Form for Testing I hereby give my consent for my participation in usability testing for Schlumberger mission control project created to meet the requirements for INF2260 Interaction Design I also understand The people responsible for this project and testing are Doreen Nalliasca Navidi J rn Emil Stensby and Qiang Gu under the supervis
14. Expressiv suit to carry out actions than Cognitiv suit This is why we will adopt Expressiv suit for our software development test in the next step Emotiv integration to Petrel The usability test shows that Expressiv suite facial expression gives more positive result than Cognitiv suite conscious thoughts Schlumberger then asked us to make a simple program using the Emotiv Soft Development Kit The purpose for this program is to integrate it with Schlumberger s Petrel software so we can communicate with Petrel by using the headset The program was mainly to check if it was able to communicate with the EmoEngine which then communicates with the headset without the need of assistance from the EmoKey program As Petrel is a commercialized product it needs highly strict requirements from the developers to develop software plug in for it The software developed by us has been divided into two stages 1 Initial Software Plug in As we cannot use Petrel on our laptop due to the limitation we have developed the software and tested on our own laptop to see whether it can realize some function as we expected 2 Second Version Software Development After we have tested the software we sent the source code to Schlumberger people they modified the software so as to connect the software with Petrel It turned out to be good version software Here we will list out the screenshot for the software 17 E S Smirklet 4 gt Smirk right Ra
15. Mission Control INF2260 Human Computer Interaction Bachelor Group Doreen Nalliasca Navidi Qiang Gu and Jgrn Emil Stensby Contents Jinngoniifenfo rp e EUER 3 Backsround MNT 3 Project Group Introduction rere ete Red eruat e Urea De Ee F d Ei LEE Race BL Ea 3 Project e rubis EM 3 Project walkthrough eee eite e rcc dete obese edet Log resp tieni bung k epu ut geeu page delt 3 Client Introduction orones teg ea PEE pu ee EAR PEL vR eeu eek E 3 Product Introduction aeree daira e d ret idera pe ERE e Eta se eee eia idera epa doe iE Mee Gea ea pit 4 OVEr VIEW T a 4 Installation and basic setup irt erede Lite Eee Pero ede ape epe sedare ae et aede vaa 5 Emotry Sn E 5 I So RAIDER M 6 Cosiitiy Sue M 7 Project Schedule c 8 Dues V M E 8 Descriptive exploration seek nitet erp EHE EHE ERU He CE Ea ee LER a ce aee ep H dE E AED ODE inng 8 Experimental Research ir ce ier te cipue ig pecie Exe E URP cen EUER Pe EFE He EE ERAN TEE 10 Gadget Noise Test amp Data Collection Phase eese 11 Ethnographic 8s nino EE 14 Usability Testing Expressive Suite facial expressions and Cognitive Suite 14 Emotiv integration to Petre ce cescesecssecesecesece
16. e actions left and right or up and down 4 cognitive actions left right up and down 6 cognitive actions left right up down zoom in and zoom out 2 cognitive actions move left and move right scenario the participant shall use to cognitive action ex Raise eyebrow and burrow assigned to 2 movement scroll down and scroll up With this experiment we shall use google earth The participant will try to move the globe using the cognitive action Equivalent value for task completed 1out3 task completed X 2 out of 3 task completed v 3 out of 3 task completed v Task Task 1 After setting up and getting signal detected The participant shall train 2 cognitive action With the help of the facilitator the trained actions will be assign to move left and move right Task 2 Move the google earth using cognitive action move left for 3 times stopping neutral in between actions Task 3 Move the google earth using cognitive action move right for 3 times stopping neutral in between actions Task4 Alternately move the google earth move left and move right for 3 times stopping neutral 4 cognitive actions left right up and down task 1 the participant shall train 2 additional cognitive action With the help of the facilitator the trained actions will be assign to move left move right move up and move down Task 2 Move the google earth using cognitive action move left for 3 times stopping
17. ection Sometimes actions don t get registered at all No matter how much you raise your eyebrows the application like Google earth doesn t react at all This is due to low signal detection of facial expression raise eyebrow The solution is to increase the detection on the Sensitivity tab Training the Cognitiv suite conscious thoughts There are two ways to train Cognitiv suite The first method is to train each action individually You can train each action until it reach up to 100 Depending on different factors sometimes reaching 100 on each action takes only 8 minutes while other time it will take quite a long time even only 80 control on the action The problem with this method of training is switching or controlling each action Controlling skill requires long training time The second method is to train all the actions at the same time The Emotiv allows user to train maximum of four actions at a time Training four actions at one time and trying to reach 100 for all actions is harder but the positive thing is that you can train your action controlling skill at the same time This method takes a long time to achieve expected result but can give more positive feedback From picture below we can see that tester has successfully got 100 control on the actions PICTURE 8 All training 100 96 Experimental Research We conduct experimental research because we want to identify if the electronic gadget is
