1
Bachelor thesis
Effort based decision making
The relationship between anhedonia and dopamine function in effort based decision making
in different immersive paradigms.
Thomas Hoedeman
Studentnumber: 10318070
Universiteit of Amsterdam
Supervisor: Jasper Winkel
Date: 28-05-2015
Word count bachelor thesis: 7503
Word count research proposal: 1886
2
Contents
Abstract
P4
Effort based decision making
P5
Presence and immersion
P6
Dopamine function and effort based decision making
P9
Anhedonia and dopamine function in effort based decision making
P11
Methods
P14
Results
P17
Discussion
P24
Conclusion
P28
Research proposal
P29
References
P36
3
‘I really don't like to take the easy way out, if I can help it, on anything I do, I like to really
make it a challenge. I don't know how to create by taking the easy routes. I've tried, you
know, I've tried to let myself, but I always struggle to compensate.’
4
Abstract
Anhedonia and dopamine are both related to individual differences in effort based decision making. More recent findings imply that dopamine is associated with the anticipation on reward instead of the liking of the reward itself. Also people who experience anhedonia mainly have a deficit in anticipatory pleasure, the experience of consummatory pleasure is relatively intact. In this study it is proposed that individual differences in dopamine functions explain these anticipatory deficits. Immersion can cause a higher sense of being present in a task, this relationship is mediated by cognition. Up till now presence has been researched as a dependent variable, but the effect of presence on cognition isn't clear yet. In this research it is proposed that a sense of presence will influence EBDM. The prediction is an increased perceived effort in a virtual reality paradigm which will reduce the motivation for rewards. 49 undergraduate students participated in the experiment. They completed an EBDM paradigm in different immersive environments (2D, 3D and VR), the listening span as an estimation of dopamine synthesis levels in the striatum and several
questionnaires. Neither the effect of immersion on EBDM nor the relationship between dopamine and anticipatory anhedonia were established, it is highly probable that the possibility of finding of these relationships was obstructed by methodological errors, a research proposal is added to suggest a better setup to study anhedonia, dopamine & EBDM.
5
Effort based decision making
We are confronted on a daily basis with a wide range of decisions that we have to make. Anticipating the results of these decisions is crucial for this process of decision making (Kurniawan et al. 2011). Especially when you have to make an effort for something, you want to make sure that your effort won’t be for nothing. After all, you usually would rather not make an effort when there is no potential benefit. Humans and animals tend to make a greater effort for larger rewards, but only when the reward is in proportion to the effort (Assadi et al. 2009; Salamone et al. 1994). Apparently a consideration has to be made between the reward and the amount of effort, this process of deciding if an action is worth the trouble is called effort based decision making (EBDM). EBDM is a process that is based on gaining the largest net profit on a short-term period (Assadi et al. 2009). So the weighing between effort and reward is based on which choice gives the biggest gain at this moment relative to the costs that have to be made to obtain this gain. The processes that are involved in EBDM are widely researched (Kurniawan et al. 2011, Assadi et al. 2009). One of the main findings is that dopamine has a crucial role in the decision making process, (Wardle et al. 2011; Assadi et al. 2009; Kurniawan et al. 2011) and that people who have a decreased ability to experience pleasure (anhedonia) have a disturbance in EBDM (Treadway et al. 2012a). This disturbance seems to be directly related to the brain network that is involved in decision making (Smoski et al. 2009), this network consists of the dorsal anterior cingulate cortex (dACC) and the nucleus accumbens (NAc). Only for EBDM this relationship hasn’t been established clearly in a single study. Also until now EBDM has been researched with a paradigm that possibly isn’t fully representative for decision making in the real world (Wardle et al. 2009; Sherdell et al. 2012; Treadway et al. 2009, 2012a). Virtual reality can be used to create an virtual environment that surrounds the participant that is similar to the real world. Because of this virtual reality as research paradigm has a high ecological validity (Bohil et al. 2011). Possibly people will behave differently when an EBDM task is presented in virtual reality. The feeling of being present in the performed task could alter the way people weigh effort and reward. The next paragraph will discuss how virtual reality can influence decision making.
6
Presence & immersion
Presence is the subjective feeling of being present in one place when you’re physically present in another (Witmer & Singer, 1998). For virtual reality this means that people have the feeling they are present in the virtual environment instead of their real environment. Presence is partially dependent on attention, a stronger focus of attention on meaningful stimuli in the virtual environment will result in a higher experience of presence (Witmer & Singer, 1998). It is important for a higher experience of presence that sensory information from the physical environment are ignored or integrated in the virtual experience. The amount of presence that people experience is influenced by individual differences, factors like personality, (Sas, O'Hare & Reilly, 2003) attachment style, (Wallach, Safir & Almog ,2009) locus of control, susceptibility to the experience of dissociation (Murray, Fox & Pettifer, 2007) and anxiety all can influence the presence experienced.
Immersion is an important concept for presence. The amount of immersion is mainly
determined by the quality of the virtual simulation. This quality is determined by the extent to which the simulation forms a constants stream of input that surrounds and includes you, so for example the amount of freedom you have to move and look around or even interact with the environment (Slater, 1999). There is some discussion in the literature about the exact relationship between immersion and presence. According to Slater (1999) immersion is the objective description of the physical qualities of the VR system, in this definition immersion is independent of presence. According to Witmer and Singer (1998) these physical qualities have a strong influence on
immersion, but they state that immersion is also dependent on individual differences in experience, just like presence. Witmer and Singer (1998) define immersion more as a part of the presence experience. In this view it isn’t possible to part the subjective experience from the factors that influence it (Slater 1999). It is true that certain aspects of an immersive system can have a different effect on people, but this will eventually by expressed in individual differences in experienced presence. Also when there are differences in experience it doesn’t mean that the qualities of the system itself are different, an immersive system always has exactly the same qualities independent
7 of the person that is using it. Because of this it is more practical to see immersion as the quantifiable quality of the system and presence as the subjective experience evoked by this system (Slater 1999; Schubert, Friedmann en Regenbrecht, 2001), therefore I will use this definition for immersion. Cognition is a mediating factor between immersion and presence and more immersive system will result in a higher experience of presence. There are several factors influencing this, the experienced presence for example increases when there is more multimodal integration (Dinh et al. 1999), by adding more sense to the simulation people will experience it as being more realistic, for example illusory and real interactions with the virtual environment cause an increase in presence
(Regenbrecht & Schubert, 2002). Also manipulating real life objects and having an avatar (virtual body) brings the performance closer to performance in the real world (Lok et al. 2003).
The more realistic your simulation is the easier people will act as if they are in the real world. Because of this you can possibly study decision making processes more like they happen in real life by studying them in VR. Usually in the EBDM studies a relatively simple paradigm is used where participants choose between two alternatives presented on a screen. One alternative has a high reward for more effort and the other has a low reward for less effort. After choosing an alternative they have to repeatedly press a button to obtain the reward (Treadway et al. 2009). This paradigm contains every element that is needed for researching EBDM, the effort however might be a bit abstract. Of course button presses require effort, but it is more effort out of persistence than real physical effort. Also there is no context for the effort that is done, because of this it is possible that the results from this paradigm are not fully representative for real world decision making. By using a VR paradigm the task will be more realistic and the virtual environment will provide context for the exerted effort, this might evoke more representative behavior. This study will compare at EBDM in different immersive environments, this means that the same conceptual EBDM paradigm will be tested in different levels of immersion. There is almost no previous research on this topic, I will however try to provide a framework to explain how immersion and presence might influence decision making.
