Perceptual elements affecting navigation of players in first person games

Bachelor of Science May 2019

Perceptual elements affecting navigation

of players in first person games

Leon Viio

This thesis was submitted to the Faculty of Computing at Blekinge Institute of Technology in partial fulfilment of the requirements for the degree of bachelor of science in digital games development. The thesis was equivalent to 15 weeks of full time studies.

The authors declare that they are the sole authors of this thesis and that they have not used any sources other than those listed in the bibliography and identified as references. They further declare that they have not submitted this thesis at any other institution to obtain a degree.

Contact Information: Author(s):

Leon Viio

E-mail: levi16@student.bth.se

University advisor: Title Petar Jercic

Department of Computer Science

Faculty of Computing Internet : www.bth.se

Abstract

Context. This thesis’ goal was to test whether or not perceptual elements which do not intrude on player movement were reliable methods of leading the player through an environment without specifically telling them where to go, focusing on players already accustomed to first person perspective(FPP) games. This was a remake of an older experiment in the hopes of finding different results than last time it was tested, by making improvements to the test chambers, and altering the application of these different elements which were used to test the participants navigation. Objectives. In the user study, the player was forced to pick between two mirrored pathways with a specific variable distinguishing them in the hopes that the players will follow a predictable pattern. By proving that none heads up display(HUD) ele-ments were reliable tools for player navigation, it could introduce new ideas for level designers and artists working within the industry to guide player pathing in a first person perspective game.

Methods. This experiment was constructed using different soft-wares commonly used in the game industry, and was made to be easily repeatable by other who would like to expand on the subject. References used were mostly scientific articles which could be found using google scholar, ACM library and researchgate. The articles in question were sourced by their use as references and the validity was assessed by their abstract, discussion thought process and own references used.

Results. Using 17 different students from Blekinge institute of technology as partic-ipants in three different test scenarios where they had to navigate their surroundings by choosing between one of two pathways. The results show promising indications for further research in the subject, with the large majority of people picking the predicted path.

Conclusions.While the data collected by the participants show promising results in favor of reliable means of navigation through the environmental effects used in the experiment, there were many altering factors at play which potentially could have altered the results. The topic of player navigation in games are a widely contested area with many factors at play. More research is necessary to reach a more conclusive answer, but shows great potential judging by the results.

Keywords: Games, Navigation, Perceptual elements, first person perspective.

Acknowledgments

I would like to acknowledge my supervisor Petar Jercic - lecturer, who helped me with the structure and details concerning the thesis.

My fellow students who helped me conduct the experiment, giving me tips and mental fortitude throughout the process of creating the experiment and writing the thesis. Erik Wadstein - technical artist, who made the original thesis which this was based upon and that made me inspired to continue research within the field.

Contents

Abstract 1

Acknowledgments 3

1 Introduction 7

1.1 Problem identification and alterations from previous works . . . 8

1.2 Aim and objectives . . . 8

1.3 Research Questions . . . 8

2 Related Work 11 3 Method 13 3.1 Perceptive elements used in the experiment . . . 13

3.2 Creation of the experiment: . . . 13

3.3 Creation and motivation for the test chambers: . . . 14

3.3.1 Base layout: . . . 14

3.3.2 Test Chamber 1 - Lighting . . . 15

3.3.3 Test Chamber 2 - Texture . . . 16

3.3.4 Test Chamber 3 - Model . . . 17

3.4 Ethical Consideration . . . 17 3.5 Consent form: . . . 18 3.6 Experiment limitation: . . . 19 3.7 Participants: . . . 20 3.8 Experiment: . . . 20 3.9 Risk management: . . . 20 3.10 Expected outcomes: . . . 20 4 Results 23 4.1 Test chamber 1 Lighting: . . . 23