18. es a 3D representation of a cube to demonstrate the Cognitiv detection There is three different tabs here Action Training and Advanced See picture 6 below VES Emotiv Control Panel 08 00 k k ie Application Connect Hep PICTURE 6 Cognitiv Suite The action tab is where the current state of the actions and the detection status Here you are able to add or remove actions one action is e g push You can also see how skill full you are in doing different actions and your overall skill rating The training tab is where you train each action The EmoEngine analyzes your brainwaves and develops an own person Hence the ability to have different user profiles Two different persons can have one user profile each with their own training and the software will remember it so one does not need to train the same action the next time one will use the software The advance tab is for customizing the behavior of the Cognitive detection The user manual does not recommend normal user to adjust these settings Project Schedule After the discussion with the customers we proposed a tentative timeline for our project In order to track the project process we created a Gantt chart Sommerville Ian 2010 to list out our expected target for each period The advantage for the adoption of the Gantt chart is that it can illustrate the start and finish dates of the terminal elements and summary elements of a project Those ele
19. hen decided to test facial expressions on Google Earth Users were asked to use the Emotiv with facial expressions to control the movements of Google Earth We mapped each action to keys corresponding Google Earth s interaction like move up and move down We decided to do three usability testing based on facial expression and we conducted the test according to the following scenarios e 2expressive actions up and down e 4 expressive actions up down left and right e 6 expressive actions up down left right zoom in and zoom out We first conducted a pilot test on each usability test to find out if the test is harmful or hard for the participants and whether it is time consuming We also tried to minimize the errors and problems we might encounter during the test The picture below shows one of the pilot testing we conducted PICTURE 14 Description Before the test implementation we asked each participant to sign a consent form that states that they understand what the test was all about and that they were willing to participate The consent form contained the groups contact details The first usability test applies to the first scenario which is two facial expressions We asked participants to use two facial expressions to manipulate Google earth The participants were then asked to follow the instruction of the test facilitator 15 Following the task on our usability test design the facilitator asked the participants to tra
20. in and performed the actions three times with pauses in between The pauses showed us if the actions were triggered and intended by the participants and not just random movements If the participants managed to do 0 to I action this was considered unsuccessful task and 2 to 3 actions are considered successful task The first test didn t take a long time to implement We then politely asked the participants if they will be willing to do second test which was the next scenario 4 facial expressions The same procedure was followed for the second test After the second test since it is hard to find willing participants we once again politely asked the same participants to do the last test on facial expressions We are thankful that all participants who participated on the first two tests graciously allowed us to do last test This last test was conducted the same manner as the first two test The only difference was the number of facial expressions with corresponding actions the participants use and do Below is the data collection table V Success X Fail Test Results Form 2012 10 07 Time 0944 1 2 Random Hm Random Operation Left Right Combination Left Right Down Up Zoom in Zoom Of Combination Combinations 5 Comments V X s Participant 1 V V V V V V Participant2 V Participant 3 lt lt lt lt lt lt lt lt lt lt lt lt lt lt lt lt
21. ion of Alma Leora Culen I understand that the testing will focus on the usability of Emotiv EPOC device I understand that the testing will include A pre testing demographic survey 5 minutes Usability testing tasks Max 10 minutes During the usability testing I will be observed and the testing will be recorded with notes I understand my information will be kept anonymous and that under no circumstances will I be identified by name or other characteristics when Doreen Navidi J rn Emil Stensby and Qiang Gu report the results of this study I understand only Doreen Nalliasca Navidi J rn Emil Stensby and Qiang Gu will have access to the non anonymous records and data collected with this study and that all data will be strictly confidential I understand that I may discontinue this study at any time I choose without penalty I understand that Doreen Nalliasca Navidi J rn Emil Stensby and Qiang Gu has agreed to answer any questions I may have about their research the acquisition of data or their procedures and that I may contact Alma Leora Culen with any further questions or concerns C Yes I m willing to participate in this usability testing research CI No I do not feel comfortable participating in usability testing research Name and Signature Contact Information For more Information on this study feel free to contact me Doreen Nalliasca Navidi doreennn ifi uio no Qiang Gu qiangg C student mat uio no J