8 The way that information is presented influences the choices that people make. For example when you formulate the same information in another way people make different choices (Tversky & Kahneman, 1981) but also the presence of visual information can influence decision making (Gamliel et al. 2013). In the literature these effects are called framing (Tversky & Kahneman, 1981).
Comparing EBDM in different immersive environments can be seen as a similar manipulation. Objectively people get the same choice alternatives, they have to make the same amount of effort for the same reward, the only difference is that information is presented in different immersive environments. It could be possible that people will act differently depending on how the task is presented just like in the framing studies. By adding more senses and more realistic elements to the standard paradigm it is expected that people will be more present in the task environment and will experience the task as being more realistic. This will provide a context for the effort that has to be made, which will make the experience of this effort more real. Because of this the effort will weigh heavier in the decision process. The expectation is that by creating a subjective feeling of presence the perception of effort will increase which will result in a relative bias for the low effort alternatives. Before discussing the exact predictions and method, I will first discuss the underlying processes of EBDM and how these processes are disturbed in anhedonia.
9
Dopamine function and effort based decision making
Because dopamine and EBDM are intensely studied there is a lot of literature on these subjects. Since it is beyond the scope of this thesis to discuss all this literature separately I will discuss the main findings based on two review articles that discuss the neural correlates of EBDM in animals and humans.
The nucleus accumbens (NAc) & the dorsal anterior cingulate cortex (dACC) together make up a network that is involved in EBDM. This network has multiple functions that can be divided in executive and evaluative processes (Assadi et al. 2009). Execution processes are involved in gathering the needed resources for the execution of a decision. Evaluation processes are involved in
anticipating the outcome of a decision by doing a cost-benefit analysis (Assadi et al. 2009; Kurniawan et al. 2011). The aim of this analysis is to maximize the net profit by comparing the reward to the effort required to obtain that reward, so actually a choice alternative becomes less valuable when you have to make an effort for it (Kurniawan et al. 2011). Execution and evaluation processes are linked together in the dACC-Nac network, so here the action is linked to the outcome in the decision making process.
The dopamine level in this network is important for the outcome of decision making. People used to think that dopamine is a reward on itself and makes you feel good. However more recent research has pointed out that dopamine isn't directly associated with a feeling of pleasure (liking) but more with motivated and goal directed behavior (wanting) (Berridge & Kringelbach, 2007). Dopamine is a resource to overcome the costs of a certain action (Kurniawan et al. 2011), here dopamine functions as a representation of the potential reward. Higher dopamine levels bias the process of decision making to high reward and low levels of dopamine bias the process to low effort options. This idea is supported by research with rats in a T-maze (Salamone et al. 1994). In this research paradigm hungry rats are trained to make a choice between a high and low reward that are localized on opposite ends of a T-shaped maze. A barrier is placed in front of the high reward, so the rats have to make a greater effort to get to it. Normally the rats go for the high reward option, but after their
10 NAc is injected with a substance that blocks dopamine receptors they show a preference for the low effort options (Salamone et al. 1994). The same results are found when lesions are applied to the dACC, NAc or the connections between these areas (Assadi et al. 2009). Similar results are also found in humans, for example the individual differences in dopamine function in the ventral striatum (PET-scan) are positively correlated with the willingness to exert effort for rewards (Treadway et al. 2012b). Also people show a bias towards high reward options without an increase of reported liking of this reward after the administration of d-amphetamine, a substance that mimics the effect of dopamine in the brain (Wardle et al. 2011). This shows that dopamine is important in the EBDM process in humans, here dopamine is a resource to overcome costs and not a reward on itself. People tend to display apathy after damage to the dACC and/or the NAc. These patients are able to
experience pleasure and are also capable of executing action on command (Schmidt et al. 2008), but they aren't able to perform voluntary actions this is because these actions aren't associated with the potential gain (Kurniawan et al. 2011). This show that the dACC & NAc are important in connecting action (execution) to reward (evaluation).
11
Anhedonia and dopamine function in the effort based decision making network
There is a clear dichotomy between motivation and the experience of pleasure (Berridge and Kringelbach, 2007). This dichotomy is not only important for dopamine and EBDM but also for anhedonia. Anhedonia is traditionally defined as the reduced ability to experience pleasure, it is a frequent symptom in different psychiatric disorders, and an important condition that must be met for diagnosing depression (Klein, 1984 in Yang et al. 2014). Anhedonia is often seen as a disturbed processing of reward, but more recent research shows that a distinction has to be made between anticipatory and consummatory pleasure (Sherdell et al. 2012; Yang et al. 2014). Anticipatory pleasure is important for motivated and goal directed behavior (wanting) and consummatory
pleasure is important for pleasure on the moment itself (liking). In people that experience anhedonia consummatory pleasure is relatively intact, it is mainly the anticipatory pleasure that is disrupted. This is supported by the fact that people with a depression show a similar response to pleasurable stimuli (Dichter et al. 2010), but show a reduced anticipation of reward which predicts motivation (Sherdell et al. 2012). There is also a reversed relationship between anhedonia and the preparedness to exert effort for rewards in EBDM paradigms both in a healthy and depressed population
(Treadway et al. 2009). These results however do not make a distinction between anticipatory and consummatory pleasure. Sherdell et al. (2012) showed that the liking of a reward and the motivation for it are dissociated in people with a depression. In this group only anticipatory pleasure and not consummatory pleasure predicted motivation for rewards, so in this group the motivation for rewards isn't reduced because they don't like the reward but because they can't adequately anticipate it. This anticipatory problem is probably caused by a defect in the integration between action and outcome, because of this the costs of the action are weighted disproportionally to the reward of the action (Treadway et al. 2012a). This is supported by the fact that people who are depressed objectively exert the same amount of effort for a reward as a healthy controls but subjectively report a larger amount of effort (Cléry-Molin et al. 2011). This tells us that people who experience anhedonia recognize a potential reward as more desirable and are prepared to make
12 more effort for it, but this isn't translated to motor output, this illustrates the defect in the
connection of action to reward. Previously I stated that the dACC and NAc are mainly involved in this process and that dopamine plays a crucial role, therefore it is reasonable to think that the defect in anticipatory pleasure in people with anhedonia is related to the functioning of dopamine in this neural network.
Both anticipatory anhedonia and variations in dopamine are related to differences in motivation in EBDM (Kurniawan et al. 2011; Wardle et al. 2009; Treadway et al. 2012b). Now the question is if these variations in dopamine can explain the differences in anticipatory anhedonia. In the previous articles this link is implied but not directly established. There seems to be a direct link between anhedonia and dopamine, for example the genes that influence the availability and binding of dopamine in the synapse are predictive for depressive symptoms in healthy and depressed populations (Pearson-Fuhrop et al. 2014). Also deep brain stimulation in the NAc relieves depressive symptoms in treatment resistant depressive patients (Bewernick et al. 2012) and people with a depression show a decreased reaction in the striatum and dACC during anticipation, selections of reward (Smoski et al. 2009). But these findings do not directly show a link between dopamine and anticipatory anhedonia in EBDM because the participants didn't have to exert effort for the reward, up till now this link has not been researched in the scope of one study.
The current study has two goals. First the goal is to establish a relationship between anticipatory anhedonia and dopamine in EBDM, here the hypothesis is that there is an inverse relationship between anticipatory anhedonia and the willingness to exert effort for rewards. The dopamine level in the NAc and dACC should be predictive for both anticipatory anhedonia and willingness to exert effort, the dopamine level should positively correlate with the willingness to exert effort and negatively with the amount of anticipatory anhedonia. Unfortunately there are not enough recourses and money for this study to use an accurate measure of dopamine levels like a PET-scan. However a recent study has shown that auditory working memory is predictive of
13 approximate dopamine levels by measuring auditory working memory. The reliability of this method is of course is a bit questionable, however because of the nature of this study this is the best option to estimate dopamine levels that might be related to EBDM and anhedonia.
The second goal of this study is to investigate to what extent people weigh effort and reward in the same way in different levels of immersion. In the literature it is implicitly assumed that the paradigm that is used to research EBDM is the way to look at EBDM processes. But the paradigm used can be a bit abstract for the participants. The effort that has to be made is in button presses and there is no contextual information (Treadway et al. 2009). It is interesting to see if there is a different response when the same task is reframed in a more immersive environment. It is know that a more immersive system causes people to experience more presence in the task (Slater, 1999; Schubert, Friedmann en Regenbrecht, 2001 ) and that it can bring performance closer to real-world
performance (Lok et al. 2003). Because of the increased presence the task in virtual reality provides a better context for the exerted effort, which makes it possible that the subjective experience of this effort also increases. If the perceived effort increases with immersion it would mean that people are less willing to exert effort for the same reward in a more immersive environment, this would mean that people would need a bigger reward to go for high effort options in an EBDM paradigm in VR compared to the same paradigm in 2D or 3D. The prediction is that people show a relative bias to low effort options in a virtual environment with respect to a 3D or 2D environment.
14
Method
Participants
The procedure was approved by the ethics commission of the university of Amsterdam. In total 49 people participated (32 female). Most of the participants were undergraduate psychology students, some participated for research credit but most participated for a €20 payment. There were also some participants that didn't participate for payment. All participants received € 0.50 extra reward. Exclusion criteria were epilepsy and other neurological disorders, heart disease, sensitivity to dizziness or nausea in rollercoaster or carousels and wearing glasses for eyesight correction. All the participants read and signed an informed consent beforehand.
Materials
There were three within-subject conditions, the 2D condition, 3D condition and the VR condition. All conditions were programmed and executed in Unreal Engine 4 (Epic Games, 2012). All
conditions were completed on the same computer with a Xbox ONE controller (Microsoft, 2013). The 2D and 3D condition were presented on a 1680*1050 resolution monitor at a refresh rate of 60Hz. The VR condition was presented on an
Oculus Rift DK2 (Oculus VR, 2014) at a refresh rate of 75Hz. The underlying EBDM mechanism is the same for each condition. Each trial started with two choice alternatives being presented to the participant. One option represented high reward/ high effort and the other low reward/ low effort. In the 2D condition the task was schematically represented (figure 1). The
participants had to place their closed fist on top of the thumb sticks of the controller. After the participant made their choice they had to move the two thumb sticks in opposite directions along the forward-backward axis. By Figure 1. Representation of the choice
alternatives in the 2D task.
Figure 2. Representation of the choice alternatives in the 3D and virtual reality task.
15 continuing these movements they had to keep the power bar displayed filled above the indicated level. When this condition was met they would make progression in the task. The different colored blocks represent different amounts of effort that had to be exerted to keep the power bar above the indicated level. Red represented high effort, orange medium effort and green low effort. The choice alternatives were constructed in a way that they both took the same amount of time to complete, this was made clear to the participants. In the 3D and VR condition the choice alternatives were represented in a similar way, but instead of a schematic representation of the progress the
participants had a first person perspective in a small cart with two handlebars that could move the cart across a track of rails (figure 2). When the participant would move the thumb sticks the
handlebars would move as well and when the power bar was filled above the indicated level the cart would move across the tracks. The trials would start by picking one of the choice alternatives, after this the cart would move through the doors and the cart would move to the track the participant chose and the participant had to move the thumb sticks to progress. The high effort parts of the track were represented by bushes that obstructed the cart. After completion of each trail the cart returned to the same room and a new trial would start. Based on the choice that was made the rewards were adapted. When the participant chose the high reward option (left) the difference between the rewards was reduced, if the participant chose the low effort option (right) the difference was increased. Based on a stair casing procedure similar to previous studies (Tversky & Kahneman, 1992) the individual point of indifference (POI) for each participant will be determined per condition. The POI reflects the weighing of effort in respect to reward for a specific person in a specific condition, when for example someone has an higher POI he or she needs a larger reward to be willing to exert the same amount of effort as someone with a lower POI. The POI represents the relationship between perceived reward and perceived effort. Because the reward is visualized in the same abstract way in each condition we assume that perceived reward is constant. If the perceived reward is constant, the differences in POI are due to differences in perceived effort. So perceived effort per condition will be measured by the POI. As a control perceived effort will also be assessed with the
16 use of a self-report scale (Borg, 1998). Besides the experimental conditions the participants had to complete a relative large set of self report questionnaires. Anticipatory and consummatory pleasure were measured by a translated version of the Temporal Experience of Pleasure Scale (TEPS; Gard et al. 2006). The scale consists of 18 items of which 10 measure anticipatory pleasure and 8
consummatory pleasure. Auditory working memory as approximation of dopamine synthesis capacity in the striatum was measured by the listening span, (Vos et al. 2001) which was presented through a headphone. Based on the mean the participants were be divided into two roughly equal groups, a high-span (N=18) and low- span group (N=20). Presence was assessed by the Igroup Presence Questionnaire (IPQ; Schubert et al. 2001). Liking of the different experimental conditions was assessed by a liking scale (Lin et al. 2012) The rest of the questionnaires are unrelated to this study. These were the Immersive Tendencies Questionnaire (ITQ; Witmer & Singer, 1998), a locus of control questionnaire (Rotter, 1966), a self constructed social-economic status questionnaire, the Short Questionnaire to Assess Health-enhancing physical activity (SQUASH; Wendel-Vos et al. 2003), a self constructed game experience questionnaire and a simulator sickness questionnaire.
Procedure
At arrival in the lab the participants first read and signed an informed consent form. Subsequently they would get verbal instruction about the task they were about to complete. There were two blocks of both approximately one hour, the order of the blocks was counterbalanced. One block consists of the listening span and most of the questionnaires. The other block consists of the three experimental conditions, the IPQ, the simulator sickness questionnaire and the liking and perceived effort scales. The order in which the conditions were presented was counterbalanced. In the instructions for the tasks it was made clear that the participants could receive an extra reward based on their choice in the conditions. After the participants complete the three conditions they filled out the questionnaires. When both blocks were completed the participants received their payment and bonus reward of €0.50.