4.2 Test chamber 2: . . . 24

4.3 Test chamber 3: . . . 26

5 Analysis and Discussion 29

6 Conclusions and Future Work 33

7 References 35

8 Appendix 37

Chapter 1

Introduction

This thesis was about finding new ways to navigate player pathing in first person games, using perceptual elements. First person refers to the perspective the camera uses when operating the game, which in this case means that the camera operates as the eyes of the player, trying to simulate what every person sees from their own perspective in real life by rendering the scene from their perspective. Perception is the baseline assessment of information and works instinctively, unlike cognition, which uses the perceptive information and further evaluates it[6]. Cognition is the process of using memory and previous knowledge to judge, use problem solving, mak-ing decisions, reason and evaluate information[6]. This process, unlike perception is not instinctive and takes a certain amount of time to assess the information at hand. The goal in the creation of the test chambers were therefor to create environmental elements which instinctively direct the player towards a certain path without them necessarily knowing why. This was a continuation of previous work done within the area[1] where the results of their experiment differs from the expectations and contradicts similar research done in the field[5]. In the previous experiment(PE)[1], using different textures, models and lights, the goal was to guide the player and predict their pathing. This was done by making the participants walk through dif-ferent environments and let them pick one out of two difdif-ferent paths. The results of that experiment contradicted the expected outcome which could be explained by the specific implementation of these different elements, which was the reason for re-vising the subject using a differing method. Every game that gives the players the option of free movement to some extent needs to make it clear where and how the player was supposed to traverse the world he/she operates in. When navigating a 3D game environment, it is not uncommon to make use of the in-game HUD which acts as an overlapping interface menu that conveys relevant information to the player, such as their health, stamina, inventory among other things depending on the game. Examples of some navigational HUD elements consists of a map of the surrounding environment or virtual arrows which point the player in the intended direction. This research could in theory help immerse the player in the world more and allow new aspects of level design by removing these navigational crutches.

8 Chapter 1. Introduction

1.1

Problem identification and alterations from

pre-vious works

Textures were clearly demonstrated to work in the PE which this was based on, but at the same time, a reason it worked could be because the textures used in that experiment mimicked a sign that specifically told the player what to do instead of emphasising differences in the two available paths which was was what done in this user study.

Lighting done in the PE was done in a similar fashion to the way it was handled in this. The difference being the paths themselves. In the PE, both pathways were completely straight from the players starting point, meaning they saw the end of both paths and knew what they were to expect when picking one or the other. Pathways were curved in this experiment in order to not show the ending of the chosen path. This was done in order to make the lighter path in theory look more inviting. The use of models in the PE were used in order to block of the player and make the path harder to navigate through, this proved to be an inefficient way of navigating the player as the results showed no clear difference between the two paths. The way models were used in this user study was not to block the players movements, but to introduce a less inviting environment in the form of water, something people in general tend not to want to walk through.

Another choice made in this experiment was to not let the players backpedal when a path was chosen, as a door locks the players into their choice when they advance far enough. This was made to make the participants decisions matter in their mind. The environments also change on a regular basis in the PE depending on the test at hand. This was also something that was changed in this experiment where the rooms and paths were identical from test to test with the one outlier being the perceptual el-ement that was currently being tested in order to reduce potential risks to the results.

1.2

Aim and objectives

The aim of this thesis was to introduce new tools for level designers in FPS games to navigate the players through a FPP game. The actual point of this experiment was not necessarily to advocate for the removal of navigational tools through HUD elements, but to introduce different tools for game developers to use within level design. This was done using perceptive elements that in theory should guide the player to their objective without them being consciously aware as to why. This could prove as a useful tool in the creation of less linear and restricted levels in the game, as it would allow there to be more open space while at the same time having a reliable way for the level designers to guide the players to their destination.

1.3

Research Questions

1.3. Research Questions 9 RQ2: Through perceptual means, in this case: textures, 3D models and lighting be an altering factor in player behavior and be used as a predictor for their pathing when faced with two alternatives?

Chapter 2

Related Work

Work related to player navigation in games were encouraged at a Game Developer’s Conference[11] where similarities between the level design of a game and the con-struction of theme park attractions were mentioned as an example of how to direct the guests / players towards attractions and objectives. Since research within the field at the time was scarce, a better understanding, better documented experiments and user studies would provide more means of level designers to create environments the players move through. After that, more scientific papers have been published where they investigate new and reliable means of directing the players towards their goals. The work which this experiment was based on, both regarding the setup of the test chambers and the elements used to test player behavior were inspired by a study made by Erik Wadstein[1], where the expected outcome of the player pathing did not match the results. According to other studies done within the field, the expectations laid out in that experiment were logical and in theory should match the outcome. This user study borrows aspects but implements them differently to try and change the outcome to better resemble the expectations.

A study made by Burigat et al.[2] aimed to compare three different navigational aids in games to guide the player by emphasising different locations and objectives while also comparing the efficiency of the three navigational tools. Whether or not it takes longer for the player to reach their destination using tool A, B or C in a user study of 48 people divided into two groups. The three navigational elements used were 2D and 3D arrows which pointed the player to their objective and a radar which showed the location of the objective.