22. ise brow gt Furrow brow Look left gt Lock right Smile Clench Awaiting calibration of gyro please hold head still for 0 5 seconds No smirk info PICTURE 17 GyroUse The source code of this program can be found in the appendix In order to test the effectiveness of the software with Petrel we asked the testers to use facial expressions to control Petrel The main actions are Move Right Move Left Zoom In Zoom Out Up and Down We carry out series tests for our software plug in in order to perfect the software as well as for better control on Emotiv Test one video on Petrel PICTURE 18 Sample from the video Link https docs google com file d OB3bTM3mhLsftaWpIY VFoaFJTc00 edit From this video we can see that after several minutes successful control on Petrel with the help of the developed software the tester herself cannot easily control her facial expressions The same results also can be found during the Facial Usability test process for the participants Final test video both on Google Earth and Petrel 18 PICTURE 20 Petrel Software Testing Video Links https docs google com file d OB3bTM3mhLsftQW xBbUI2TnhaQjg edit In this video we asked the tester to find Oslo Norway As you can see from the video the tester successfully carries out those operations according to our expectations Due to the business restrictions we are not able to use Petrel sof
23. lored the Emotiv device to find out how effective or viable Emotiv is Each member trained and used the different facial and cognitive actions on Emotiv The following are the summary of our exploration Good signal is the key for Emotiv to get the best result Signals affect the use of different actions When you re using the Cognitiv or Expressiv suite it is harder to switch from one action to other actions When you have low signals the actions sometimes don t get registered due to poor detection It is also harder for a user to train a particular action Among the three classes Expressiv Affectiv and Cognitiv only Expressiv and Cognitiv contain actions that can be trained Before you can use the Emotiv fully you are required to train neutral state and a few actions Training neutral actions is necessary so that a particular brain signal pertaining to a particular action can be compared to the neutral signals Training neutral a user needs to be in a relax state especially with expressive actions The user should avoid twitching while training the neutral state or Expressiv suite The Twitching can be detected as neutral state For instance if you manage to smirk while training when you do smirk it will be registered as neutral and will not be registered as smirk Training the expressive suite facial expressions The Expressiv suite have a universal roster of actions for any user to use but expressive suite has a training functions for some
24. ments comprise the work breakdown structure of the project The Gantt chart also shows clear relationships between activities We can track current schedule status using percent complete shadings A B c D E F G J K L M N o P Q RIS T U V W X Y Z AA AB AC AD 2 Schlumberger Emotiv Project Plan Oslo University amp Schlumberger 3 2012 09 4 Planned Actes Show Gantt for Planned Show Status gt 2012 9 13 2 2012 98 2012 9 17 2 List of Activties cen Start Dur Start Dur Done 2012 9 13 ermet s sespe 4e sieme siie semmi se ee sineme deeem dem nemen deem nem nm me 2012 10 2012 9 18 2012 9 18 4 2012 9 21 2012 9 18 4 2012 9 20 2012 9 21 2012 9 21 2012 9 24 j amp Exec 1 Act 2012 9 27 15 8 ent th and perform act 2012 10 1 mmm u V yl Uy J PICTURE 7 Project plan Research methods We have conducted the following methods Informal interview descriptive exploration experimental research and usability testing In the following we will explain how we contacted and implemented each of the methods Descriptive exploration First impression In our initial use of the Emotiv device we encounter a big problem at once which is how to make it properly work The manual stated that all sensors must physically touch the scalp or skin and because most of us have thick and long hair this gave us a challenge with mounting the headset In the span of approximately two hours the seven
25. n Gender Female Time z Comments Long hair PICTURE 10 No Noise Place Emotiv Noise Tests Data Noise Time ET Comments PICTURE 11 Noise Place As we can see that we have also made comments regarding the gender hair and sensor In order not to make the tests data analysis quite complicated here we will not take those extra elements into considerations as the tests are mainly focusing on the signal influences from digital products In order to analyze the data we have assigned below numbers to those different color lights Green 4 Yellow 3 Orange 2 Red 1 Black 1 The total of those numbers can be used as the quality measurement for signal strength The bigger the total number is the better the quality is We multiply the number of greens to the corresponding points As we have highlighted in the No Noise Place in PICTURE 4 the user get 14 green 56 points 1 yellow 3 points and 1 orange 2 points This equals to the total points of 61 We have collected the data from 10 participants with below summary table provided 12 No Noise Place Total Points Creen Yellow Orange Red Black Noise Place Noise Place a E Total Points Green Yellow orange Red Black Oo el 4 52 36 en 2 858 0 of 9 PICTURE 12 Noise test summary table Here we start to calculate the data like below No Noise Test Data Total Points 561 Average Points 56 1 Noise Test Data Total Point