17
Results
The first 9 participants were excluded from analysis because the instructions were changed after they participated. One participant was excluded because she did not complete all three conditions. Another participant was excluded because the self reported questionnaires he/she completed weren’t saved due to an error. After these removals there were 38 participants left (26/38 female). All data was further analyzed using IBM SPSS statistics for Windows version 20.0 (IBM Corp, 2011). Four participants did not complete the TEPS questionnaire due to a system error, these participants will be excluded for analysis that include the TEPS but still will be used for the other analysis. Table 1 shows the number of participants that completed each measure and the mean and standard deviation of each measure. Many participants had an average POI over all three conditions of near zero (N=16). This means that their POI probably was not accurately assessed by the
procedure that was used, their willingness to exert effort for the reward was higher than the highest amount they could exert, this means that there was a ceiling effect for these participants. Because of this the POI results were analyzed both with (N=38) and without these participants (N=22).
Table 1. number completed (n), mean and standard deviation (SD) for all measures used.
n Mean SD
Age 38 23.29 2.93
Anticipatory scale of the TEPS 34 36.06 5.40
Consummatory scale of the TEPS 34 37.56 5.09
Listening span score 38 4.36 0.97
Perceived exertion 2D 38 1.5 1.61 Perceived exertion 3D 38 1.56 1.17 Perceived exertion VR 38 2.26 1.58 Point of indifference 2D 38 10.45 15.86 Point of indifference 3D 38 9.84 15.98 Point of indifference VR 38 11.39 18.48 IPQ 2D 38 21.58 6.20 IPQ 3D 38 33.32 9.01 IPQ VR 38 48.42 7.20 Liking 2D 38 11.24 4.77 Liking 3D 38 15.05 4.22 Liking VR 38 22.53 7.83
18 The cut-off for the listening span groups was based on the mean, this resulted in a high-span group (N= 18) and a low-span group (N=20). On average the participants in the low span didn't differ on the consummatory scale of the TEPS (M = 36.39, SE = 5.135), compared to those in the high span group (M = 38.88, SE = 4.87). The difference, 2.49, BCa 95% CI [5.994, 1.022], was not significant t(32) = -1.443, p = .159. On average the participants in the low span didn't differ on the anticipatory scale of the TEPS (M = 36.22, SE = 5.68), compared to those in the high span group (M = 35.88, SE = 5.25). The difference, 0.347, BCa 95% CI [-3.492, 4.186], was not significant t(32) = 0.184, p = .855. The consummatory scale of the TEPS was significantly related to the average perceived exertion, r = - .398, p = .02, and the anticipatory scale was not significantly related to the average perceived exertion, r = - .07, p = .695.
Effect of immersion on presence
Mauchly’s test indicated that the assumption of sphericity has not been violated for the main effect of condition on the IPQ score, 𝜒𝜒2(2) = .858, p = .651. Therefore degrees of freedom were not corrected. There was a main significant effect of the condition on the amount of presence
experienced, F(1.65, 60.88) = 300.596, p < .001. Planned contrasts revealed that the presence in the VR condition was significantly higher than in the 2D condition, F(1, 36) = 631.214, p < .001, r = .97, and in the 3D condition, F(1, 36) = 227.344, p < .001, r = .927.
Figuur 1. Effect of immersion on presence, error bars represent the standard error of the mean, as computed over within-subject normalized data.
19 Effect of condition on self-reported perceived effort
Mauchly’s test indicated that the assumption of sphericity had been violated for the main effect of condition on perceived effort, 𝜒𝜒2(2) = 13.977, p = .001. Therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (ε = .757 for the main effect of
condition). There was a main significant effect of the condition on self-reported perceived effort, F(1.51, 55.98) = 5.172, p = .015. Planned contrasts revealed that the perceived effort in the VR condition was significantly higher than in the 3D condition, F(1, 37) = 5.732, p = .022, r = .37, and the 2D condition, F(1, 37) = 6.752, p = .013, r = .39.
Figuur 2. Effect of immersion on perceived effort, error bars represent the standard error of the mean, as computed over within-subject normalized data.
20 Effect of condition on reported liking
Mauchly’s test indicated that the assumption of sphericity had been violated for the main effect of condition on liking, 𝜒𝜒2(2) = 25.63, p = .001. Therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (ε = .663 for the main effect of condition). There was a main significant effect of the condition on reported liking, F(1.33, 49.029) = 38.915, p < .001. Planned contrasts revealed that the liking of the VR condition was significantly higher than of the 3D condition, F(1, 37) = 48.75, p < .001, r = .75, and the 2D condition, F(1, 37) = 29.157, p = .001, r = .66.
Effect of condition on POI
For the effect of condition on POI I conducted two analysis: both including and excluding the participants who had an average POI of near zero.
Including: Mauchly’s test indicated that the assumption of sphericity has not been violated for the main effect of condition on point of indifference, 𝜒𝜒2(2) = 4.22, p = .121. Therefore degrees of freedom were not corrected. The main effect of condition on point of indifference was not
Figuur 3. Effect of immersion on reported liking, error bars represent the standard error of the mean, as computed over within-subject normalized data.
21 significant, F(2, 74) = 0.131, p = .878. Planned contrast revealed that the POI in the VR condition was not significantly higher than the POI in the 2D condition , F(1, 37) = 0.08, p = .78, and the 2D
condition, F(1, 37) = 0.384 , p = .54.
Excluding: Mauchly’s test indicated that the assumption of sphericity has not been violated for the main effect of condition on point of indifference, 𝜒𝜒2(2) = 2.408, p = .300. Therefore degrees of freedom were not corrected. The main effect of condition on point of indifference was not
significant, F(2, 42) = 0.086, p = .918. Planned contrast revealed that the POI in the VR condition was not significantly higher than the POI in the 2D condition , F(1, 22) = 0.022, p = .88, and the 2D condition, F(1, 37) = 0.254 , p = .62.
Figuur 4. Effect of immersion on POI including non-repsonsive patricipants, error bars represent the standard error of the mean, as computed over within-subject normalized data.
22 Figuur 5. Effect of immersion on POI excluding non-responsive participants, error bars represent the standard error of the mean, as computed over within-subject normalized data.
Relationship between POI and presence
To look at the relationship between POI and presence the differences between the VR condition and 3D condition and the difference between the 3D condition and 2D condition were computed. A Pearson’s correlation was performed on these scores. The results are displayed in table 2 including the participants with an average POI of near zero and in table 3 excluding these
participants. There was no significant relationships between the differences between IPQ VR and 3D, IPQ 3D and 2D , presence VR and 3D and presence 3D and 2D. Neither for the data including the non-responsive participants not the data including the non-non-responsive participants.
Table 2 Correlation matrix of the computed differences of the IPQ and POI between the VR and 3D and 3D and 2D conditions, for the data including the participants with an average POI of near zero.
POI VR-3D POI 3D-2D IPQ VR-3D POI 3D-2D Pearson’s r - .546 1 Sig. (2-tailed) .028 N 37 37 IPQ VR-3D Pearson’s r .068 - .052 1 Sig. (2-tailed) .691 .760 N 37 37 37 IPQ 3D-2D Pearson’s r .041 .181 - .343 Sig. (2-tailed) .810 .284 0.037 N 37 37 37
23 Table 3 Correlation matrix of the computed differences of the IPQ and POI between the VR and 3D and 3D and 2D conditions, for the data excluding the participants with an average POI of near zero.