This experiment’s aims were similar but the execution of the trials and use of navi-gational tools differ wildly.

Another work related to this subject was made by Moura et al.[5] where the aim was to guide the player towards a certain objective in different environments. The objective changes depending on the test chamber they were in, ranging from finding a hidden path behind a waterfall, finding a lever hidden on top of platform high into the air and backpedaling to find the correct way in a maze. They used lights to highlight the objective in a mostly somber and dark environment to differentiate it from the surroundings, using player pickups to indicate the possibility of climbing up platforms and visual cues like vines to give the player a logical indication that it is possible to climb up tall walls. This study showed varying results, but served as a great inspiration for the implementation of the elements used in this experiment. Their implementation of lights to highlight a desired objective showed promising re-sults, which contradicted the results in the experiment made by Wadstein[1], which

12 Chapter 2. Related Work was the main reason it was tested again in this user study.

Decisions made regarding the implementation of the elements in the different test chambers stem from the path of least resistance, which acts as a metaphorical pathway for objects when faced with different alternatives. A common example of this is the way the water flows along a surface.

Chapter 3

Method

3.1

Perceptive elements used in the experiment

Textures:

2D images used to represent the surface of a 3D model. Textures differ depending on what kind of surface the designers want to convey. Common examples of textures used in the game industry are: Diffuse/albedo maps which specify the color value of the 3D model. Normal maps introduce different levels of surface curves on the model which is useful for faking complexity. Roughness maps are used to specify the roughness and smoothness of a surface.

Lighting:

Lighting is used in the game industry to convey different atmospheres and moods. Lighting is also used to create shadows and emphasise detail on complex geometry through the surface normals. Another common use of light is to point out important parts of a game environment or the player’s objective by highlighting it with a light source[5].

3D Models:

3D Models refer to meshes that consist of vertices, faces and edges that are stitched together to represent a polyhedral object in 3D graphics which is often used to represent the tangible objects in the game.

3.2

Creation of the experiment:

The foundational design aspects of this experiment was made to replicate the exper-iment in the thesis by Erik Wadstein[1] since this was a continuation of that work to some extent, but with a few differences that potentially could change the results. The baseline structure of the experiment was to introduce two different paths for the player to pick between. The two paths differ from each other in regards to one perceptive element. When the player enters one path, the entrance of that path closes off to enforce the choice and reduce backpedaling. Both paths lead to identical next choice scenarios without the player’s knowledge to make the player think their choice mattered. The result of their navigation was recorded by automatically creat-ing and writcreat-ing a .txt file that correlates to the individual participants when a path was chosen and the player collided with a trigger box at the entrance of each tun-nel. Another trigger box was located at the ends of the tunnels which automatically

14 Chapter 3. Method restarted another scene in a random order. The participants were given in total of three choices of paths that test a certain perceptive element presented in a random order in order to reduce faulty data. The player movement within the experiment worked like the standard FPP game controllers on a desktop computer. Using WASD to move around and the mouse for looking around. The experiment was constructed using the free student version of Unity(Unity Technologies. 2017) where the scripts correlated to player movement and recording of the navigational results were made using Visual Studio 2018(Microsoft. 2018) using C# as programming language. The assets used in the experiment were constructed using a combination of different soft-wares. Maya 2018(Autodesk 2018) was used to construct all the 3D models besides the water one, which was made using the standard asset pack which comes along with Unity. The textures used was created using Substance Painter (Allegorithmic. 2018). The consent form was made in collaboration with Blekinge Institute of Tech-nology following a standard template for user studies to make sure the experiment was ethical in nature and performed in an orderly manner, which the participants had to read and agree to before their inclusion in the experiment (The consent form can be found in the appendix part of the thesis). After the participants chosen paths were gathered, the data was analysed using an online frequency distribution calcula-tor to generate the mean and standard deviation to be used for a normal distribution bell curve, where the probability of the results were calculated and then compared to the expectations for each test.

3.3

Creation and motivation for the test chambers:

3.3.1

Base layout:

3.3. Creation and motivation for the test chambers: 15

Figure 3.1: Base model of test chamber

3.3.2

Test Chamber 1 - Lighting

This test chamber’s only difference from the others was its use of light. Because the tunnels were curved, it introduces an aspect of the unknown, not being able to see the end result of the path chosen could in theory act as a great motivator to take the path of least resistance, the more inviting path. Darkness is often seen as a negative, representing fear of the unknown, often represented as such in the horror genre, both in games and movies, which in theory would make the brighter lit path more inviting for players.

