Virtual reality and effort-based decision making

Virtual Reality and Effort-Based Decision Making

Name: Amber Astara de Wit

Student number: 10357920

Subject: Bachelor Scriptie

Teacher: dr. Jasper Winkel

Index

Abstract

3

1.Introduction

4

1.1 Work, Money, Pleasure: A study in the Interaction of Effort-Based

Decision-Making, Financial Status, Anhedonia

4

1.2 The Influence of Depression on Sense of Presence and EBDM: How

Virtual Reality Can Influence Our Idea of Effort

9

2 Methods an Materials

13

2.1 Main Question 1: Work, Money, Pleasure

13

2.1.1 Participants

13

2.1.2 The Decision Task: POI

13

2.1.3 Striatal DA: Listening span

17

2.1.4 Questionnaires

18

2.2 Main Question 2: Effort

19

2.2.1 Questionnaires

19

2.3 Procedure

20

3 Results

22

3.1 Main Question 1: Work, Money, Pleasure

22

3.2 Main Question 2

25

4. Discussion

33

4.1 Main Question 1

33

4.2 Main question 2

35

Literatuur

38

Appendices

41

Abstract

Over the last years virtual reality has become a tool to conduct psychological research. However, up to this date it has not extensively been used to study effort-based decision-making, which represents the making of a choice based on its costs and rewards.

This current study hypothesizes that participants will have a higher sense of presence in a virtual reality environment compared to a normal 2D or 3D environment. And further hypothesizes that due to the higher sense of presence, the effort that has to be conducted

in a effort-based decision-making task is perceived as more intense. Therefore the participants will choose the low effort option more often than the high effort option. This study will evaluate the effect of anhedonia, the lack of-, or diminished interest, and ones financial situation on effort-based decision-making as well. Where it is predicted that a high

amount of anhedonia and a high financial situation will lead to a less amount of effort conducted, compared to a low amount of anhedonia and a low financial status. Twenty

participants conducted a decision task in three different settings (2D, 3D and virtual reality). Participants reported a higher sense of presence and higher perceived effort in the

virtual reality setting compared to the 3D and 2D setting. However this did not influence participants choices in choosing the low or high effort condition more often. Furthermore there were no differences between anhedonic and non-anhedonic participants or between

participants of different financial situations. This could mean that although participants perceived presence and effort as higher, that this was not enough to influence their

“What destroys a man more quickly than to work, think and feel without inner necessity, without any deep personal desire, without pleasure - as a mere automaton of duty?” ― Friedrich Nietzsche, The Anti-Christ, Ecce Homo, Twilight of the Idols & Other Writings

1.Introduction

1.1 Work, Money, Pleasure: A study in the Interaction of Effort-Based Decision-Making, Financial Status, Anhedonia

Since the beginning of time it’s been crucial for survival to make the right decisions. Our ancestors had to decide whether to fight the giant mountain lion that threatened their camp, or flee into the higher grounds and hope that the mountain lion would go away. Nowadays, not all decisions are those of life and death, but they still impact our daily lives. The question is, what makes us decide whether we fight or flee, whether we get up and get ready for work or press the snooze button one more time and risk getting late?

It has been proposed (Doya, 2008) that to make a decision a person has to take four steps. The first step is the recognition of the current state or situation the person is in. The second step is the evaluation of the action options and their possible reward or

punishment. The third step is selecting the preferable action. And the last step is the reevaluation of the action based on the outcome. Looking at decision making, there are several kinds of decision making, one of which is Effort-Based Decision-Making (EBDM). This portrays the making of a decision between two rewards, for which the higher rewards needs a greater effort and the lower reward a lower amount of effort. In everyday life we all make choices based on EBDM, for example: one can choose to go to an expensive

restaurant and enjoy a nice and nutritious meal, or one take the less costly choice and hop into a fastfood joint and eat a not so nutritious burger. In psychological research more and more research is done in EBDM. A fine example for the kind of tasks that are being used is

given by Treadway, Bossaller, Shelton, & Zald. (2013), who use the Effort-Expenditure for Rewards Task. In this task participants are given two possible outcomes, one smaller($1) and one larger(ranging from $1.24 to $4.30). For the smaller reward the participants have to complete an easy task, which is the pressing of a button, with the dominant index finger for 30 times in seven seconds. The hard task, which will result in the higher reward, needs the participant to press the button with their non-dominant pinky finger, a 100 times in 21 seconds. For each trial a probability rate(12%, 50% or 88%) is given for the easy, as well as for the hard task. It is up to the participant to choose between the hard or easy task.

Heyman and Ariely (2004) discussed the standard model of labor based on Fiske’s (1992) relational model. They propose that there is a difference in the relationship between effort and payment, depending on if the situation is monetary or social. If the payoff is social (e.g. no payoff, or in gift form) the amount of effort is independent of the payoff, meaning that there won’t be a rise in effort, given a higher payoff. However, when the payoff is monetary, the amount of effort will increase given a higher payoff. Looking at EBDM one can expect that individuals will choose the high effort task more often when given a higher monetary payoff contrary to a low effort task with a lower monetary payoff. One question is whether individuals with different amounts of income will all choose the higher effort/higher reward option. It appears that when an individual’s basic income is high (e.g. $45.000 a year) this person will be less satisfied when given a bonus of $1.000 compared to an individual whose income is low (e.g. $16.000 a year), this is a concept named diminishing marginal utility (Easterlin, 2005). It is possible that an individual with a higher income is less motivated to work harder in an EBDM task because the high reward is not satisfactory enough compared to the amount of work the individual has to perform. Conversely, the individual with a lower income is more motivated to choose the high effort choice, because the high reward is worth the high effort.

The theory described in the above makes it seem that individuals with a lower income are more motivated to work harder, but it appears that people with a lower income are at higher risk for multiple mental problems like schizophrenia and depression as well (Saraceno, Levav, & Kohn, 2005). Depression is an mental illness with at least five out of the nine following symptoms: Depressed mood, diminished interest or pleasure

(anhedonia), significant weight loss or gain, insomnia or hypersomnia, psychomotor

agitation or retardation, fatigue or loss of energy, feelings of worthlessness or excessive or inappropriate guilt, diminished ability to think or concentrate, recurrent thoughts of

death/suicidal ideation ((American Psychiatric Association, 2013). At least one of the first two symptoms, depressed mood or anhedonia, has to be present to be diagnosed with depression. Anhedonia is a lack of interest or motivation in rewards (Der-Avakian & Markou, 2012). Research shows that depressed individuals or individuals with anhedonia show less effort in multiple EBDM tasks (Forbes, May, Siegle, Ladouceur, Ryan, Carter, & Dahl, 2006). It is possible that depressed individuals don’t have the motivation to work for an higher reward(Treadway, Buckholtz, Schartzman, Lambert, & Zald, 2009).