POI VR-3D POI 3D-2D IPQ VR-3D POI 3D-2D Pearson’s r - .339 1 Sig. (2-tailed) .123 N 22 22 IPQ VR-3D Pearson’s r .103 .034 1 Sig. (2-tailed) .647 .882 N 22 22 22 IPQ 3D-2D Pearson’s r -.067 .205 - .623 Sig. (2-tailed) .766 .361 .002 N 22 22 22
Table 4. Correlation matrix of the C-TEPS, A-TEPS, Listening span and average POI
C-TEPS A-TEPS Average POI
A-TEPS Pearson’s r .313 1
Sig. (2-tailed) .179
N 20 20
Average POI Pearson’s r .281 .203 1
Sig. (2-tailed) .230 .882
N 20 20 20
Listening Span Pearson’s r .169 - .023 - .132
Sig. (2-tailed) .477 .922 .578
N 20 20 20
Relationship between POI, TEPS and listening span
To test the relationship between POI, TEPS and the listening span a Pearson’s correlation was performed on the anticipatory and consummatory scale of the TEPS, the listening span and the average POI over the three conditions. This analysis was only performed on the data excluding the participants with an average POI of near zero. The results of this analysis are displayed in table 4. There were no significant relationships between the consummatory and anticipatory scale of the TEPS, listening span and the average POI.
24
Discussion
The present study had two goals. Firstly the goal of this study was to establish a relationship between individual differences in anticipatory anhedonia and dopamine in de EBDM network, this relationship was not found. There are several possible explanations why the predicted effects were not established. The main explanations are as follows: (1) auditory working memory is not a good estimation of dopamine related to EBDM, (2) the TEPS could not assess individual differences in anhedonia properly for this research question and (3) the EBDM paradigm that was used was not properly designed to measure individual differences in weighing of effort in respect to reward. I will discuss these explanations in detail. As you might notice all these explanations are focused on the methodology of the study and not on the theory that was put forward. The main reason for this is the fact that findings of previous studies on EBDM were not replicated, because of the nature of this study (it being a study for a bachelor project and a relative new field) is more likely that something went wrong in the execution or measurement than that these previous findings are to be
reconsidered. Because the previous findings were not replicated it is hard to make any substantial claims about the theory and predictions based on this study, that’s why the main focus of this discussion will be on the methodology.
(1) Auditory working memory as estimate of dopamine levels.
It was not a big surprise that the listening span didn’t yield any significant results. Although there is evidence from a previous study that the listening span is predictive for dopamine synthesis capacity in the striatum (Cools et al. 2007). This however does not necessarily mean that you can measure dopamine levels using the listening span. Even when the listening span would be an highly accurate measure of dopamine levels in the striatum there are still several explanations for the insignificant results. Firstly the participants scored relatively high on average compared to other studies (Vos et al. 2001; Cools et al. 2008), maybe there wasn’t enough variation in the sample to measure individual differences in dopamine levels in the striatum that are relevant for EBDM and anhedonia. Another possibility is that that the listening span did accurately measured individual
25 differences in dopamine levels in the striatum but that these levels are not involved in the neural network that is involved in EBDM and anhedonia. It is true that the NAc is located in the ventral part of the striatum, but it might as well be true that the dopamine synthesis capacity that was related to the listening span score in the study of Cools et al. (2008) was not directly related to the dopamine that is involved in the neural network discussed previously. Overall it would have been interesting to find differences in POI and/or anhedonia scores between people with a high listening span and a low listening span, but it was not surprising that this wasn’t the case. What was more surprising was that neither consummatory nor anticipatory pleasure were predictive for differences in EBDM. The prediction was the anticipatory pleasure would be more predictive for EBDM then consummatory pleasure, but neither were predictive. This finding could have multiple explanations, there could be a problem in the measurement of anticipatory and consummatory pleasure, or in the measurement of EBDM.
(2) Measurement of anticipatory and consummatory pleasure.
There is the possibility that anticipatory and consummatory pleasure weren’t measured by the TEPS. Although this certainly is possible it isn’t the most likely explanation, the TEPS is a well validated questionnaire that is regularly used to measure anticipatory and consummatory pleasure, both in EBDM (Yang et al. 2012) as in other research paradigms (Gard et al. 2006). What however is a more likely possibility is that the TEPS is not sensitive enough to measure individual differences in anhedonia that can be related to EBDM in a healthy population. In previous EBDM studies the TEPS was used to compare clinically depressed patients with a healthy control group (Yang et al. 2012) but not to investigate individual differences in only a healthy population. There however is previous research that established a relationship between EBDM and individual differences in anhedonia in a healthy population (Treadway et al. 2009) but this study used the Snaith Hamilton Pleasure Scale (SHAPS, Snaith et al. 1995) to measure anhedonia . Maybe the SHAPS is a more sensitive instrument than the TEPS in detecting variation in anhedonia that will be reflected in EBDM. The SHAPS was not
26 used in this study because it doesn’t differentiate between anticipatory and consummatory
anhedonia and it takes a relatively long time to administer.
(3) The procedure used wasn’t properly designed to measure effort based decision making.
There are multiple explanations that probably all contributed more or less to the fact that maybe the paradigm used didn’t measure EBDM correctly. Firstly the reward was relatively low, the participants were told they could receive a maximum of €0,50 based on their choices in the
paradigm. After they received the reward most participants did not seem to react as if they were really excited about it. It is possible that the participants weren’t motivated to obtain this relatively small reward, this would result in choices that aren’t based on the potential reward, which would mean that the POI’s didn’t reflect an EBDM process. Another possibility is that the amount of effort was not high enough. The thumb sticks of the Xbox controller almost have a weak resistance this makes it relatively easy to move them, this is reflected in the fact that the perceived exertion scores that were overall low, so it is possible the participants didn’t feel like they had to make an effort. In the original design of this study it was planned to use joysticks instead of the Xbox controller, however the task could not be programmed in time to work with the joysticks. The use of these joysticks would have increased the effort that had to be exerted, in combination with a larger reward this might have resulted in a better estimation of the POI. The combination of a relative small reward and low effort could have resulted in participants not making decisions based on the weighing of reward and effort. This reflected in the fact that there were a lot of participants who didn't seem to respond to the manipulation. They had an average POI of near zero. This means that they always choose the high reward options and ignored effort. This means that their POI is not based on a decision making process were effort and reward are weighted.
27 Immersion
The second goal of this study was to establish to what extent people weigh effort and reward in the same way in different levels of immersion. The prediction was that people would have a higher POI in the VR condition than in the 3D and 2D condition. The manipulation of immersion was
effective, people experienced higher presence in the VR condition than in the 3D and 2D condition. This means that the qualities of the immersive system were adequately inducing a sense of being present in the performed task. There was however no significant difference between the 2D, 3D and VR condition, this would mean that people do not weigh effort and reward differently in different levels of immersion. The amount of presence experience also wasn’t correlated with the POI’s, so there is no relationship between presence and individual differences in EBDM. This would mean that it doesn’t matter what paradigm you use to measure the weighing of effort and reward. There are however some factors that could also have caused these findings, the main explanation for these findings is the possibility that the paradigm used wasn’t properly designed to measure EBDM, this is already discussed in the previous paragraph. Another contributing factor can be the fact that the verbal instructions were given by varying experimenters, although the instructions were roughly standardized there was some variation in the exact instructions that the participants received. Because of this it is possible that some participants did not fully understand the instructions, this might result in participants making choices at random instead of basing their choices on the reward and effort.