16 Chapter 3. Method

Figure 3.2: Test chamber with different light setups

3.3.3

Test Chamber 2 - Texture

Test number two two different path and entrance textures. One path (right) has the same texture used in the other tests, using a concrete material for the walls of the tunnel and a dark tar material for the floor. The other path (left) uses a more modern and maintained texture, closely resembling plastic with a metallic beehive pattern around it for the walls of the tunnel with an almost teflon metal as floor. The texture on the entrance mesh was also changed to more closely represent the cleaner aesthetic of the tunnel. The purpose for these choices was to differentiate the textured path from the dirtier older look the rest of the room and tunnel entail. The hypothesis was that a cleaner and different environment would be more intriguing and inviting for a player.

3.4. Ethical Consideration 17

3.3.4

Test Chamber 3 - Model

Test number three uses the base layout of the experiment, both paths having the same properties and light setup, with the sole difference being a square mesh without collision being laid out on the floor of the left tunnel with a panning normal and albedo texture to imitate water. The assets being used are part of the Unity standard asset pack included in the creation of a new project. That includes the material of water, the script necessary for reflective and refractive properties of water, and the shader included to create the panning X-Y coordinates of the textures. The thought behind the water rather than other objects was to find a 3D model that does not interfere with player movement. As shown in the PE, geometry acting as an obstacle do not prove to be an effective mean of altering player behavior. Water on the other hand is merely an inconvenience, something that do not require walking around of jumping above, the consequences of water means merely slowing the player down or the act of swimming. By having mirrored pathway tunnels, it creates an expectation of the floor level being equal and should in theory make it obvious that the water level was at ankle height and not be an obstacle which requires swimming or another type of navigation that differs from the alternative path.

Figure 3.4: Test chamber with water 3D model

3.4

Ethical Consideration

Performing a user study means actively involving outside people in the experiment, people who might be ill informed about the study, the risks and their responsibilities. Because of this, it is important to be transparent and mitigate and preferably elim-inate any aspect of the experiment which could risk the mental and physical health of the participant. Ethical considerations can be separated into different categories which were handled throughout the experiment[12].

• Purpose of the research:

18 Chapter 3. Method • Offer to withdraw at any time for any reason:

• Potential benefits to participant or society: • Potential risks or discomforts to participant: • How privacy will be protected:

• Who administrator was and how to reach them:

Many of these points were stated in the consent form, how their privacy was con-cerned and handled, the name and email of the administrator for the experiment and being able to leave and withdraw from the experiment at any time being ar-guably the most important aspect. The participants were verbally informed of the time the experiment would take to complete. The experiment was low risk in nature, taking place in a room with a laptop and mouse being the only tools used and the only foreseeable damage that could come from the participant was them navigating a 3D environment and looking around, which could result in motion sickness related discomforts and nausea from people not experienced in the area. This risk was con-trolled by only allowing the participants to take part if they already have previous experience in FPP games before. The participant was mostly kept in the dark re-garding the aim of the user study to mitigate risks of them altering their behavior to further the goal of the study. They were instructed to pick the right path in order to simulate a real game scenario where they have some goal or quest to follow. The publication of the thesis under Blekinge Institute of Technology and that the results would be publicly available after the completion was verbally stated when asked, and implied in the consent form through the data handling explanation and bth email. Potential benefits to the participant and society was not communicated to the participant, as it would inform them of the aim of the research and could not be shared after their participation, since they could reveal it to future participants, as they were all students picked at random.

The ethical elements regarding this experiment was discussed and approved by the supervisor of the thesis before the user study commenced.

3.5

Consent form:

3.6. Experiment limitation: 19

3.6

Experiment limitation:

Because of time management, the number of perceptual elements which can be stud-ied during the user study was limited both because of the increased length of carrying out the experiment, but also because of the creation of the assets used within the experiment. There were a number of things that could alter the results of the exper-iment. A few examples of these include but are not limited to:

• Participant’s previous experience with FPP games. If the player was not famil-iar with FPS games controller’s or feel could influence their stress level, cause nausea or disorientation during the experiment. This was controlled by making the participants read the consent form before taking part in the application of the experiment. The consent form specifies that in order to be a part, they have to have had previous experience in FPP games.

