1 Rollercoaster Love:
Testing Excitation Transfer in (Virtual) Reality Charlie Helmijr
5815916 Master’s Thesis
Graduate School of Communication Master’s Programme Communication Science
Dr. Jeroen Lemmens 28 February 2019
2
Abstract
A field experiment and a laboratory experiment was conducted among 154 young adults between the ages of 18 and 30, to replicate earlier findings of studies on the Excitation Transfer Theory, and to examine if virtual reality (VR) experiences can elicit the same misattribution of residual
arousal on interpersonal attraction that is seen in real-world experiences. It was examined if experiecing a rollercoaster, whether in real-life or in VR, lead to an increase in attraction to someone of the preferred sex and a decrease in attraction to someone of the non-preferred sex. Furthermore, it was examined what effect different VR experiences with 6 degrees of freedom
(VR 6 DoF) and 3 degrees of freedom (VR 3 DoF) had on this effect and if this effect was mediated by arousal and by presence. No effect was found of experiencing a rollercoaster on interpersonal attraction. No mediation effect was found for arousal and presence, though spatial
presence was significantly higher in the VR 6DoF experience. The real-life, VR 6DoF and VR 3DoF all had significant effect on arousal. The increase in arousal was statistically between VR 3DoF and VR 6DoF and the real-life rollercoaster. Interestingly, the increase in arousal was not statistically different in VR 6DoF and the real-life setting. This sameness in increase in arousal
indicates that with certain experiences VR can recreate a similar perceived arousal response as real-world experiences. This effect on arousal should be examined further in future research, and
3
Rollercoaster Love: Testing Excitation Transfer in (Virtual) Reality
The excitation transfer theory proposes that residual excitement from a stimulus that arouses or excites can influence the excitatory response to a stimulus or situation experienced thereafter (Zillmann, 1971). In studies by Dutton and Aron (1974) and Meston and Frohlich (2013) researchers concluded that residual excitement from physiological arousal lead to misattribution of emotional valence. Meston and Frohlich (2013) found that participants rated photographs of individuals of the opposite sex higher on attractiveness and desirability when they had completed a rollercoaster ride than those that were about to go in a rollercoaster. Similarly, Dutton and Aron (1974) asked male participants to walk over either a shaky
suspension bridge or a stable solid bridge. When the men were asked to write a short story about a drawing of woman, participants that walked over the shaky suspension bridge had stories with much more seual imagery. Though both the bridge experiment and the rollercoaster experiment are often cited and noted studies in social science, there are few examples of studies that have empirically tested and replicated these findings. The Excitation Transfer Theory being a pinnacle of Communication Science, this study aims to not only corroborate these previous findings that have been hard to replicate, but to apply the theory to a new, emerging and exciting media: virtual reality (VR).
Virtual reality enables a lot of new of possibilities in all types of fields: (social) media, health and fitness, productivity, and more. As is often the case with new technology and new media, it can be a quite divisive subject: while some worry about its consequences on our society, others praise it as the next step in media and go so far as to describe it as the ultimate “empathy machine” (Milk, 2015). VR research is not new and VR as a consumer product has
4 been pushed before and failed. After disappearing for 25 years, VR is yet again speedily rising in popularity. This year, VR sales are doubling, device owners are increasing and the VR market is expected to grow to ten times larger in five years. VR is currently already leveraged for different types of research regarding empathy, different types of biases, health treatment and gaming effects. VR is used to treat phobias (Suso-Ribera et al., 2018), autism (Kandalaft, Didehbanu, Krawczyk, Allen, & Chapman, 2013) and PTSD (Diemer, Lohkamp, Mühlberger, & Zwanzger, 2016).
VR can also offer new possibilities as a research tool. Situations can be recreated in a virtual environment that in the real world might be too expensive, unethical, or plain impossible.. But in order to maximize the benefits as not only a consumer product but also as a research tool, we must understand the processes through which VR “works”. Furthermore, we must understand the possible effect of VR compared to other media. Seeing as VR can create a virtual
environment, emulating the real word, it is essential to also compare if and how VR can recreate processes and effects we see in real-life settings and how these differ in VR.