But what causes this drop in the effort in EBDM? It’s theorized that dopamine plays an important part in EBDM in rats (Bardgett, Depenbrok, Downs & Green, 2009). Floresco, Tse, & Ghods-Sharifi (2008) showed how dopamine blockage or increase could change the amount of effort done by rats. When DA was blocked rats showed less effort, when there was a medium increase in DA the rats showed higher effort. And finally when there was a high increase in DA the rats showed less effort, which gives an inverted U-shape. It even appears that individual differences in Dopamine also have an influence on EBDM in humans (Treadway, Buckholtz, Cowan, Woodward, Ansari, Baldwin, Schwartzman, Kessler & Zald, 2012). In their research they showed that there was a positive correlation in the DA activity in the striatum and ventral medial prefrontal cortex and the the amount of times an individual chooses the high effort/high reward option. This shows that a higher

amount of DA activity in some way influences EBDM, but in what way? According to Berridge and Robinson(2003) there are different components of reward namely ‘learning’, ‘wanting’ and ‘liking’, all with their own brain mechanisms. The ‘liking’ of a reward is

hedonic response to a reward whereas ‘wanting’ is the motivational response to a reward. It seems that there is a distinction in forms of anhedonia as well, namely the

consummatory anhedonia, which means the decreased liking of a reward, and the anticipatory anhedonia, which means a decreased wanting of the reward (Gard, Gard, Kring, & John, 2006). But what causes this diminished liking or wanting in anhedonia? According to Der Avakian and Markou (2012) the ventral striatum plays an important part in the experience of pleasure as well in the motivation to receive rewards. Looking at the relation between anhedonia, DA and reward, it seems there is a negative correlation between DA activity in the nucleus accumbens in respons to monetary reward with

individuals with anhedonia (Wacker, Dillon & Pizzagalli, 2009). This makes it possible that individuals with anhedonia have a lowered DA activity in the ventral striatum, which in turn causes the lower effort in EBDM tasks. Sherdell, Waugh and Gotilib (2012) showed that especially anticipatory anhedonia can predict the amount effort done in individuals. In their study they showed that depressed and non-depressed individuals had the same ‘liking’ response to cartoons. But when the participants were asked how much effort they were willing to exert to watch these cartoons, it showed that depressed and non depressed individuals both were willing to work for the cartoons, with exception of depressed

individuals with a high anticipatory anhedonia. This could indicate that there is a distinction between anticipatory and consummatory anhedonia in depression.

In this current study we are interested in multiple hypotheses with regard to the prior described theories. First of all there is the presumed relationship between anhedonia and DA. We hypothesize that anhedonia is caused by a lowered amount of baseline striatal DA. We predict that there is a correlation between the the severity of anhedonia and the

amount of baseline striatal DA. To be precise we predict that there is a negative correlation, meaning the lower the amount of baseline striatal DA, the higher the

severeness of anhedonia will be. Because the use of PET scans are out of the scope for this current study the listening span will be used to approximate striatal DA, which will be explained in further detail in the methods and materials section. According to Cools, Gibbs, Miyakawa, Jagust and D’Esposito (2008) the outcome of the listening span correlates highly the baseline striatal DA, where a higher listening span indicates a higher baseline striatal DA and a lower listening span indicates a lower baseline striatal DA. They state that a higher listening span reflects a higher working memory, which itself reflects a higher baseline striatal DA. We focus primarily on anhedonia in this study instead of depression because not all individuals with depression suffer from anhedonia. Furthermore, because anhedonia is a diminished interest and lowered motivation we assume that this has a higher influence on EBDM compared to taking depression as a primary independent variable.

The second hypothesis of interest is whether individuals with a lower financial status (FS) truly are more motivated to work harder for a reward. Here we expect a negative correlation between FS and the amount of times an individual chooses the high effort compared to lower effort. As described in the above, individuals with a lower FS are of higher risk for psychiatric disorders such as depression. We hypothesize that because of the higher risk for depression, individuals with a low FS have a higher risk for anhedonia as well. Which makes us predict that there is a negative relation between FS and

anhedonia where participants with a low FS will have a higher anhedonia.

Finally we hypothesize that there is an interaction between FS, anhedonia and the amount of effort. Here FS will be the predictor for effort, but it will be moderated by

anhedonia. As described above, a lower FS will be negatively correlated with amount of effort, meaning the lower the FS, the higher the amount of effort. Anhedonia will moderate

this, meaning that a lower FS, combined with a more severe amount of anhedonia will lower the amount of effort. In this last analysis we will look at the different kinds of anhedonia as well. We predict that especially anticipatory anhedonia will moderate the effect of FS on the amount of effort compared to consummatory anhedonia

1.2 The Influence of Depression on Sense of Presence and EBDM: How Virtual Reality Can Influence Our Idea of Effort

“What is the Matrix? Control. The Matrix is a computer-generated dream world built to keep us under control in order to change a human being into this.”

-Morpheus, The Matrix, 1999.

For those who have seen the movie The Matrix, virtual reality (VR) may seem like something futuristic and even as a scary way to control us, as seen in the quote. Others may think of VR as a future way to escape from the real world and lose yourself in a virtual gaming world. With the growing expertise in technology in the last couple of years VR techniques have been used for gaming as well as for experimental and scientific needs. For example, VR is being used in clinical settings for the treatment of post traumatic stress disorder (Difede, Cuckor, Jayasinghe, Patt, Jedel, Spielman, Gioson & Hunter, 2007), or specific phobias, for example acrophobia (Krijn, Emmelkamp, Biemond, de Wilde de Ligny, Schuemie & van der Mast, 2005). One factor that is important for the successof VR is the sense of presence. Presence in VR is the sense of actually being in the virtual

reality(Sanchez-Vivez & Slater, 2005).

There are individual as well as technological aspects that have an influence on the perceived sense of presence. One example of an individual aspect that influences sense of presence is that of locus of control. Rotter (1966) describes locus of control as how an individual perceives the causal relationship of an event. When an individual feels as if he cannot control an event, it is said that this individual has an external locus of control.

Conversely if an individual feels that he causes an event, or feels as if he is in control of an event, it is said that the individual has an internal locus of control. Study shows that having an external locus of control will give a higher sense of presence (Murray, Fox and Petifer, 2007). Given that depressed individuals show an external locus of control (Benassi,

Sweeney & Dufour, 1988) it is possible that depressed individuals will experience a higher sense of presence when doing a VR task compared to non-depressed individuals.

One other, more technological, important aspect of presence in VR is immersion, which is the extent in which the VR environment can influence the senses (Slater and Wilbur, 1997). According to Wilbur and Slater (1997) there are several aspects that can influence immersion in VR which are Inclusiveness (the extent in which the external environment is excluded), Extensiveness (the extent of the sensory modalities that are accommodated), Surrounding (if the VR view is panoramic in stead of narrow) and Vivid (which includes resolution, color and quality of display). Wilbur and Slater (1997) state that presence is a function of immersion, meaning that when immersion is high, presence will be higher as well. In this current study the newest version of the Oculus Rift is used, the Oculus Rift Development Kit 2 (DK2)(Oculus VR, 2014). This is a head mounted virtual reality display with head tracking which should provide a higher immersion and

consequently a higher sense of presence.

One question that can arise from the differences in sense of presence in multiple environments is what the influence of it can be on given tasks in a VR environment. Bowman and McMahan (2007) argue that participants experience a higher level of

realness in a VR environment compared to a classroom environment in a military training. According to Wilbur and Slater (1997) the more sense of presence that the participant experiences, the more likely the participant will act as if it is in its own natural environment. So one may say that VR increases the realness of the environment that the participant is in, which in his turn might increase the ecological validity. Looking at EBDM, what can one

expect of the effect of VR on how EBDM is conducted? Since it is hypothesized that the higher the presence, the more real the participants experience the environment they’re in, one could expect that the decision that participants make in a VR environment are more closely to the ones they make in everyday life. When focusing on EBDM, where

participants have to exert a certain amount of effort before receiving a reward, it is possible to assume that the effort that is being exerted is more intense in VR compared to a 3D task where participants only look at a screen. When effort is perceived as more intense it is possible that this will influence the way individuals make their decisions concerning effort and reward.