• The location and setting of the experiment could be an altering variable be-tween the participants which could affect the way they operate within the experiment possibly because of differing lighting setups depending on the day or their level of tiredness depending on the time. This was easily fixed by not altering the location and time frame of the experiment, which includes the PC which the experiment takes place on and having the experiment’s time frame limited to a couple hours on that specific day when the user study was conducted.

• The hardware which was used in the experiment. If the hardware differs from experiment to experiment it could affect different aspects of the user study. If the screen shows different hue values, the screen updates at different rates, the frames per second differs or the response rate of the keyboard used for moving around was different from each participant to the other, could alter the results gathered and could play a part in differences which were unaccounted for in the data. This was solved by making the application of the user study take place on the same computer which uses the same mouse and keyboard. The computer used in the experiment was a HP Pavilion 14-al003ng, using its standard monitor and keyboard with a Steelseries Rival 310 mouse.

• Variables that could alter the participants choice of navigation. By making the paths in the experiment mirrored in every way besides a certain aspect depending on the perceptive element that was being tested at that time, it limits the amount of information that could influence the player.

• Left or right tendencies, if the player specifically or humans in general have a tendency for picking either left or right, it could alter the results and provide faulty data that either confirms or rejects the variable being tested.

20 Chapter 3. Method

3.7

Participants:

The test was conducted by students in Blekinge Institute of Technology that willingly agree to the consent form that contain information that states that their inclusion in the experiment was held completely anonymously, their test results were recorded and used in the writing of this thesis and that their participation was voluntary and that they could exit the experiment at any time.

3.8

Experiment:

The experiment took place in Blekinge Institute of Technology and the participants were students picked at random who were willing to take part in the tests. They were asked to come to the room where the experiment took place, and were instructed to read the consent form already in full view on the computer screen. The participants were asked to find the correct way and no further instructions. After agreeing to the consent form, the participants were asked to complete their assigned task of the experiment, and the administrator started the application. The computer used to run the experiment and collect the data were a HP Pavilion 14-al003ng, using an Intel Core i5-6200U processor, an NVIDIA GeForce 940MX GPU and 8GB of RAM. This achieved 47 - 52 frames per second when running the experiment made in Unity. The data were collected automatically as the participant progressed through the test.

3.9

Risk management:

The main risks involved in the experiment were faulty data. That the participants are not in their right state of mind or that outside forces change the result of the collected data. The main point of this thesis was to explore perceptive aspects of level design to influence the participants choices of navigation and if they take extra time to consider their options when choosing a path, could result in overthinking, none instinctual behavior which would not represent realistic player behavior in the game, and be outlying data. This problem could be resolved by limiting the time the players have to choose where they are headed, but was this aspect was overlooked because of the implementation of these elements, making them as subtle as possible. The time limit to finish the experiment and writing of the thesis was also a risk, if things took longer than they were calculated to. Plan B would be to reduce the amount of perceptive elements being tested in the experiment, which not only would reduce the time it takes to complete the Unity scene the project takes part in, but it also reduces the time it takes for participants to complete the experiment. This could also increase the amount of participants that take part, which would increase the validity of the result.

3.10

Expected outcomes:

3.10. Expected outcomes: 21 proof based on the data collected that supports this hypothesis. The reason the ex-pectations for the user study to produce positive results was because of the changes done from the PE.

The expected outcome behind the test using different lighting schemes inside the two paths was that the user would pick the one with a light source, because of the reasons stated in Creation and motivation for the test chambers. The expectation was that the user would pick the predicted path - path two.

The expectations behind the test using different textures on the wall and the en-trance was that the player would be intrigued and chose the path with altered tex-tures because of the reasons stated in Creation and motivation for the test chambers and chose the left path - path one.

Chapter 4

Results

This part of the thesis was going through the data collected and analysing it to find out whether it was likely that there was a reliable way of guiding the player towards their destination without implicitly telling them or giving hints on where to go.

4.1

Test chamber 1 Lighting:

The results showed that out of 17 participants, 15 people chose the path with a more inviting light setup. 11.76% (2 / 17) of the participants chose the left path while the remaining 88.23% (15/17) chose the left path.