Thus with the rise in popularity of VR devices and the prospect of VR technology becoming more accessible, while also becoming more advanced. It is crucial that we understand which factors affect VR experiences in what way. Seeing as VR is popular as a gaming device (Roettl & Terlutter, 2018), that users use to relax or have fun, and with more attention going to social VR (Rubo & Garner, 2018), it is worthwhile examining if and how VR can recreate arousal responses and if that response can mimic responses in the real world. In attempting answer to these questions, we are concurrently presented with a way to empirically test propositions of the Excitation Transfer Theory. This study will look at the role of VR, arousal
5 and presence on the misattribution of arousal, resulting in a change in subsequent response or emotions. Thus investigating both main goals of this study.
Excitation Transfer Theory
The excitation transfer theory proposes that residual excitement from a stimulus, or response to that stimulus, that arouses or excites, can influence the excitatory response to a stimulus or situation experienced thereafter (Zillmann, 1971). In a study on the effects of
physical exercise on the misattribution of residual arousal on sexual arousal, Cantor, Zillman and Bryant (1975) explain that arousal of the sympathetic nervous system does not acutely stop or disappear once the situation or condition inducing the arousal has stopped. Rather, this arousal slowly dissipates, this allows for ‘residual arousal’ to be able to spill over into the next moment or situation. It can be proposed that residual excitement can be transferred to a subsequent situation and the emotions in that situation.
Cantor, Bryant and Zillman (1974) proposed that certain aspects can increase the residual arousal that is transferred. First, that the more arousing a stimulus is, the more residual arousal there is to be transferred onto the subsequent response. Second, the switch to the subsequent stimulus should not be jarring or alert the participant of an attempt to transfer excitement from one response to the other. In practice this means that participants should not be fully aware of their heightened state of arousal when the subsequent situation or response is brought on. Similarly, White and Kight (1984) concluded that misattribution of the residual arousal onto the subsequent stimulus or response is more likely when participants are not self-conscious about the effect of the arousing stimulus and about their heightened state of arousal. This is aided by stimulus that produce high levels of arousal: this will interfere with the attention being paid to the subsequent response, so as to be unaware of the misattribution of residual arousal. Lastly,
6 timing is important: arousal decay sets in as soon as the first response ceases and there might be minimal residual arousal to be transferred in a matter of minutes.
Studies empirically testing the excitation transfer theory are few. Though in a study by Meston and Frohlich (2013), researchers investigated excitation transfer by looking at how participants that just went on a rollercoaster ride rated photographs of people of the opposite sex. The researchers found that participants that went on a rollercoaster ride exhibited a transfer of residual arousal: participants exiting the rollercoaster gave higher ratings for attractiveness and dating desirability than participants entering the ride. Residual (physiological) arousal resulted in misattribution of this unrelated arousal and caused an increase in interpersonal attraction.
Similarly, White, Fishbein and Rutsein (1981) found that physiological arousal from exercising, resulted in males liking an attractive female confederate more than the control group. They also found that the aroused subjects liked an unattractive female confederate less than the control group. This implies that residual arousal can cause an increase in interpersonal attraction towards a person the participants finds attractive, but can also cause a decrease in interpersonal attraction towards a person the participants find unattractive. These results indicate that the misattribution of arousal leads to liking more or finding an (according to the participant) attractive person more attractive: and that misattribution of arousal leads to also liking less, or finding an (according to the participant) unattractive person even attractive.
Although, the excitation transfer theory is a pinnacle of communication science, few studies have empirically tested the transfer of residual arousal on to subsequent emotions or responses. Even fewer studies have investigated how misattribution of arousal can negatively influence attraction. This study aims to replicate the results found in the aforementioned studies.
7 But more importantly, aims to also apply the excitation transfer theory to a new field in media (studies) and communication science: virtual reality.
Virtual reality
Virtual reality (VR) can be defined as a computer-generated 3D environment that’s capable of simulating real-world settings and situations (Wei, Qi, & Zhang, 2019). VR enables people to immerse themselves in a virtual world, where they can navigate and possibly interact with this virtual environment (Lau & Lee, 2015; Neumann et al, 2018). VR is an emerging field in media, communications, health and many other areas. VR offers new possibilities when it comes to entertainment purposes, different health-related therapies and educational purposes. VR has been shown to be able to simulate and substitute real-world situations. Suso-Ribera et al. (2018) found that in treating phobia of small animals, a VR and AR treatment could be just as effective as an in vivo treatment. In a literature review of 12 studies examining VR as a way to provide treatment for PTSD, researchers concluded that the studies found VR to be just as effective as in-vivo exposures (Botella, Serrano, