Log in / Register
Home arrow Computer Science arrow Augmented and Virtual Reality
< Prev   CONTENTS   Next >

3.4 Emotional and Game-play Data Processing

During the experiment three pre-processed emotional signals – excitement, frustration and engagement/boredom – were recorded using Emotiv Epoc device. Values of the three signals varied from zero to one. If excitement, frustration and engagement were low, the value was close to zero and if they were high, the values of parameters were close to one. The signals were recorded with the sampling period of T0=0.5 s.

The coordinates of a spaceship two times per second and the exact time of the col-

lision were recorded as well. Two derivative parameters were calculated and added after the experiment. Using the coordinates of the spaceship, manoeuvring speed and acceleration of a spaceship were calculated and later used in the statistical analysis.

SPSS 13.0 software was used for statistical analysis. Nonparametric correlations and criteria for two related samples comparison were used because of the small sample size. Matlab software was used for processing the required data for later statistical analysis.

Relations between personality traits, gaming experience, emotional signals and spaceship manoeuvring speed as well as acceleration were calculated using the data from the second (two asteroids per second were generated) and the third (ten asteroids per second were generated) stage of the game.

There was more specific analysis performed, using excitement, frustration, and spaceship manoeuvring speed and acceleration signal intervals of 14.5 s around the collision action (seven seconds before the collision and seven seconds after the collision). Interval of seven seconds before the collision was chosen as an asteroid becomes visible on the monitor screen six-seven seconds before the possible collision. The same length interval was taken after the collision as the reaction time to visual and audible stimuli for every person is different.

These intervals were selected from each volunteer data, but the statistical analysis was performed for the whole group at once. Only single collisions in the selected intervals were analysed. There were lots of cases where several collisions were present in the 14.5 s length interval, but they were not taken into the analysis as there is a need of larger amounts of data for more complex analysis.

Fig. 4 shows all the single intervals of frustration signal around the collision for one of the volunteers in the experiment Stage II. Vertical dotted lines denote the collision moment and the time is set to zero at this point. Fig. 5 shows all the single intervals of frustration signal around the collision for the same volunteer in the experiment Stage III. It can be seen from the figures that the signals behave differently. To analyse the data in detail the intervals of 14.5 s around the collision were divided into three smaller intervals. The first sub-interval included signal values from -7 s to -2.5 s before the collision; the second sub-interval included the values from -2 s to 1.5 s – the close collision environment and the third included values from 2 s to 7 s after a collision. Such sub-intervals were selected based on the fact that the collision starts to become unavoidable 2 s before a collision and taking into account that it takes some time for the sound and visual collision effects to disappear. Maximal values from each sub-interval were calculated for excitement, frustration, manoeuvring speed, and acceleration signals to analyse the changes in the sub-intervals that is to investigate if there are significant changes in the signals before the collision, during it and after it. Engagement/boredom signals were not analysed this way as they did not vary in the vicinity of the collisions.

Fig. 4. Samples of frustration signal around the collision in the experiment stage 2. Vertical dotted lines denote the collision moment.

Fig. 5. Samples of frustration signal around the collision in the experiment stage 3. Vertical dotted lines denote the collision moment.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science