24 Chapter 4. Results Using frequency table calculator to calculate the normal distribution of the re-sults, using 0 and 1 as numerical values representing path one and two respectively, it showed that with a mean of 0.882, and a standard deviation of 0.322, the Area(probability) = 0.0031 when tested below 0 which was used as a numerical value representing the left (none lit) path. A probability value <= 0.05 indicates a strong possibility of certainty and can be qualified to be reliable.

Figure 4.2: Normal distribution showing probability area for test 1

4.2

Test chamber 2:

4.2. Test chamber 2: 25

Figure 4.3: Chart showing test chamber 2 pathing distribution

Using the same frequency distribution calculator as before, with the numerical values 1 and 0 representing the left and right path respectively. The mean was calculated to be 0.823, the standard deviation 0.381 and calculating the area below 0 to get the area(probability percentage) of 0.0154. A probability value <= 0.05 indicates a strong possibility of certainty and can be qualified to be reliable.

26 Chapter 4. Results

4.3

Test chamber 3:

Out of 17 participants, 5 people chose the left path, while the remaining 12 people chose the right path. This means 29.41% (5/17) of people chose the left path with water in the tunnel, while 70.58% (12/17) people chose the right path without water.

Figure 4.5: Chart showing test chamber 3 pathing distribution

4.3. Test chamber 3: 27

Figure 4.6: Normal distribution showing probability area for test 3

All and all, the tests imply a strong indication that there was a correlation be-tween the environmental effects and player pathing. Both test one and test two showed a probability factor < 0.05, or five percent when tested, meaning there is, with a statistical certainty a strong possibility of being true, while test three showed a high probability of it not being by chance also.

28 Chapter 4. Results

Chapter 5

Analysis and Discussion

While the results of the tests conducted were promising, with a high probability of supporting the research questions, this does not mean there was a 100% chance of the research conducted in this user study being absolute. The limitations in the research shows flaws in the execution and possible problems which could alter the data for better or worse. While people had to agree that they had previous first person experience in order to take part in the experiment do not mean they cannot lie or misunderstand the statement. While the purpose for making the participants say they have previous experience in FPP games was to limit risks involving the par-ticipants like nausea, disorientation or motion sickness, it also creates a level playing field for the participants, a baseline from which one could measure the experiment from that sole perspective. If there were inherent behavioural differences in people who were used to or have some experience in FPP games from people who do not, this could alter the data when tested on another group of people.

Left and right navigational tendencies were a hot topic, with research showing different takes and results from it. In a study from 2015[7], it shows that students have a tendency of picking a seat on the left side of the classroom, while another study show that right handed people and americans tend to favor turning to the right[8]. The tendencies of left to right was debatable at best and researchers try to interpret the data differently, ranging from left and right handed correlation or social norms like traffic regulations as an example. A test to see whether or not the participants had such tendencies by making a test chamber without any perceptual elements tested by using the base meshes and materials could be a solution for this, but could psychologically alter their behavior in the tests that actually test these variables. An example of this would be if they feel like they were going left too many times. This was documented as a pattern where the participant will not feel comfortable making the same decision in extended matching item choices, which was used in the creation of multiple choice tests[9]. Another experiment which could fix the tendency problem would be to duplicate the tests with opposite direction variables as shown in figure 5.1.

30 Chapter 5. Analysis and Discussion

Figure 5.1: Mirrored variable placement

The reasons this kind of test was not done in the experiment was because it also could introduce faulty data. The theory behind it was that after being introduced to one path environment and then facing that same environment again, the participant would be more inclined to take the other path, which they skipped the first time encountering that specific test chamber. For example. Being introduced to the light-ing focused test chamber and picklight-ing the brightly lit path the first time, and then being introduced to the flipped version later, knowing the consequences of taking the brightly lit path and out of curiosity taking the other path the second time around. This could be solved by making another test group of the same amount of people and redoing the entire experiment but with flipped variable locations. The reasons this was not done was because the tendencies of picking a certain direction are not conclusively proven to not be entirely individualistic.

The changes made in this test that differs from the last experiment were made with the hope of minimising faulty results and making the results more resemble the predicted outcome. The results collected from this test show that the variables being tested in the test chambers with a high likelihood alter the pathing of the participants. This do not mean the data was without fault or interpretation, a few examples of other interpretations of the data are:

Regarding RQ1 and RQ2, it is safe to say that the data collected in the experiment support the research questions that there were reliable perceptive means to lead the player towards a certain path when faced with two different alternatives using perceptual elements in the environment.

31 choice based on that information, either to further the agenda of the experiment or to sabotage it.