26. of the l amp electroderecesses are Jfttec with amast elt pad IF the pads are not akeady moist met them with saine soluton before jrsering inta the headet or alierratively use a medicine dropper to carefidly moisten the pads hie already in place Step 2 Swtch on the Eroty headset and verify hat the bult n battery s charged anc is provding power by locking fcr the bue LED located near the power switch at the back of the headset If the headset battery needs charging then set the power switch to the off position and pug the headset irto the Emotiv battery charger using the mri JSE cable provided with the jneadeet Alon the headset battery to charge for at east 15 mutes before sving agan Step 3 Venty that the Wireless Signal reception is reported as Good by looking at the Engne Sas box n the Emoty Control Perel If itis rot make sure that he Eriotv Done is rsen led nto a USB port on your computer ard that the inge LED on the top half cf the donde i on continuously ar fictering very ragidiv 2f the LED s birkng slony ar is not luminated then remove the dongle fron the computer reinsert it and try again Remove any metalic or dense physcal obstructions located near the donde or the headset and nove away from any poser sources of ekecromagnetc interference such as microwave ovens or hgh ponered adn Ivorcmiters Step 4 Put on the Erot headset ay genty puling apart the headband and lowering the sensor ams onto your head fom the too down
27. of us alternately tried to make the Emotiv work Unfortunately none of us manage to make Emotive EPOC work This we found weird because the headset Hakeem had which is the Emotiv Researcher headset we got working on two of us After this experience we all wondered if the room Sonen we were staying at was one factor that could influence the Emotiv We thought that all the gadgets inside Sonen interfered with the signal detection of Emotiv As to Petrel it is professional software for petroleum industry The setup and use of the software will take quite long time due to the special requirements for this professional software People from SLB help us to go install the software on SLB workstation and guide us through some basic functions for this software Due to the special requirements for this 8 software we cannot install Petrel on our own laptop and try it at home or in the university It means that we will have limited time on learning or use this software so we will limit our research target to some basic functions under this situation Emotiv training Since Emotiv is just almost decade young technology we decided to find out all about Emotiv We would like to find answers for questions like Does it really work How does it work Do Emotiv rely on the software or does it need people skill to be able to make it work and How long does it take a user to train and start using it Each member of control group tried and exp
28. opping in between actions Task 4 Move the google earth using expressive action scroll up for 3 times stopping in between actions Task 5 Move the google earth using expressive action scroll down for 3 times stopping in between actions Task 6 Move the google earth using expressive action Zoom in for 3 times stopping in between actions Task 7 Move the googie earth using expressive action zoom out for 3 times stopping in between actions Task 8 Randomly move the google earth left right up down zoom in and zoom out following the sequence the facilitator instructs For 3 times stopping in between actions 25 Appendix C Usability Testing Cognitiv Suite conscious thoughts 26 Usability Testing Cognitiv Suite conscious thoughts Goal In this test we would like to find out how many cognitive actions can we use to integrate to an application ex Google Earth how many actions can we control and If cognitive actions is more accurate and viable than expressive actions facial expressions Participants This usability testing will have atleast 10 participants Male and female age ranging from 15 to 50 The participants must have a little knowledge with computer particularly on using computer mouse The participant s responsibilities are to try to complete the tasks given to them and to provide feedback about the usability of the device Scenarios Each scenarios is a independent usability testing 2cognitiv
29. plication ex Google Earth how many actions can we control and If facial expression is more accurate and viable than cognitiv Participants This usability testing will have atleast 10 participants Male and female age ranging from 15 to 50 The participants must have a little knowledge with computer particularly on using computer mouse The participant s responsibilities are to try to complete the tasks given to them and to provide feedback about the usability of the device Scenarios Each scenarios is a independent usability testing 2expressive actions up and down A4expressive actions up down left and right 6 expressive actions up down left right zoom in and zoom out 2 expressive actions move left and move right scenario The participant shall use two expressive action ex Raise eyebrow and burrow assigned to 2 movement scroll down and scroll up With this experiment we shall use google earth The participant will try to move the globe using the expressive action Equivalent value for task completed D 10ut3taskcompleted X 2outof3task completed v 3outof3task completed v Task task 1 After setting up and getting signa detected The participant shall train 2 expressive action With the help of the facilitator the trained actions will be assign to move left and move right Task 2 Move the google earth using expressive action move left for 3 times stopping in between actions Task 3 Move