The current study focuses on the influence of presence on EBDM and mainly on how the extent of effort is experienced in the three different settings: 2D, 3D and VR which will be further explained in the method section. First of all, the idea that depressed

individuals have an external LOC will be examined. As in the first main question this study focusses primarily on anhedonia and to be more specific anticipatory anhedonia. We predict that participants with an external LOC have a higher extent of anhedonia compared to participants with an internal LOC.

The second hypothesis focusses on the effect of the three different settings on the sense of presence and consequently on the perceived effort. It is expected, that due to the higher immersive qualities of the VR setting, participants will have a higher sense of

presence in the VR setting compared to the 3D setting, and a higher sense of presence in the 3D setting compared to the 2D setting. Consequently it is expected that the effort that is being exerted is perceived as more intense in the VR setting compared to the 3D setting and more intense in the 3D setting compared to the 2D setting. Since the effort is

expected to be more intense, it is expected that participants will choose the low effort option more often than the high effort option in the VR setting compared to the 3D setting, and more in the 3D setting compared to the 2D setting.

Next, the role of anhedonia on the sense of presence and perceived effort in the three settings will be evaluated. Since it’s hypothesized that depressed, or in our case individuals with anhedonia, have an external locus of control, we predict that they will have a higher sense of presence compared to non-anhedonic individuals. And that this will be higher in the VR setting compared to the 3D setting, and higher in the 3D setting

compared to the 2D setting because of the higher immersive power of the VR setting. Subsequently is is expected that participants with a higher extent of anhedonia experience a higher perceived effort compared to non-anhedonic participants. Because of the

differences in perceived effort and sense of presence it is expected that anhedonic

participants will choose the low effort condition more often in all three settings compared to the non-anhedonic participants.

Furthermore there is the question whether the extent of anhedonia and the extent of presence could predict the way participants make their choices. It is expected that

anhedonic participants have a higher sense of presence, and therefore choose the low effort condition more often compared to non-anhedonic participants. And this will be even more so in the VR setting compared to the 3D setting, and more in the 3D setting

compared to the 2D setting.

Lastly there is the question whether the extent of anhedonia, the extent of

perceived effort and the extent of presence could predict the way participants make their choices. As explained before it is predicted that when participants experience a higher sense of presence, their effort will be more realistic and therefore more intense. Since it is predicted that anhedonic participants will have a higher sense of presence, and therefore experience the effort as more intense, it is expected that they will choose the low effort condition more often than the high effort, as compared to non-anhedonic participants.

2 Methods an Materials

2.1 Main Question 1: Work, Money, Pleasure 2.1.1 Participants

There was a total of 40 participants (n = 28 female, n = 12 male), which were recruited through the university database, posters on the school campus and verbal advertisement. There was a screening process prior to the recruitment, which excluded participants with the following: (a history of) epilepsy, (a history of) neurological problems, heart problems, dizziness and/or nausea after fast movements as in roller-coasters or merry go rounds, glasses with astigmatism or glasses for farsightedness, younger than 18 years or older than 28 years. One participant was excluded because of age(30 years). One participant was excluded after wrongfully finishing the task. One participant was excluded after the participant didn’t fill out the questionnaires. 15 participants with a mean POI between minus three and three were excluded. This is because they solely chose high reward and ignored the amount of effort. Two participants didn't fill out the complete questionnaires, so they were excluded as well. This leaves 20 participants (n = 15 female, n = 5 male ) for analysis. The age ranged from 19-28, (mean age = 23.2, SD = 2.80). The participants either received 20 euros or 2 proefpersoon punten (UvA students only). A small part of the participants volunteered to participate for free, but they received a box of Merci as token of gratitude. All participants (volunteers as well) received a so called bonus of 0.50 euro’s for the decision tasks.

The study was approved by the Ethics Commission of the UvA. All participants were given an informed consent and an informational folder with details of the experiment which had to be signed before participating.

Participants had to choose between two tracks, each with a different amount of reward (represented by two numbers, adding up to a total of 2000) and each with a different amount of effort. Effort was represented by 5 bars (green, orange and red which stand for low, medium and high effort respectively). The time to complete the tracks was identical to make sure that the time to complete the tracks wasn't a confounding factor. The participants had to hold their hands in a fist and lean with their pinky on the two pokes of an Xbox one controller (Microsoft, 2013). When on the bottom of the screen the words “choose track” were shown, participants had to move both pokes to the left or the right to choose a track. To finish the tracks the participants had to keep a powerbar above a certain value, to make sure they would keep moving. The speed at which they moved forward was the same, no matter how much the powerbar was above the said value. To make sure the powerbar was green, participants had to move the two pokes in opposite directions. For the different amounts of effort the pokes had to be moved more or fewer times. The red bars needed more effort which meant a higher amount of moving of the pokes, the green bars needed almost no effort which meant moving the pokes about one time, and the orange bars needed moving the pokes a medium amount of times. For each trial the amount of effort was set, but the difference in reward between the two tracks was calculated based on the track chosen on the previous trial. When the low effort track was chosen, the difference in reward between low and high effort was higher on the next trial. Conversely, when the high effort track was chosen, the difference in reward between the low and high effort track was lower on the next trial.

It was expected that an equivalence point would be reached after 10 trials, which is a point where the participant will fluctuate between two values; the Point of Indifference (POI). The POI is the reward per unit over the amount of trials. A stair-casing procedure (Tversky and Kaneman, 1992) was used, where it is expected that on every trial, the participants will choose between the high effort and low effort choice based on how they

value the given effort and reward. When participants value reward more than effort, and hence chose the high effort condition, the difference between the reward in the high and low effort condition will decrease, which will result in a low POI. Conversely when

participants value effort more than reward, and hence chose the low effort condition, the difference between reward in the high and low effort condition will increase, which will result in a high POI (see figure ). To summarize, participants with a high POI were

prepared to exert a small amount of effort for a higher reward. Participants with a low POI were prepared to exert a lot of effort for a smaller reward (see figure 1). Since it was expected that the POI was reached after 10 trials, the mean of the final four trials was used to calculate the POI.

Figure 1. POI which represents the reward per unit effort over the amount of trials. Here a high POI means

the participant is prepared to do little work for a high reward, and a low POI means that the participant is prepared to do a lot of work for a small reward.

The decision task was performed on a computer screen in a secluded room. Each participant had to complete the task in the three different settings: 2 Dimensional (2D), 3 Dimensional (3D) and a Virtual Reality (VR) setting. To create the three different

environments for the decision task Unreal Engine 4 (Epic Games, 2012) was used. The 2D setting was a black screen, with on each top side of the screen a number representing the

reward. Below each number were several bars aligned which represented the amount of effort. In the middle of the bottom of the screen the powerbar was represented together with the words “Choose Track”, “Waiting for Tracks” and “ Go” (see figure 1 and figure 2 in the appendices for an example of the 2D version).

In the 3D version as in the VR setting participants saw the same effort bars and reward numbers but in stead of a black screen it looked as if they were in a cart on a rail. At the starting screen for each trial the cart was in a house and in front of the cart was a door. After choosing the track the doors opened and the cart moved outside where the track split into two tracks. The cart moves towards the chosen track and the participant is told to wait. When the word “Go” appear the participant had to exert effort by moving the pokes as told before. Low effort was represented by the track going through a plain without any constructions or blocking. But medium and high effort was presented by long

grass and bushes on the tracks respectively. After each trial the participant was back in the house to choose another track (see figure 3 to figure 6 for the examples of the 3D and VR version).