• Because the probability area of the normal distribution does not reach zero, there was the chance that the results look like they do completely by accident. • While the implementation of these elements were made with intent and thought behind them, there was nothing saying there are not other ways to implement these elements in other ways that could result in better navigational tools for developers. The thought behind this, was that it is uncertain whether these navigational elements were underutilized or if they would work at all in different scenarios.

• There could be other perceptual elements which could alter player pathing to a greater or lesser extent than those used in this thesis. Examples of some would be sound cues, animations, particles or some form of implementation of shadows.

• The results were excluding any other game setup besides FPP, it is possible that the pathing would differ widely in a game using top-down, virtual reality, third person or side scrolling perspectives.

Chapter 6

Conclusions and Future Work

While many variables could be in effect to vary results, the data indicates with high certainty that certain perceptual elements in the environment could be useful tools in the future for game and level designers when creating games and specific levels. The researach question: Through perceptual means, in this case: textures, 3D models and lighting be an altering factor in player behavior and be used as a predictor for their pathing when faced with two alternatives? Was answered and confirmed from the data collected through the user study.

According to the data collected in this experiment - the conclusion whether or not perceptual elements in a FPP game environment could be used as reliable tools of player navigation is leaning heavily in favor of yes. This data is only certain regard-ing the textural and lightregard-ing implementation recorded in the test chambers, while it is worth mentioning that the 3D model used also showed promising results, Further research regarding the implementation of these elements is still an unknown, while other potential tools like particles or sound effects are still unclear.

For future research in the subject, testing other perceptual elements would be the first thing, in this test, only textures, lights and 3D models were used as elements, but games often use many different types of environmental effects to create their levels. Animations, sounds and particles were elements that could have been included to increase the amount of tests and would be a great starting point to further research within this area. Trying the same tests but using other devices or perspectives and see whether or not the FPP aspect of this test was necessary for the results would be another thing which could be worth exploring. Making the participants fill out a questionnaire which includes individual aspects like age, gender, left or right hand preference and then comparing those to their results, would be a way to further test whether those aspects had a correlation to their navigational choices could be an-other aspect worth exploring. While the data collected show promising results, it is clear that there are many factors at play which could potentially influence the player behaviour and indicates the complexity of this type of experiment. More research should be done in the field to pinpoint these circumstances and eliminate the varying factors at play to reach a solid conclusion.

Chapter 7

References

1. Erik Wadstein. Artistic Techniques to Influence Navigational Behavior in 3D-Games. Blekinge Institute of Technology. 03/05, 2013

2. Stefano Burigat and Luca Chittaro. Navigation in 3D virtual environments: Effects of user experience and location-pointing navigation aids. International Journal of Human-Computer Studies, 65(11):945 – 958, 2007.

3. Veronica Sundstedt. Perception and Cognition. Blekinge Institute of Technol-ogy. 2019

4. David Milam and Magy Seif El Nasr. Design Patterns to Guide Player Move-ment in 3D Games. School of Interactive Arts and Technology July 28 - 29, 2010.

5. Dinara Moura and Lyn Bartram. Investigating players’ responses to wayfinding cues in 3D video games. School of Interactive Art and Technology. April 26 -May 01, 2014.

6. Douglas Bernstein. Essentials of Psychology. Pp 123-124. 25/4, 2011.

7. Victoria L. Harms, Lisa J. O. Poon, Austen K.Smith, Lorin J. Elias. Take your seats: leftward asymmetry in classroom seating choice. Frontier in Human Neuroscience. Aug 17. 2005.

8. Scharine A. A., McBeath M. K. (2002). Right-handers and Americans favor turning to the right. Hum. Factors 44, 248–256. 10.1518/0018720024497916 9. Mohammed O Al-Rukban. Guidelines for the construction of multiple choice

questions tests. J Family Community Med. PMC. Sep-Dec 2006.

10. Toby Tan. (2008). Color Psychology in Football: The Effect of Shirt Color on a Team’s Performance in the Dutch Eredivisie. Erasmus University Rotterdam. 11. ROGERS, S. Everything I Learned About Level Design I Learned from Dis-neyland presented at the Game Developers Conference - Game Design Track, San Francisco, CA, 2009

12. Burgess, R. G. (Ed.).The ethics of educational research (Vol. 8). Psychology Press. 1989.

Chapter 8

Appendix

Figure 8.1: Consent form used in the study