Because there are so many factors that could have contributed to the results it’s difficult to make any claims about if immersion has no influence whatsoever on EBDM or that there is an effect but that this effect wasn’t found due to methodological problems. This is of course always the case when you do not find significant results but this paradigm is new so it is more likely that there were errors due to methodology that obstructed the testing of the theory. This however doesn’t mean that the effect exist but wasn’t found, it means that we can’t make any substantial claims about the influence of immersion on EBDM at this point in time.
28
Conclusion
This study yielded disappointing results, it was predicted that there would be a relationship between individual differences in anhedonia and EBDM, and that dopamine would be predictive for this relationship. It was also predicted that more immersive environments would lead to an increase of perceived effort. These relationships were however not found in any way. There is a high
probability that at least part of this can be explained by faults in the method and execution of this study and not by the theory itself. Based on previous research it is still highly probably that individual differences in levels of dopamine in the neural network associated with EBDM are predictive for individual differences in anticipatory anhedonia. This however is up till now not proven in the scope of one study. Because it is probable that the methodology of this study was not sufficient to establish the relationship between dopamine and anticipatory anhedonia I will include a research proposal that might be better designed to do so. For the predicted effect of immersion on EBDM it is much harder to make any claims, since there is almost no previous research on this topic. But this doesn’t mean it’s not interesting and useful to explore the possibilities and limitation of VR research paradigms in the future.
29
Research proposal
It isn't worth the effort:
The relationship between anticipatory anhedonia and dopamine in decision making processes.
In the previous study it was proposed that there is a direct relationship between the reduced ability to anticipate on potential rewards (anticipatory anhedonia) and individual differences in dopamine function in the brain network that is responsible for weighing effort and reward in decision making processes. This network consist of the nucleus accumbens (NAc) and de dorsal anterior cingulated (dACC). In the NAc-dACC network reward and action are connected, in this network dopamine represents the potential reward and is a resource to overcome effort in the decision making process (Kurniawan et al. 2011). It was proposed that individuals who experience anhedonia have a reduced function of dopamine in these regions. This relationship between dopamine and anticipatory anhedonia however was not established, it is highly probable that the possibility of finding of these relationships was obstructed by methodological errors. These
methodological errors were mainly caused by the nature of the study, it was a study for a bachelor project. This means that there were limited time and resources to properly asses the research question. In this research proposal I will define two experiments that might be better suited to investigate the relationship between anticipatory anhedonia and dopamine function in the effort based decision making network. Based on previous research it is already known that anticipatory anhedonia is predictive for a reduced motivation for rewards in effort based decision making (EBDM) paradigms (Yang et al. 2012; Sherdell et al. 2012). It is also known that dopamine function is
predictive for motivation for rewards in EBDM paradigms (Wardle et al. 2011). From these findings the link between anticipatory anhedonia and reduced dopamine levels is easily made. There has been research that implies a direct link between (anticipatory) anhedonia and dopamine function both in clinically depressed and healthy subjects (Padrao et al. 2013; Tye et al. 2013), but here hasn't
30 been a study that investigates this relationship directly in humans with an experimental measure of the preference for reward versus effort. The establishment of this relationship however can have important implication for our knowledge and treatment of major depressive disorder (MDD) which will be discussed later.
It is possible that people who experience more anticipatory anhedonia have less dopamine available in the brain network related to EBDM, but it is also possible that people who experience more anticipatory anhedonia have the same amount of dopamine available but are less responsive to the effects of it. Because it isn't clear if availability, responsiveness or both are related to anticipatory anhedonia and EBDM this study will investigate the variation between individuals in the availability as well as the individual differences in response to dopamine, this will be done with two
experiments. The first experiment will investigate if individual differences in dopamine responsivity are related to individual differences in anticipatory anhedonia and effort based decision making in a healthy population. This will be done by looking at the effects of the administration of d-amfetamine. The effects will be measured at the behavioral level as well as at the neuronal level (positron
emission tomography, PET). The second experiment will compare dopamine availability (PET), EBDM and anticipatory anhedonia in people who are diagnosed with MDD with healthy subjects.
31
Method
Participants
In study 1 30 healthy subjects will participate (50% female). In study 2 about 50 subjects will participate (50% female), 25 with a diagnosis of major depressive disorder (MDD) and 25 healthy subjects.
Materials
In both studies the individual differences in the weighing of effort and reward will be measured by the Effort Expenditure for Rewards Task (EEfRT; Treadyway et al. 2009) the task is based on the same concept as the classical T-maze paradigm that is commonly used in EBDM research with rats (Salamone et al. 1994). In study 1 individual differences in dopamine responsivity will be assessed similar to a previous study by Treadway et al. (2012b). Positron emission
tomography (PET) with [18F] Fallypride after administration of d-amphetamine will be contrasted with PET after administration of a placebo for each participant. [18F] Fallypride is a D2/D3 -specific ligand that binds to striatal and extrastriatal regions (Treadway et al. 2012b), this allows to look at cortical as well as striatal areas. By contrasting the d-amphetamine scan with the placebo scan you show how and where the brain responds to dopamine. In study 1 only healthy subjects will
participate since it is unethical to give currently depressed patients a pleasurable substance that can be addictive. In study 2 individual differences in dopamine synthesis will be measured by PET with 18F- DOPA which assesses presynaptic dopamine synthesis in striatal and extrastriatal regions (Egerton et al. 2011). The PET of the MDD group will be contrasted with the PET of the healthy subjects which will show where these groups differ in synthesis capacity. In both studies
consummatory anhedonia will be measured by the Snaith-Hamilton Pleasure Scale (SHPS; Snaith et al. 1995) and the consummatory pleasure scale of the translated version of the Temporal Experience of Pleasure Scale (TEPS; Gard et al. 2006). Anticipatory anhedonia will be measured by administering
32 the Hamilton Rating Scale for Depression (HAM-D; Hamilton 1967) and the anticipatory pleasure scale of the translated version of the Temporal Experience of Pleasure Scale (TEPS; Gard et al. 2006).
Procedure
In study 1 the participants will complete three sessions. There will be a week between each session. In the first session the participants will fill out the questionnaires and complete the
interviews. In the last two session the scans and the EEfRT will be completed, the administration of amphetamine will be double blind placebo controlled. The participants will first get either the d-amphetamine or the placebo, each participant will received 0.43mg/kg d-d-amphetamine. The peak of the drug effect is three hours after administration(Treadway et al. 2012b), the scans and EEfRT will be administered around this peak. Two hours after administration the participants will complete the EEfRT, the scans will be made three hours after administration of the drug. The scans will be made after injection of 5.0 mCi slow bolus injection of [18F] Fallypride.
In study 2 the participants will complete two sessions. In the first session the scan will be completed, in the second the EEfRT, questionnaires and interviews. The method of this scan will be similar to a previous study by Egerton et al. (2010). 30 seconds after the scan starts 150 MBq of 18F-DOPA will be administered by bolus injection.