In the VR setting the participants had a Oculus Rift Development Kit 2 (DK2) on which provided the VR environment (Oculus VR, 2014). The DK2 is a head mounted VR glasses with positional tracking with a near infrared CMOS sensor so that the participants get a full VR experience. It has an internal tracking as well through the help of a

gyroscope, accelerometer and a magnetometer. It has a 960 x 1080 HD resolution per eye and a refresh rate of 75Hz, 72Hz and 60Hz. The DK2 was plugged in the computer

through a USB 2.0 port. The DK2 provided the participants with the ability to look around in the environment.

2.1.3 Striatal DA: Listening span

To measure striatal DA the Dutch version of the listening span task by Salthouse and Babcock (1991) was used. The listening span was originally modeled after the reading span task by Daneman & Carpenter (1980). In the current listening span task participants were presented by a sentence and were asked to answer a question on paper about the sentence while hearing it and remember the last word of the sentence. All questions had three answer possibility's. After the word “herinner” participants had to turn the page and write down the last word and turn the page back again to listen to the next sentence. In total there were 7 blocks, each with three trials. In the first block each trial contained a set of one sentence. In the second block each trial contained a set of two sentences. This continued until the seventh block where each trial contained a set of seven sentences. The word “herinner” was spoken after the end of each trial, meaning that in the seventh block participants had to remember 7 words.

The listening span was run on VLC media player on a computer in a secluded room with noise canceling headphones on. It was a paper version meaning that participants had to fill out the form with a pen. Before the start it was explained that the participants had to turn the page after hearing the word “herinner” and that they had to turn the page back again until they finished al three trials of the blocks. Participants were aloud to answer the questions of the sentences while hearing the sentence. Because the original paper version was particularly old, it was retyped. This had the consequence that the page numbers were not identical with the page number spoken on the listening span task. Participants were therefore asked to ignore the page numbers.

The scoring of the listening span was as followed: for each question answered right and all the words from one set remembered right 1/3 point was given. Meaning that there is a maximum of 1 point for each block and a maximum total of 7 points. According to Cools et al. (2008) the higher the score on the listening span, the higher the baseline striatal DA has to be. A score higher than four points is considered as having a high striatal DA baseline and below four points is considered as having a low striatal DA baseline.

2.1.4 Questionnaires

Because the experiment was conducted in the Netherlands, and all participants were native Dutch or speaking Dutch as a second language, all questionnaires were translated to Dutch. The participants completed the questionnaires in a secluded room on a computer screen in google forms(Google, 2015).

Anhedonia: The Temporal Experience of Pleasure Scale (TEPS) was used to measure the amount of anhedonia. The questionnaire was developed by Gard, Gard, Kring & John (2006). It exists of 18 items which measures both anticipatory (TEPS-ANT, 10 items) and consummatory anhedonia(CON, 8 items). An example of a TEPS-ANT item is “When I hear about a new movie starring my favorite actor, I can’t wait to see it”. An example of a TEPS-CON item is “ The smell of freshly cut grass is enjoyable to me”. The TEPS has a six points Likert scale (1 = very false for me, 6= very true for me) with a total of 108 points where a higher score reflects a lower extent of anhedonia. Since this study focusses on anticipatory anhedonia, only the TEPS-ANT items will be analyzed which will give a total of 60 points. Participants scoring between 0-30 will be considered as highly anhedonic, and between 31-60 participants will be considered as low anhedonic.

Financial Status: A self-made FS questionnaire was used which tested the current FS of the participants. The questionnaire consists of 11 questions where the answer possibilities could range depending on the kind of question. A few examples to give a global idea of the questionnaire are: “Do you have a part-time/full-time job?” “What is your

current education?” and “What is your monthly after tax income, including study loan, jobs and other kinds of income?”. Because a large number of participants are students the quota of having a high or low FS may differ from the normal income quota. Participants will be considered to have a low FS when their total income is less than 300,- Euro, and when they have less than 150,- Euro left at the end of the month. Participants will be considered to have a normal FS when their total income is between 300,- Euro and 800,- Euro, and when they have less then 300,- Euro left at the end of the month. Participants will be considered to have a high FS when their total income is higher than 800,- Euro, and when they have more then 300,- euro left at the end of the month. When, for example,

participants will have a lower income, but still have a high amount of money left at the end of the month due to low monthly expenses, the amount of money left at the end of the month will be taken as detrimental factor to classify the participant.

2.2 Main Question 2: Effort 2.2.1 Questionnaires

All questionnaires were translate to Dutch for the same reasons as described in the first main question. The questionnaires were conducted in a secluded room in Google Forms on a computer screen.

Anhedonia: the TEPS was used to measure anhedonia. For a detailed description of the TEPS see the description in 2.1.3.

Locus of Control: To measure LOC the LOC Questionnaire (LOCQ) was used (Rotter, 1966). The LOCQ consists of 29 questions each with two answer possibilities which either stands for a internal or a external LOC. 23 of the 29 questions are used to calculate the LOC where each external answers provides one point. This leads to a total score of 23 points, where a high score is considered as an external LOC and low score is considered as an internal LOC.

Perceived Effort: To measure the perceived effort a self-made questionnaire consisting of four questions was conducted.. For the following two questions participants had to answer for the three different settings (2D, 3D, VR): “State for each task, how physically exerting it felt”, “State for each task, how much effort you exerted”. These questions had a 10-point and seven-point Likert scale respectively .The other two

questions were: “The distance I had to cover felt in the virtual reality task” with a five-point likert scale (ranging from “much longer than in the 3D task” to “much shorter than in the 3D task”) and “the distance I had to cover felt in the 3D task” with a five-point Likert scale (ranging from “much loner than in the 2D task" to “much shorter than in the 2D task”). A higher score was represented as a higher perceived effort in the settings.

Igroup Presence Questionnaire (IPQ): To measure the amount of presence perceived in each of the three setting in the decision tasks the IPQ was used (Schubert, Friedmann, and Regenbrecht, 2001). The IPQ consists of a total of 14 questions and consists of three independent sub-scales and one general item. The general item measures the sense of being in the virtual environment. The three sub-scales were the following: 1) Spacial presence; the feeling of being physically in the virtual environment. 2) Involvement; the involvement that the participant experienced and the attention that the participant devoted to the virtual environment. 3) Experienced Realism; how the participant had a realistic experience in the virtual environment. The IPQ has a five-points Likert scale (ranging from completely disagrees to completely agrees). This will give a total of 70 points where a higher score represents a higher sense of presence. Since there were three

different settings (2D, 3D and VR) the IPQ was conducted three times to measure the sense of presence for each of the settings.

2.3 Procedure

All the sessions were conducted at the University of Amsterdam in two secluded rooms. There were sessions of two hours in which one or two participants were tested.

Each session was divided in two blocks, consisting of either the listening span task and the questionnaires (IPQ and perceived effort excluded) or the three decision tasks

followed by the IPQ and perceived effort questionnaire. Because there was only one DK2 available, part of the participants started with one block and the other part started with the other block. After finishing the blocks participants switched or moved to the other room where they finished the other block.