30. pressive Suite facial expressions and Cognitive Suite We use usability testing to test Expressive Suite and Cognitive Suite on Emotive to find out if those suits are feasible enough to integrate into software Because this is high fidelity commercialized product we decide to adopt summative test We chose to use Google earth because the interaction use on Google earth and Petrel software s model part is almost the same According to the project requirements we re supposed to have atleast10 participants for the test We manage to gather a total of 10 participants that were willing to use and test the Emotiv 14 We tried to get both gender and different age bracket We know that to represent the general population we need to have representation from different groups such as disable people We know that using people with physical disabilities requires contacting the appropriate institution for participants Due to limitation and time constraints including unavailability of the participants we recruited the students as convenience sampling for the project In this test we would like to find out e How many facial expressions cognitive actions can we use to integrate to an application Google Earth e How many actions can we control e If facial expression is more accurate and viable than cognitive and vice versa After learning that Emotiv has three classes of detection which we use to map up different actions to different keystrokes We t
31. rel and two using Emotiv information to quantify user satisfaction The project will start with the analysis and implementation of basic 3D scene interaction and later will evolve into more elaborate workflows and interactions From the project description Project walkthrough In the beginning of this project we had brainstorming on how could we tackle this project and we gather questions that will pave way for our progress The following are the questions regarding this project 1 What is Emotiv and how does it work 2 Is it possible for this device to replace the mouse 3 Whatis Petrel software 4 What features of Emotiv can we use to integrate with Petrel 5 How can we integrate Emotiv with Petrel After we explore the first question more questions pop up that lead us to the progression of this project Client Introduction Schlumberger is the leading oilfield services provider in the world The company provides solutions for oil and gas companies and has offices around the world The company aims to develop products services and solutions that optimize customer performances In Norway Schlumberger has research centers both in Oslo and Stavanger Our project is carried out in Oslo office 3 Petrel is a Schlumberger owned Windows PC software application It is used to aggregate oil reservoir data from multiple sources Users can use this application to interpret seismic data perform well correlation build reservoir model
32. s 519 Average Points 51 9 If all the participants achieve all green lights then the maximum average points should be 64 Based on this information we can get below percentage number for each group No Noise Percentage 56 1 100 64 87 6 Noise Percentage 51 9 100 64 2 81 196 When we compare those two sets data we can see that participants can get better signal in the no noise place The test was carried out at the beginning of our project When we conducted this test we were not aware that the moisture level for the signal detector on Emotiv has significant influences on the test At first we had thought that it was good enough to just put several drops of Saline solution on the signal detector but we found later if we add enough drops on the detector we can get all green lights for the signal detection The sample for the participants is small and the environment is not strictly noise free as we carry out the test in the university building Ideally the sample size should be large and the test room should be completely noise free laboratory compared with noise place 13 Ethnographic Observation Our target users are experts who have many years experiences in IT industry They have very strong knowledge on the software development and IT technology The products which we are to use also are very professional ones The more observation we have the better insight we can get The challenge is how to get in contact with
33. s suitable for simulation submit and visualize simulation results calculate volumes produce maps and design development strategies to maximize reservoir exploitation Schlumberger aims to build this software into a single application portal which will reduce the need of multitude of highly specialized tools and can control the lifecycle of the petroleum data Our project is based on two products Emotiv EPOC and Petrel The first one is a combination of software and hardware product while the second one is the software application Emotiv EPOC is a device which can detect and measure brain waves It is a high fidelity Brain Computer Interface BCI which reads and interpreters both conscious and unconscious thoughts from the users After meeting with people from Schlumberger we have some information listed and those listed information are what people in Schlumberger expect us to do with Emotiv and Petrel 1 Is it feasible to use Emotiv in practical areas in our daily life 2 How far can we integrate Emotiv device with Petrel software 3 Is it possible to use Emotiv to control some programs We need to test the performance of the Emotiv with quantifiable results According to those listed requests and the available testing products our research project is based on high fidelity prototype or we can say that it is based on finished commercial product Product Introduction Overview Emotiv EPOC is a device which can detect and mea
