Monday, 20 February 2012

2012-02-20

Last Friday, I finished the final presentation. This week which is also the last week, I should focus on the report. The report should cover all what I've done in the past 12 weeks, and discussions about the result.

Friday, 10 February 2012

2012-02-10

Up to now, I have finished almost all the data analysis and the results are to be reviewed after preparing the presentation. Some of the results are not significant as expected, and I am trying to think about the reasons. It may cause by learning effect or other factors like gender, age and so on.

I plan to focus on editing the report from next Monday and hope I can finish it before the 24th. The final presentation will be on next Friday.

Thursday, 2 February 2012

2012-02-02

I am now doing the ANOVA test to three main features -- percentage of higher than upper bounds, average pressure and number of overcorrection. As Excel cannot do the one-way repeated measures ANOVA,  I am learning to use SPSS.

From the book, I understand that I should firstly draw some graphs like frequency histograms to get an overview of the data before doing the intensive analysis.

I still have some problems in data normalization. I am still not sure what type of data needs normalization and which formula is the best.

Monday, 30 January 2012

2012-01-27

I am now focusing on analysing the data. Besides the Python scripts used in the pilot study, I added some more feature analysis, like overcorrection.

From simple t-test and graph in Excel, only some of them had significant p-value, and the graph didn't contain any trends.The results from SPSS (only for the feature percentage of higher than upper bounds) showed that the Estimated Marginal Means has a clear trend, the cognitive load increased from Level 0 to Level 3. Next week, I will continue the analysing and start to edit some parts of the report.

Wednesday, 18 January 2012

2012-01-17

Today, I finished experiments for 7 subjects, including Session 1 and Session 2. I added a question today about whether they often use tablet, so that it can be taken into consideration in the data analysis.

Currently, I have to enter the subjective feedback data into computer and then combined files of User 0 correctly.

Monday, 16 January 2012

2012-01-16

The actual experiment started last Friday (January 13th) and continued this week. After the Session 1 of the first subject, the program was modified to include different repeats instead of manually starting the program after every sections. Up to now, I've finished the Session 1 for four subjects, and I am on the schedule. Hopefully, the experiment will be finished at the end of this week.

Tuesday, 10 January 2012

2012-01-09

For the past few days, I was doing the pilot study, which include the procedure of experiment, user interface and difficult levels. I also made some changes to the experiment according to the data analysis result. In order to increase the difficulty, I designed another two versions of secondary task -- distinguishing between even and odd number of tones and using flash effect of border instead of playing beeps. The above two versions will be tested within this week, and I think I can use more than one in the final experiment.

Friday, 6 January 2012

2012-01-05

I've been working on the script for data analysis of the pilot study for two days. Up to now, I finish the computations of average pressure (for the whole experiment), average pressure for different levels, percentage of time that out of bounds (0.3 - 0.5) and correctness of tone counting task answer. During the testing, I also improved the format of the logging file, adding some fields to facilitate the analysis.

Wednesday, 4 January 2012

2012-01-03

Today is the first day after the long break, and I think the most important for me is to catch up the schedule. Firstly, I started to read the recommended article about designing experiments to learn more theories. Then, I reviewed the codes and modified some small parts. In the meeting with the supervisor, I got a clearer plan for this week and some ideas of data analysis (for the pilot study).

Wednesday, 14 December 2011

2011-12-14

Yesterday, I run a set of pilots about monitor angle, how to hold the electronic pencil and the time. I gave them two to three minutes as training session, and then asked them to do the pressure control. I collected 15 groups of the data and some valuable feedback. However, the problem for me is I am not sure what to do with those data. Should I analyse them in details? Or  mainly consider the subjective feeling and feedback? What is more, they think with the larger angle (inclining from horizontal), the task is harder. But for me, I have to decide whether to make it harder or easier.

Tuesday, 13 December 2011

2011-12-13

In the last few days, I focused on the preparation of presentation in the group meeting. Actually, during the preparation, I concluded some current unsolved problems which has to be checked in the pilot. I found that I also have to do some more researches about whether the kinesthetic processing is under the central executive. Yesterday, I finished the implementation of tone counting task and add a confirm button on the interface.

Today, I'll do the pilot of tablet angle, pen angle and duration.

Wednesday, 7 December 2011

2011-12-06

Today, I learned some details of java.util.logging, understanding it's main components and basic usages. I also discussed my designing with Ronnie and Natalie to clarify some questions, especially whether my experiments can increase the cognitive load to the expected levels. They gave some suggestions about what data to collect in the pilot study.

I also edited a draft of my presentation next week and refactored the experiment code. I think tomorrow I should finish the draft slide and understand the working principle of "listener".

Monday, 5 December 2011

2011-12-05

Today, I attended the regular Monday group meeting in the morning. I think what Dan talked about in his presentation made me rethink about the validity of my experiments, particularly the difficult levels. In my experiments, the primary task remains on the same level, and the reason is that it may be very hard if I shrink the range further down -- but still needs to be proved. Meanwhile, the secondary task has three levels. In the pilot study, I should focus on analyzing the data and to find evidence that participants indeed feel Level 3 is harder than Level 2.

I also organized the steps of each tasks and wrote a detailed description of my experiments containing screenshots of UI and timeline of the experiment. It is in CVS.

2011-11-30

Today, I finished implementing the first version of my primary and secondary tasks.

There are still some points to modify, especially the logging part. The original way in my program to write to a file has two main disadvantages. Firstly, it is hard to separate different participants' data, which increases difficulties in data analysis. Then, open and close files too many times waste time and decreases the speed.

Another important point to think about is how to give feedback of pen pressure to the participants.

Tuesday, 29 November 2011

2011-11-29

Today, I continued to implementing the experiments. Now, I finish a simple version of primary task, and the UI of the secondary task.

As I haven't got much experience in implementation of user interface, I learned how to use AWT and Java Swing for almost one day. The interface of my secondary task is not complicated, which has 20 radio buttons and two ordinary buttons for the subject to select the sum number of tones.

Additionally, Ronnie suggested use Java Swing instead of AWT. I think the reason maybe Swing is lightweight and independent of platforms.

Monday, 28 November 2011

2011-11-28

Last Friday, after the meeting with the supervisor, I decided to implement the experiment measuring static pen pressure and the secondary task once used by Gerard P. van Galen in "Error, stress and the role of neuromotor noise in space oriented behavior". So today, I started to use JTablet API to implement the primary experiment. Up to now, there are still some problems unsolved. I plan to finish it before Wednesday, and do some pilot testing within the group.

Friday, 25 November 2011

Getting started

Totally, there will be 13 weeks for me to do this project. Ronnie and Natalie have made a clear schedule for me to follow and at the end of the February, I am expected to hand in a report and do a presentation to demonstrate my work to NICTA.

The whole process mainly contains reviewing literature, designing the experiment, conducting the experiment, collecting and analyzing data and writing the report. Currently, I am reading published articles.They helped me to understand some fundamental concepts of working memory and cognitive load. Particularly, one of them summarized three different methods to measure cognitive load and it gave me basic ideas of designing my experiment.

Yesterday, I focused on searching for more articles about experiments to prove relations between pen pressure and memory load, which are close to my topic. Besides that, I kept drawing sketches for different designing ideas.