All test leaders were given a set explanation for all of the tasks to prevent observer expectancy bias. For the decision tasks the test leaders had to demonstrate the 3D setting of the task to make sure the participants understood the task. Participants were explained about the effort bars, the reward numbers and how they had to use the pokes. It was told that the participants would conduct the same decision task three times, but each time in a different setting. The order in which the participants conducted the three tasks was counter balanced to prevent order effect. Participants were told that they could receive a bonus of 0.50 euro dependent on how they completed the task, but all participants were given 0.50 euro bonus independent of how they completed the task. After finishing the three tasks the participants had to fill out the IPQ and the perceived effort questionnaire. For the listening span task participants were told they would conduct a memory task. Participants were told that they would have to answer a question of each of the heard sentence while

remembering the last word of each sentence. Participants were told that they started easy with one sentence at a time, but that an extra sentence would come on each block. After the explanation the test leader started the listening span and waited too see if the

participants understood the explanation. If seen that the participant made a mistake in the first block the test leader would explain the task a second time and restart the task. The test leader would leave the room quietly if the participant conducted the first block in the right order. Note that the test leader did not check for the right answers, but only saw to it that the participant would turn the page when needed.

3 Results

3.1 Main Question 1: Work, Money, Pleasure

Descriptive Statistics: All analysis were done over 20 participants in SPSS. The mean of the TEPS-ANT was 34.6 (SD = 5.52). For the POI the average over all three blocks was used (for more information about POI please see the second main question). This gave a mean POI of 15.267 (SD = 9.19). For FS participants were classified in low (n = 9), normal (n = 7) or high (n = 4) FS. The mean of the Listening Span task was 4.267 (SD = 1.168) and participants were classified in low(n = 6) or high (n = 14) listening span.

Anhedonia vs. Listening span: A one-way anova was conducted to see whether the score on the listening span could.predict the amount extent of anhedonia. Levene’s test was not significant F(1,18) = 3.604 p = .074. There wasn't a significant effect of the score of the listening span task, representing baseline DA on the TEPS-ANT outcome F(1,18) = .019, p = .892 (see figure 2). Planned comparison showed that having a high listening span didn’t significantly mean a lower score on the TEPS-ANT, t(18) = .138, p = .892.

Figure 2. The mean ANT-TEPS score in the different listening Span scores: Lage Listening Span = low score on the listening span, Hoge Listening Span = high score on the listening span.

Anhedonia vs. FS: A one-way ANOVA was conducted to see whether financial status could predict the extent of anhedonia. Levene’s test was nog significant, F(2,17) =.925, p=.416. There was no significant effect of FS on the TEPS-ANT outcome F(2,17) = .027, p = .974 (see figure 3). There was no significant linear trend F(1,17) = .002, p = .963, indicating that as Financial Status increased, the score on anticipatory anhedonia didn’t increase as well. Planned Contrast showed that having a low FS didn’t significantly mean a lower score on the TEPS-ANT compared to having a high or normal FS, t(17) = -.156, p = .878. Having a low FS didn’t significantly mean a lower score on the TESP-ANT

compared to a high FS, t(17) = .048, p = .963.

Figure 3. The mean TEPS-ANT score on the three different FS: Laag = low FS, midden = normal FS, hoog = high FS.

Financial status vs. POI: A one-way ANOVA was conducted to see whether FS could predict the overall POI. Levene’s test for homogeneity of variance was not

= .512, p = .609 (see figure 4). There wasn’t a significant linear trend F(1,17) = .426, p =.523, indicating that as FS increased, the POI didn't increased or decreased as well. Planned Contrast showed that having a low FS didn’t significantly lower the POI compared to having a normal or high FS, t(17) = -.956, p = .352. Planned contrast showed that

having a low FS didn't significantly lower the POI compared to having a high FS, t(17) = -.653, p = .523.

Figure 4. The Mean POI score on the three different FS: Low, normal and high.

Financial Status, POI and Anhedonia: A moderation analyses was done to see whether anhedonia would moderate the relationship between FS and overall POI. When the TEPS-ANT was low, there was no significant relation between FS and POI, b = -5.847 95% CI [-18.113, 6.420], t = -1.011, p = .327. When the TEPS-ANT was at mean level, there was no significant relation between FS and POI, b = -2.210, 95% CI[-8.245, 3.824], t = -.777 p = .449. When the TEPS-ANT was high, there was no significant relation between FS and POI, b = 1.426, 95% CI[-5.031, 7.883], t = -.468, p = .646. These results tell that the relationship between FS and POI is not significantly effected by the extent of

Figure 5. Moderation effect of Anhedonia on Financial Status and POI.

3.2 Main Question 2

Descriptive Statistics: The descriptives for anhedonia can be found in the first main question results section (3.1).

Anhedonia vs LOC: A one-tailed bivariate correlation was used to see whether there was a relationship between anhedonia and LOC. There was no significant

correlation between the TEPS-ANT and the LOCQ, r(1,20) = .182, p .221, which indicates that there is no relation between external or internal LOC and the extent of anhedonia.

Condition on IPQ: A repeated measure was conducted to see whether the condition (2D, 3D or VR) had an effect on the sense of presence, where sense of presence was the within-subjects variable and the order of the conditions was the between-subjects variable. Mauchly’s test of Sphericity for the effect of condition was not violated, X^2(2) = 4.739, p = .094. There was a significant main effect of condition on the sense of presence, F(2,28) = 118.385, p < .001, n^2 = .894 (see figure 6). Contrasts revealed that presence was

significantly higher in the VR setting compared to the 3D setting, F(1,14) = 127.790, p < .001, n^2 = .901. Contrast revealed that the presence was significantly higher in the 3D setting compared to the 2D setting as well, F(1.14) = 40.745, p < .001, n^2 = .744. There was no significant effect of order, indicating that the scoring of the order that the

participants were in were similar, F(5,14) = .404, p = .838, n^2 = .126. There was no significant interaction effect of condition and the order in which the participants conducted the conditions on sense of presence, F(10,28) = .725, p = .694, n^2 = .206. This indicates that the presence sensed in the different conditions did not differ between the orders that the participants were in. Contrast revealed no significant effect comparing the order in the VR setting compared to the 3D setting, F(5,14) = 1.392, p = .286, n^2 = .332. Contrast revealed no significant effect comparing the order in the 3D setting compared to the 2D setting, F(5,14) = .785, p = .577, n^2 = .219.

Figure 6. The mean IPQ score on the three different conditions.

Condition on Perceived Effort: A repeated measure was conducted to see whether the condition (2D, 3D or VR) had an effect on the perceived effort, where perceived effort was the within-subjects variable and order of the conditions was the between-subjects variable. Maulchy’s test for sphericity wasn't violated for the effect of condition, X^2(2) =

4.455, p = .108. There was a significant main effect of condition on perceived effort F(2,28) = 11.123, p < .001 (see figure 7). Contrasts revealed that participants perceived effort as higher in the VR setting as compared to 3D setting, F(1,14) = 16.436 p = .001, n^2 = .540. Participants did not perceive effort as higher in the 3D setting compared to the 2D setting, F(1,14) = .033, p = .858, n^2 = .002. There was no significant effect of order, indicating that scoring from the different orders were similar, F(5,14) = .834, p = .547, n^2 = .230. There was no significant interaction effect between the conditions and the order, F(10,28) = 1.220, p = .321, n^2 = .303. This indicates that the perceived effort in the different conditions did not differ between the different orders in which the conditions were

conducted. The contrast revealed no significant effect when comparing the order scores to the VR setting compared to the 3D setting, F(5,14) = 1.093, p = .407, n^2 = .281. Neither revealed the contrast a significant effect when comparing the order scores to the 3D setting compared to the 2D setting, F(5,14) = .754, p = .597, n^2 = .212.

Figure 7. The mean perceived effort score on the three different conditions.

Condition on POI: A repeated measures analysis was used to see whether condition had an effect on the POI. Here condition was the within-subject variable and order in which the conditions were conducted was the between-subject variable. Mauchly’s test of Sphericity for the effect of condition was not violated, X^2(2) = 3.292, p = .193.

There was no significant effect of condition on POI, F(2,28) = .095, p = 909, n^2 = .007 (see figure 9). Contrast revealed that POI was not significantly higher in the VR setting compared to the 3D setting, F(1,14) = .156, p = .699, n^2 = .011. Contrast revealed that the POI was not significantly higher in the 3D setting compared to the 2D setting, F(1,14) = .183, p = .676, n^2 = .013. There was no significant effect of order, indicating that the scoring of the order that the participants were in were similar, F(1,14) = .939, p = .486, n^2 = .251. There was no significant interaction effect of condition and the order participants conducted the conditions in on POI, F(10,28) = 1.951, p = .080, n^2 = .411. This indicated that the POI in the different conditions did not differ between the order that the participants were in. Contrast revealed a significant effect comparing the order in the VR setting

compared to the 3D setting, F(5,14) = 3.132, p = .042, n^2 = .528. Contrast revealed no significant effect comparing the order in the 3D setting compared to the 2D setting, F(5,14) = 1.571, p = .232, n^2 = .359.

Figure 8. The mean POI in the three different conditions.

Anhedonia vs IPQ: A repeated measures was conducted to see whether the extent of anhedonia had an influence on the sense of presence in the different conditions. Here

sense of presence was used as a within-subject variable, and TEPS-ANT as a between-subject variable. Order in which the participants conducted the conditions was used as a covariate. Mauchly’s test of sphericity was not violated, X^2(2) = 5.590, p =.61. There was no significant effect of TEPS-ANT indicating that scores from low and high anhedonic participants were similar, F(1,17) = .051, p = .824, n^2 = .003. There was no significant effect of order, indicating that scores from the different orders the participants conducted the conditions were similar, F(1,17) = .148, p = .705, n^2 = .009. There was no significant interaction effect of anhedonia and the sense of presence in the different conditions, F(2,34) = .798, p = .459, n^2 = .045. Contrasts revealed that there was no significant interaction when comparing the high-, and low anhedonic-participants to the VR condition compared to the 3D condition, F(1,17) = 1.968, p = .179, n^2 = .104. Contrast revealed that there was no significant interaction when comparing the high and low anhedonic-participants to the 3D condition compared to the 2D condition, F(1,17) = .950, p = .343, n^2 = .053.

Figure 9. The mean IPQ score of low and high anhedonic participants in the three different conditions. Low = low-anhedonic, high = high anhedonic.

Anhedonia vs perceived effort: A repeated measures was conducted to see whether the extent of anhedonia had an influence on the the perceived effort in the three different conditions. Here anhedonia was the between-subjects variable where participants were classified between high or low anhedonic. Perceived effort was the within-subjects variable and the order in which participants conducted the conditions was the covariate. Mauchly’s test for significance was not violated, X^2(2) = 4.846, p = .089. There was no significant effect of order, indicating that the scoring from the different orders were similar, F(1,17) = 3.995, p = .062, n^2 = .190. There was no significant effect of TEPS-ANT, indicating that scoring from low- and high-anhedonic participants were similar, F(1,17) = .851, p = .369, n^2 = .048 (see figure 10). There was no significant interaction effect of perceived effort and TEPS-ANT, F(2,34) = 2.692, p = .082, n^2 = .137. Contrasts revealed that there was no significant interaction when comparing high-, and low-anhedonic participants in the VR setting compared to the 3D setting, F(1,17) = 3.875, p = .066, n^2 = .186. Contrast

revealed that there was no significant interaction when comparing high-, and

low-anhedonic participants in the 3D setting compared to the 2D setting, F(1,17) = .104, p = .751, n^2 = .006.

Figure 10. Mean perceived effort score of low and high anhedonic participants in the three different settings. Low = low anhedonic, high = high anhedonic.

Anhedonia vs POI: A repeated measure was conducted to see whether the extent of anhedonia had an effect on POI in the three different settings (2D, 3D and VR). Here TEPS-ANT was the between-subject variable, POI was the within subject-variable and the order in which participants conducted the conditions was taken as a covariate. Mauchly’s test of Sphericity was not violated, X^2(2) = 3.189, p = .148. There was no significant effect of order, indicating that the scoring between the order in which participants

conducted the conditions were similar, F(1,17) = .320 p = .579, n^2 = .019 (see figure 11). There was no significant effect of TEPS-ANT, indicating that the scoring between the high-, and low-anhedonic participants were similarhigh-, F(2high-,17) = .272high-, p = .609high-, n^2 = .016. There was no interaction effect of POI and TEPS-ANT, F(2,34) = .678, p = .515, n^2 = .038. Contrasts revealed that there was no significant interaction when comparing high-, and low-anhedonic participants in the VR setting compared to the 3D setting, F(1,17) = .071 p = .793 n^2 = .004. Contrasts revealed that there was no significant interaction when comparing high-, and low-anhedonic participants in the 3D setting compared to the 2D setting, F(1,17) = 1.690, p = .211, n^2 = .090.

Figure 11. Mean POI of low and high anhedonic participants in the three different settings. Low = low anhedonic, high = high anhedonic.

Anhedonia vs IPQ vs POI: A multiple regression was conducted to see if extent of anhedonia and the sense of presence predicted the POI in each of the three settings. Using the enter method it was found that anhedonia and sense of presence (2D) did not explain a significant amount of variance in POI in the 2D version, F(2,17) = 2.71, p = .766, R2 = .031, R2adjusted = -.083. Analysis showed that sense of presence didn't significantly predict the POI-2D (Beta = -.048, t(18) = -.199, p = .892. Anhedonia didn't significantly predict the POI-2D as well (Beta = .166. t(18) = .693, p =.498). Using the enter method it was found that anhedonia and sense of presence(3D) didn’t explain a significantly amount of variance in POI in the 3D version, F(2,17) = 1.008, p = .386, R2 = 1.06, R2adjusted = .001. Analysis showed that anhedonia did not significantly predict the POI-3D (Beta = .174, t(18) = .747, p = .465. It showed that sense of presence(3D) did not significantly predict the POI-3D either (Beta = .248, t(18) = 1.064, p = .302). Using the enter method it was found that anhedonia and sense of presence did not explained a significant amount of variance in POI-VR, F(2,17) = .565, p = .579, R2 = .062, R2adjusted = -.048. Analysis showed that anhedonia did not significantly predict the POIVR (Beta = .037, t(18) = -.158, p = .876. Analysis showed that sense of presence (VR) did not significantly predict POI-VR as well (Beta = .250, t(18) = 1.061, p = .304).