33
Data-analyses & predictions
In study 1 there will be multiple analyses. First the scores on the EEfRT after administration of d-amphetamine will be compared within subjects to the scores after the administration of the placebo.
It is expected that after d-amphetamine the participants will show a relative bias to higheffort/ high reward options compared to the placebo condition. The PET scan data will be correlated to the scores on the anhedonia scales and the EEfRT. It is expected that anticipatory anhedonia and not consummatory anhedonia will correlate negatively with dopamine responsivity, especially in the NAc
and dACC. It is also expected that anticipatory anhedonia and not consummatory anhedonia will negatively correlate with willingness to exert effort for reward in the EEfRT. And it is expected that
the EEfRT in the placebo condition will correlate positively with the contrasted scan, so higher dopamine reactivity shows a higher willingness to exert effort in the EEfRT. The difference on the EEfRT between the d-amphetamine and placebo conditions will be correlated to the PET data and anhedonia scales. Study 2 will compare the average group scores on the EEfRT and anhedonia scales
between the MDD group and the control group. It is expected that the MDD group will show a reduced motivation to exert effort compared to the control group. It is also expected that the control
group will score lower on anticipatory anhedonia than the MDD group. Also the groups will be compared on presynaptic dopamine levels. It is expected that the control group will have a higher
overall presynaptic dopamine level than the MDD group. This difference should especially be present in the NAc and dACC. These levels should be related to the anhedonia scores and willingness
34
Interpretation of the results
These two studies have been designed to establish a direct relationship between anticipatory anhedonia and dopamine function in the EBDM network. Study 1 will try to do this for dopamine responsivity in a healthy population and study 2 will compare dopamine synthesis in subjects with MDD with healthy subjects. Together these studies could show if and how the relationship between anticipatory anhedonia and dopamine function expresses itself throughout the whole distribution of motivational levels. The establishment of this relationship can have important implication for our knowledge and treatment of major depressive disorder. Anhedonia is a key symptom in major depressive disorder and is proposed to be a endofenotype for depression. If this anticipatory deficit can be directly linked to the function of dopamine in the effort based decision network this would mean that dopamine dysfunction could be one of core neurological deficits of major depressive disorder. This would have important implications for the psychopharmacological treatment of depression, it would mean that treatment of depression should not only focus on the serotonin function of depressed individuals but also on the dopamine function and/or availability. These treatments could prove to be effective especially for those patients who are resistant to the commonly used treatments, which mainly focus on the availability of serotonin.
Psychopharmacological treatments that target the availability of dopamine could be, for example methylphenidate or d-amphetamine. It has already been proven that these psycho stimulants can be effective in reducing depressive symptoms in patients with treatment resistant depression (Stoltz et al. 1999), but there has been no direct explanation for these results. The results of this study could help in designing psychopharmacological treatments that are based on specific characteristics of individual patients. For example when a anticipatory deficit is the core problem you could use a dopamine based treatment and when a consummatory deficit is the core problem you could use a serotonin based treatment, or a combination of both. The role of dopamine in depression could also help explain how serotonin based treatments work. Serotonin based treatments have a direct effect on serotonin but the cognitive effects take a few weeks. Dopamine could prove informative in
35 explaining this gap between neurological and cognitive effect. There probably is a complex
interaction between serotonin and dopamine in mood, motivation and decision making (Cools et al. 2010). This might also help explain why some people don't seem to react to the treatment.
It is however also possible that this study won't establish a relationship between anticipatory anhedonia and dopamine in the EBDM network. This could mean that people who experience anticipatory anhedonia don't necessarily have reduced dopamine function. This could mean that reduced motivation for rewards in people with and without MDD is possibly regulated by (a defect in) a different brain network. The EBDM network could be intact but for example the outcome of this decision making network could be inhibited or ignored by another process which will result in the same reduced motivation on the behavior level that can result from dopamine depletion but isn't the same deficit on the neural level.
36
References
Alsina-Jurnet, I., & Gutiérrez-Maldonado, J. (2010) Influence of personality and individual abilities on the sense of presence experienced in anxiety triggering virtual environments. Human-Computer Studies, 68, 788-801.
Assadi, S. M., Yücel, M., & Pantelis C. (2009). Dopamine modulates neural networks involved ineffort-based decision-making. Neuroscience and Behavioural Reviews, 33, 383-393.
Berridge, K. C., & Kringelbach, M. L. (2007) Neuroscience of affect: brain mechanisms of pleasure and displeasure. Current Opinion in Neurobiology, 23, 294–303.
Bewernick, B. H., Kayser S, Sturm V, Schlaepfer T. E. (2012). Long-term effects of nucleus accumbens deep brain stimulation in treatment-resistant depression: evidence for sustained efficacy. Neuropsychopharmacology, 37, 1975–1985.
Bohil, C., Alicea, B., & Biocca, F. A. (2011) Virtual reality in neuroscience research and therapy. Nature Reviews: Neurosciene, 12, 752-762.
Borg G. A. V. (1998) Borg’s rating of perceived exertion and pain scales. Champaign, IL: Human Kinetics, viii-104.
Cléry-Melin, M., Schmidt, L. , Lafargue, G., Baup, N. , Fossati, P., Pessiglione, M., & Laks, J. (2011). Why don't you try harder? an investigation of effort production in major depression, PLoS ONE, 6, e23178.
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, 28, 1208-1212.
Cools, R., Nakamura, K., & Daw, N. D. (2010). Serotonin and Dopamine: Unifying affective, activational and decision functions. Neuropsychopharmacology, 36, 98-113.
Dichter, G.S., Smoski, M.J., Kampov-Polevoy, A.B., Gallop, R., & Garbutt, J.C. (2010).Unipolar depression does not moderate responses to the Sweet Taste Test. Depression and Anxiety, 27, 859–863.
Dinh, H., Walker, N., Hodges, L., Song, L., & Kobayashi, A. (1999) . Evaluating the importance of multi-sensory input on memory and the sense of presence in virtual environments. IEEE Virtual Reality,222–228.
Egerton, A., Demjaha, A., McGuire, P., Mehta, M. A., & Howes, O. D. (2010). The test–retest reliability of 18F-DOPA PET in assessing striatal and extrastriatal presynaptic dopaminergic function. NeuroImage, 50, 524-531.
Epic Games. (2012). Unreal Engine 4. Retrieved from https://www.unrealengine.com/unreal-engine-4.
37 display and attribute representation in attenuating framing bias. Judgement and decision Making, 8, 482-491
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. J. Res. Pers., 40, 1086–1102
Hamilton, M. (1967). Development of a rating scale for primary depressive illness. The British journal of social and clinical psychology, 6, 278–296.
IBM Corp. (2011). IBM SPSS Statistics for Windows Version 20.0. retrieved from www-01.ibm.com/software/analytics/spss/.
Knutson, B., Bhanji, J. P., Cooney, R. E., Atlas, L. Y., & Gotlib, I. H. (2008). Neural responses to monetary incentives in major depression. Biological Psychiatry, 63(7), 686-692 Kurniawan, I. T., Guitart-Masip, M., & Dolan, R. J. (2011) Dopamine and effort based decision
making. Frontiers in Neuroscience, 5, 1-10.