34. seeeseeesneeeseeeseeesaecaaecsaecsaecsaecsaeesseeeeeeeseeeseeeees 17 Nuntia senecio ieaie e aE EEE EEEa E EE N TEN EEA E EE EN a ENE 20 Va 001 116 1 60 EE EE EE EA EAEE EEA ETE 22 Appendix A Informed Consent Form ssrernvnrvvarenrnevnnenvnsnvnenvnsnvnnnvnnsnnnrnenenrnsvraseresevessnsssnensnvnennesnnee 22 Consent Form for Testing oe eese Ere eS ETE RA HEP Ce ED ER LE EE eT LEE Tee p LH ERR RP EE ka ladon 22 Appendix B Usability Testing Expressiv Suite facial expressions eese 23 Appendix C Usability Testing Cognitiv Suite conscious thoughts esee 26 Appendix D Raw Data for Cognitiv and Expressiv Suits esses 28 Introduction Background Project Group Introduction Mission Control Project group has three members Doreen Qiang and J rn Emil The team members are from different cultural background Philippine China and Norway We are bachelor students from University of Oslo studying Informatics Design Use and Interaction Since we all have the basic year here at UiO for interaction design we have some knowledge on human computer interaction design As the team members are from different background it will give us different insights into this project Project Assignment This project involves the analysis of Emotiv information as to improve human computer interaction in two areas one on replacing or complementing mouse interaction with emphasis on its use on Pet
35. sor Units Saline solution and Battery Charger The USB transceiver will connect the headset with the computer The Hydration Sensor pack is where the sensors will be stores when they are not connected to the headset The saline solution is used to moistening the felt in front of each sensor and it also disinfect As to Petrel it is professional software for petroleum industry The setup and use of the software will take quite long time due to the special requirements for this professional software People from SLB help us to go install the software on SLB workstation and guide us through some basic functions for this software Due to the special requirements for this software we cannot install Petrel on our own laptop and try it at home or in the university It means that we will have limited time on learning or use this software so we will limit our research target to some basic functions under this situation Emotiv Control Panel The Emotiv Control Panel is one of the user interfaces and the one that controls and connects with both EmoComposer and EmoEngine We ll come back to this The first one will see is the Headset setup picture 4 This will provide the user with a headset setup guide which basically just is guidance regarding how to fit the neuroheadset on the head As you can see on the picture below are all the sensors green which indicates very good signal Step 1 Before putting on the Enotv headset ensure that each
36. sure brain waves using 16 sensor units See picture 1 Each arm has a sensor at the end which is a sensor unit It is as previously stated a high fidelity Brain Computer Interface BCI which reads and interpreters both conscious and unconscious thoughts from the users g 3 e PICTURE 1 Headset placement Petrel is a Schlumberger owned Windows PC software application It is used to aggregate oil reservoir data from multiple sources Users can use this application to interpret seismic data perform well correlation build reservoir models suitable for simulation submit and visualize 4 simulation results calculate volumes produce maps and design development strategies to maximize reservoir exploitation Schlumberger aims to build this software into a single application portal which will reduce the need of multitude of highly specialized tools and can control the life cycle of the petroleum data Installation and basic setup The installation of the Emotiv software was pretty straightforward and we did not encounter any problems at all We installed it only on Windows After reading the user manual we started by moistening the sensor knobs and charging the batteries which is also easy and straightforward Picture 2 has all the items in the EPOC Headset Kit PICTURE 2 Overview of equipment From the top left Neuroheadset USB Transceiver CD Installation disk Windows only Hydration Sensor Pack with 16 sen
37. tware outside Schlumberger office during the whole process We take advantage to use Petrel software when we are in Schlumberger office The Google Earth controlling was done through EmoKey a subprogram of Emotiv Control Panel which allows users to assign different actions both Expressiv and Cognitiv to a specific keystrokes on the keyboard The people from Schlumberger modified this program to integrate it to the model part of the Petrel software The video from above was taken at one of the testing sessions we conducted during our stay at Schlumberger During this test the tester show confusion when the facilitator asked her to do some actions She executed actions opposite of what the facilitator was asking her to do Fatigue could also be the reason for her lack of concentration But overall result was a success despite the fact 19 that the software itself needs modification We can say that this program is still work in progress Summary Our project is about Emotiv EPOC as mentioned earlier it is a Brain computer device that allows user to control an object like computer or even wheelchair We worked together with Schlumberger which is one of the leading service providers for the petroleum industries Initially Emotiv EPOC is developed for gaming purposes Emotiv developer wanted to give users a different gaming experience But does it really end there or can Emotiv have more possibilities This curiosity leads to this project
Download Pdf Manuals
Related Search