POI vs perceived effort vs IPQ vs anhedonia: A multiple regression was conducted to see if the extent of Anhedonia, the perceived effort and the sense of presence predict the POI in each of the three settings. Using the enter method it was found that anhedonia, sense of presence and perceived effort did not explained a significant amount of variance in POI-2D, F(3,16) = 1.861, p = 1.77, R2 = .259, R2adjusted = .120. Analysis showed that sense of presence(2D) didn’t significantly predict the POI-2D (Beta = -.036, t(17) = -.165, p = .871). Analysis showed that anhedonia didn’t significantly predict the POI-2D (Beta = .103, t(17)= .472, p =.644). And finally analysis showed that perceived effort(2D) did

significantly predict the POI-2D(Beta = .482, t = 2.217, p =.041). Using the enter method it was found that anhedonia, sense of presence(3D) and perceived effort(3D) did not

explained a significantly amount of variance in POI in the 3D version, F(3,16) = 2.428, p = .103, R2 = .313, R2adjusted = .1844. Analysis showed that anhedonia did not significantly predict the POI-3D (Beta = .291, t(17) = 1.342, p = .198). Sense of Presence(3D) did not significantly predict the POI-3D as well (Beta = .087, t(17) = .392, p = .701). Finally perceived effort(3D) did significantly predict the POI-3D (Beta = -.490, t(17) = -2.194, p = .43). Using the enter method it was found that anhedonia, sense of presence(VR) and perceived effort(VR) did not explain a significant amount of variance in POI-VR, F(3,16) = .289, p = .762, R2 = .068, R2adjusted = -.107.. Analysis showed that anhedonia did not significantly predict the POI-VR (Beta = -.004, t(17) = -.016, p = .988. The sense of

presence (VR) did not significantly predict the POI-VR (Beta = .224, t(17) = .906, p = .378). Lastly, perceived effort (VR) did not significantly predict the POI-VR as well (Beta = . 085, t(17) = .314, p = .758.

4. Discussion

The results of the study aren’t consistent with the predictions and hypotheses as described in the introduction. First of all, having a higher striatal DA baseline did not predict a lesser extent of anhedonia. The second hypotheses, namely that a lower FS would be accompanied by a higher extent of anhedonia wasn’t found in the results either. The hypotheses that individuals with a lower FS would be more motivated to work for a reward wasn’t found in the results. Lastly anhedonia did not moderate the relationship between FS and overall POI.

This study does have several discussion points. Starting with the hypotheses concerning striatal baseline DA and the listening span: There was no prior research that indicated that the listening span would be a valid indicator of anhedonia. Although Cools et

al. (2008) did show that there was a positive correlation between the amount of baseline striatal DA and working memory, one can not conclude from these results that there is a correlation between anhedonia and the listening span as well. Furthermore, although DA is considered as an important factor in EBDM, its precise role in anhedonia is still

unknown. Future research could investigate whether the scores in the listening span task correlate with the extent of anhedonia. Unfortunately neuro-imaging techniques were out of the scope of this study, but future research could use neuro-imaging techniques, such as PET-scans, to further investigate whether anhedonia is indeed accompanied by a lower baseline striatal DA.

One other discussion point could be the relative small sample size. One needs enough power to find smaller or more subtle effects. Since in this study there were only 20 participants left to analyse, one could say that the sample size was too small to find these small effects, if they exist. One factor that caused the small sample size was the large amount of excluded participants. Partly due to the fairly new method that was used, namely VR for EBDM, a few mistakes were made what caused 9 participants to be excluded due to technical improvements. Furthermore some participants were excluded who did not finish all questionnaires, also due to some technical difficulties.

Two other discussion points concerning the participants were possible gender and education effects. In this sample 75% of the participants were female, which could

influence the effects, especially since the sample was so small. Furthermore, although participants were recruited from the streets as well, in stead of solely from psychology students, 14 out of 20 participants either finished or were still studying on the University. This current study did not correct for education effects, but future study could try to make the sample more representative for the entire population.

4.2 Main question 2

The results for the second main question are to a great extent inconsistent with the prior predictions. No relationship was found between LOC and anhedonia. Since there is a great extent of prior research which did show this relationship, one could state that there were some faults in this current study. But before looking at the discussion points, the other hypotheses will be described first. It seemed that condition did have an effect on sense of presence, and that individuals experienced a higher sense of presence in the VR condition compared to the 3D condition, and a higher sense of presence in the 3D

condition compared to the 2D condition. When looking at the differences in immersion between the conditions, one could say that the main difference between the 2D and 3D condition was the visual environment, or as Slater and Wilbur (1997) might say the vividness and extensiveness. Since in the 2D environment participants only saw a black screen and the information needed to conduct the task, the vividness and extensiveness were rather low. Conversely in the 3D setting participants saw the track, and the cart and the environment they were in. So one could say that immersion was higher in the 3D condition compared to the 2D condition. But what about the difference between 3D and VR? The visual aspects were quite similar, since participants saw the same visual

environment. However in the VR condition participants could look around and were more excluded from the external environment, which increases inclusiveness as well as

surrounding of the environment, and consequently increases immersion even more. So one could say that the amount of immersion does increase the sense of presence that participants experience.

The following hypothesis was that a higher sense of presence would result in a higher perceived effort in the VR condition compared to the 3D condition and in the 3D condition compared to the 2D condition. The results showed that participants did

experience effort as more intense in the VR condition compared to the 3D condition, but that it was not more intense in the 3D setting compared to the 2D setting. From these

results one could hypothesize that inclusiveness and surrounding are the most important factors to influence the realness of the environment, and as such increases the perceived effort. Future research could research this in more detail.

The next hypothesis was whether the condition would influence ones decisions. However the results did not show any effect of the condition on the participants’ decisions. So although participants did report a higher presence in the VR task and perceived the effort as more intense, this did not influence the way that they made their decisions. However, one important point of discussion for this study is the extent in which the effort was perceived as realistic in the tasks. Since 15 of the participants were excluded because they ignored effort, one could say that the effort in this current study was not realistic enough. Since participants used an X-box controller, one could say that this did not fully simulate the effort that was depicted in the task. If participants could hold similar levers as depicted on the screen, and these levers have some sort of resistance this could simulate the effort in a more realistic manner, which in turn could increase the sense of presence and the perceived effort. One other point concerning presence is that in this study the participant did not have a body or hands in the VR or 3D conditions. So when the cart was moving in the 3D setting, the participant only saw the levers move without hands

controlling the lever, beside their own hands on the X-box controller. In the VR setting the participant not only saw the levers without the hands, but when looking down the

participants saw a chair, instead of a body. Having a body in VR might enhance the sense of presence. One counter argument for a body could be that the body in a virtual

environment could differ from one's own body (differences in size, gender, race). However Slater, Spanlang, Sanchez-Vives and Blanke (2010) showed that ownership, the extent in which the participant feels the virtual body is his or hers, can be high even when the virtual body differs from ones own body. So having a virtual body could increase presence, even

when the virtual body is different from the actual body. It is still necessary to study this phenomenon in more detail, especially whether it truly increases presence.

It seemed that the extent of anhedonia did not influence perceived effort or sense of presence, or even influenced the way participants made their decisions. However in this study mainly students were included, and only one of the participants was clinically depressed. To study the influence of anhedonia in a higher extent, it is necessary to include more clinically depressed or anhedonic participants in future research. Participant-wise the same discussion points used in the first discussion could be used.

Overall one could say that VR is a possible tool for the study of EBDM, but there are still some challenges ahead for future research. The fact that VR has shown to have a higher sense of presence compared to 3D tasks, makes it possible for VR to be a useful method in psychological research. After all, we want participants to behave in experimental studies as they do in their everyday life. Therefore it is necessary to use the growing

knowledge of technology to improve this method. This way, we may be able to build our own matrix as a means of psychological research.