Lin, J. W., Duh, H. B. L., Parker, D. E., Abi-Rached, H., & Furness, T. A. (2002). Effects of field of view on presence, enjoyment, memory, and simulator sickness in a virtual environment. In Virtual Reality, 2002. Proceedings. IEEE (pp. 164-171). IEEE.
Lok, B., Naik, S., Whitton, M., & Brooks Jr., F. P. (2003). Effects of handling real objects and avatar fidelity on cognitive task performance and sense of presence in virtual enviroments. Presence: Teleoperators & Virtual Environments, 12, 615-628.
Microsoft. (2013). Xbox One Wireless Controller. Retrieved from www.xbox.com/en-US/xbox- one/accessories/controllers/wireless-controller.
Murray, C. D., Fox, J., & Pettifer, S. (2007). Absorption, dissocation, locus of control and presence in virtual reality . Computers in Human Behaviour, 23, 1347-1354.
Oculus VR. (2014). Oculus DK2. Retrieved from https://www.oculus.com/dk2/.
Opstal, F., Van Laeken, N., Verguts, T., van Dijck, J., De Vos, F, & Goethals, I. (2014). Correlation between individual differences in striatal dopamine and in visual consciousness. Current Biology, 24, 265-266.
Padrao, G, Mallorqui, A., Cucurell, D., Marco-Pallares, J., & Rodriguez-Fornells, A. (2013).
Neurophysiological differences in reward processing in anhedonics. Cognitive Affective & Behavioral Neuroscience, 13, 102-115.
Pearson-Fuhrhop, K. M., Dunn, E. C., Mortero, S., Devan, J. W., Falcone, G. J., Lee, P. et al. (2014). Dopamine genetic risk score predicts depressive symptoms in healthy adults and adults with depression. PLoS ONE, 9, e93772.
Regenbrecht, H. & Schubert, T. (2002). Real and illusory interactions enhance presence in virtual environments. Presence: Teleoperators & Virtual Environments, 11, 425-434.
38 Rotter, J. B. (1966) Generalized expectancies for internal versus external control of reinforcement.
Psychological Monographs, 33, 300-303.
Salamone, J.D., Cousins, M.S., Bucher, S., (1994). Anhedonia or anergia? Effects of haloperidol and nucleus accumbens dopamine depletion on instrumental response selection in a T-maze cost/benefit procedure. Behavioural Brain Research 65, 221–229.
Santos-Ruiz, A., Garcia-Rios, M. C., Fernandez-Sanchez, J. C., Perez-Garcia, M., Munoz-Garcıa, M. A., & Peralta-Ramirez, M. I. (2012). Can decision-making skills affect respones to psychological stress in healthy woman? Psychoneuroendocrinology, 37, 1912-1921.
Sas, C., O'Hare, G. M. P. , Reilly, R. (2003). Presence and task performance: an approach in the light of cognitive style. Cognitive Tech Work, 6, 52-56.
Schmidt, L., d’Arc, B. F., Lafargue, G., Galanaud, D., Czernecki, V., Grabli, D., Schüpbach, M., Hartmann, A., Lévy, R., Dubois, B., and Pessiglione, M. (2008). Disconnecting force from money: effects of basal ganglia damage on incentive motivation. Brain 131, 1303–1310. Schubert, T., Friedmann, F., & Regenbrecht, H. (2001). The experience of presence: factor analytic
insights. Presence: Teleoperators & Virtual Environments, 10, 266-281.
Smoski, M. J., Felder, J., Bizzell, J ; Green, S. R. ; Ernst, M., Lynch, T. R., & Dichter, G. S. (2009). fMRI of alterations in reward selection, anticipation, and feedback in major depressive disorder. Journal of Affective Disorders, 118(1), 69-78.
Snaith RP, Hamilton M, Morley S, Humayan A, Hargreaves D, et al. (1995). A scale for the assessment of hedonic tone the Snaith-Hamilton Pleasure Scale. Br J Psychiatry, 16,: 99–103.
Sherdell, L., Gotlib, I. H., & Waugh, C. E. (2012). Anticipatory pleaure predicts motivation for reward in major depression. Journal of Abnormal Psychology, 121, 51-60.
Slater, M. (1999). Measuring presence: a response to the Witmer and Singer presence questionaire. Presence-Teleoperators And Virtual Enviroments, 8, 560-565.
Stotz, G., Woggon, B, & Angst, J. (1999). Psychostimulants in the therapy of treatment-resistant depression Review of the literature and findings from a retrospective study in 65 depressed patients. Dialogues in clinical neuroscience, 1, 165-74
Tye, K. M., Mirzabekov, J. J., Warden, M. R., Ferenczi, E. A., Tsai, H., Finkelstein, J., Kim, S., Adhikari, A., Thompson, K. R., Andalman, A. S., Gunaydin, L. A., Witten, I. B., & Deisseroth, K. (2013). Dopamine neurons modulate neural encoding and expression of depression-related behaviour. Nature, 24, 537-541.
Treadway, T. M., Bossaller, N. A., Shelton, R. C., & Zald, D. H. (2012a). Effort-based decision making in major depressive disorder: a translational model of motivational anhedonia. Journal of Abnormal Psychology, 121, 553-558.
Treadway, T.M., Buckholtz, J. W., Cowan, R. L., Woodward, R. D., Li, R., Sib Ansari, M., Baldwin, R. M., Schwartzman,A. N., Kessler,R. M., & Zald, D. H. (2012b). Dopaminergic mechanisms of
39 individual differences in human effort-based decision-making. The Journal of Neuroscience, 32, 6170-6176.
Treadway, M. T., Buckholtz, J. W., Schwartzman, A. N., Lambert, W. E., Zal, H. Z. (2009). Worth the ‘EEfRT’? the effort expenditure for rewards task as an objective measure of motivation and anhedonia. PLoS ONE, 4, e6598.
Tversky, A, & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211, 453-458.
Tversky, A. & Kahneman, D. (1992) Advances in prospect theory: Cumulative representation of uncertainty, 5, 297-323.
Vos, S. H., Gunter, T. C., Kolk, H. H. J., & Mulder, G. (2001). Working memory constraints on syntactic processing: An electrophysiological investigation. Psychophysiology, 38, 41–63.
Wallach, H. S., Safir, M. P., & Almog, I. (2009). Attachment and sense of presence in a virtual enviroment. Virtual Reality, 13, 205-217.
Wardle, M. C., Treadway, M. T., Mayo, L. M., Zald, D. H., & de Wit, H. (2011). Amping up effort: effects of d-amphetamine on human effort-based decision making. The Journal of Neuroscience, 31, 6597–16602.
Wendel-Vos W. G. C., Schuita, J. A., Sarisc. W. H. M., & Kroumhout, D. (2003). Reproducibility and relative validity of the Short Questionnaire to Assess Health-enhancing physical activity. Journal of Clinical Epidemiology, 56, 1163–1169
Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual enviroments: a presence questionnaire. Presence, 7, 225-240.
Yang, X., Huang, J., Zhu, C., Wang, Y., Cheung, E. F. C., Chan, R. C. K., & Xie, G. (2014). Motivational deficits in effort-based decision making in individuals with subsyndromal depression, first-episode and remitted depression patients. Psychiatry Research, 220, 874-882.