Literatuur

-American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.

-Bardgett, M. E., Depenbrock, M., Downs, N., & Green, L. (2009). Dopamine Modulates Effort-Based Decision-Making in Rats. Behavioral Neuroscience, 123(2), 242–251. doi:10.1037/a0014625.Dopamine

-Benassi, V. a, Sweeney, P. D., & Dufour, C. L. (1988). Is there a relation between locus of control orientation and depression? Journal of Abnormal Psychology, 97(3), 357– 367. doi:10.1037/0021-843X.97.3.357

-Berridge, K. C., & Robinson, T. E. (2003). Parsing reward. Trends in Neurosciences, 26(9), 507–513. doi:10.1016/S0166-2236(03)00233-9

-Bowman, D. a, Mcmahan, R. P., & Tech, V. (2007). Virtual Reality: How Much Immersion Is Enough? (Cover story). Computer, 40, 36–43.

-Cools, R., Gibbs, S. E., Miyakawa, A., Jagust, W., & D’Esposito, M. (2008). Working memory capacity predicts dopamine synthesis capacity in the human striatum. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 28(5), 1208–1212. doi:10.1523/JNEUROSCI.4475-07.2008

-Daneman M, Carpenter P (1980) Individual differences in working memory and reading. J Verbal Learning Verbal Behav 19:450 – 466.

-Der-Avakian, A., & Markou, A. (2012). The neurobiology of anhedonia and other reward-related deficits. Trends in Neurosciences, 35(1), 68–77. doi:10.1016/j.tins.

2011.11.005

-Difede, J., Difede, J., Cukor, J., Cukor, J., Jayasinghe, N., Jayasinghe, N., … Hoffman, H. G. (2007). Virtual reality exposure therapy for the treatment of posttraumatic stress disorder following September 11, 2001. The Journal of Clinical Psychiatry,

-Doya, K. (2008). Modulators of decision making. Nature Neuroscience, 11(March), 410– 416. doi:10.1038/nn2077

-Easterlin, R. a. (2005). Diminishing marginal utility of income? Caveat emptor. Social Indicators Research, 70, 243–255. doi:10.1007/s11205-004-8393-4

-Floresco, S. B., & Ghods-Sharifi, S. (2007). Amygdala-prefrontal cortical circuitry regulates effort-based decision making. Cerebral Cortex, 17(February), 251–260. doi:10.1093/cercor/bhj143

-Forbes, E. E., Christopher May, J., Siegle, G. J., Ladouceur, C. D., Ryan, N. D., Carter, C. S., … Dahl, R. E. (2006). Reward-related decision-making in pediatric major

depressive disorder: An fMRI study. Journal of Child Psychology and Psychiatry and Allied Disciplines, 47, 1031–1040. doi:10.1111/j.1469-7610.2006.01673.x

-Gard, D. E., Gard, M. G., Kring, A. M., & John, O. P. (2006). Anticipatory and

consummatory components of the experience of pleasure: A scale development study. Journal of Research in Personality, 40, 1086–1102. doi:10.1016/j.jrp. 2005.11.001

-Google, (2015) received from: https://docs.google.com/forms/d/

1EbO2blgIBLi8Dv6a0WlKVgsaOMN9DU_YSG8y5U-TMCw/edit -Heyman, J., & Ariely, D. (2004). Effort for payment - A tale of two markets.

Psychological Science, 15, 787–793. doi:10.1111/j.0956-7976.2004.00757.x -Krijn, M., Emmelkamp, P. M. G., Biemond, R., De Wilde De Ligny, C., Schuemie, M. J., & Van Der Mast, C. a P. G. (2004). Treatment of acrophobia in virtual reality: The role of immersion and presence. Behaviour Research and Therapy, 42, 229–239. doi: 10.1016/S0005-7967(03)00139-6

-Murray, C. D., Fox, J., & Pettifer, S. (2007). Absorption, dissociation, locus of control and presence in virtual reality. Computers in Human Behavior, 23, 1347–1354. doi: 10.1016/j.chb.2004.12.010

-Oculus VR (2014). Development Kit 2. received from: https://www.oculus.com/dk2/ -Rotter, J. B. (1966). Generalized expectancies for internal versus external control of

reinforcement. Psychological Monographs: General and Applied, Vol 80(1), 1966, 1-28. doi: 10.1037/h0092976

-Sanchez-Vives, M. V, & Slater, M. (2005). From presence to consciousness through virtual reality. Nature Reviews. Neuroscience, 6(April), 332–339. doi:10.1038/ nrn1651

-Saraceno, B., Levav, I., & Kohn, R. (2005). The public mental health significance of research on socio-economic factors in schizophrenia and major depression. World Psychiatry : Official Journal of the World Psychiatric Association (WPA), 4, 181– 185.

-Schubert, T., Friedmann, F., & Regenbrecht, H. (2001) The experience of presence: factor analytic insights. Presence: Teleoperators & Virtual Environments, 10, 266-281. -Slater M, Spanlang B, Sanchez-Vives MV, Blanke O (2010) First Person Experience of

Body Transfer in Virtual Reality. PLoS ONE 5(5): e10564. doi:10.1371/journal.pone. 0010564

-Slater, M., & Wilbur, S. (1997). A framework for immersive virtual environments (FIVE).. Presence: Teleoperators & Virtual Environments, 6(6), 603.

-Sherdell, L., & Gotlib, I. H. (2012). NIH Public Access, 121(1), 51–60. doi:10.1037/ a0024945.Anticipatory

-Treadway, M. T., Bossaller, N., Shelton, R. C., & Zald, D. H. (2013). Translational Model of Motivational Anhedonia, 121(3), 553–558. doi:10.1037/a0028813.Effort-Based -Treadway, M. T., Buckholtz, J. W., Schwartzman, A. N., Lambert, W. E., & Zald, D. H.

(2009). Worth the “EEfRT”? The effort expenditure for rewards task as an objective measure of motivation and anhedonia. PLoS ONE, 4(8), 1–9. doi:10.1371/

-Treadway, M. T., Buckholtz, J. W., Cowan, R. L., Woodward, N. D., Li, R., Ansari, M. S., … Zald, D. H. (2012). Dopaminergic Mechanisms of Individual Differences in Human

Effort-Based Decision-Making. Journal of Neuroscience, 32(18), 6170–6176. doi: 10.1523/JNEUROSCI.6459-11.2012

-Tversky, A., Kahneman, D. (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and Uncertainty, 5, 297-323.

-Wacker, J., Dillon, D. G., & Pizzagalli, D. a. (2009). The role of the nucleus accumbens and rostral anterior cingulate cortex in anhedonia: Integration of resting EEG, fMRI, and volumetric techniques. NeuroImage, 46(1), 327–337. doi:10.1016/

j.neuroimage. 2009.01.058

-Xbox One Wireless Controller. Retrieved from http://www.xbox.com/en-US/xbox- one/ accessories/controllers/wireless-controller

figure 1. Choosing the track in the 2D version. Here participants have to choose between the high effort/high reward (right track) and low effort/low(left track) option.

figure 2. Exerting effort in the 2D version. Participant has to keep the powerbar in green to finish the track. Each bar turns blue when participants finished that part of the track.

Figure 3. Choosing the track in the 3D version. Participants have to choose between the high effort/high reward (right track) or low effort/low reward (left track) Here the rewards differ in great extent, whereas the effort is almost similar.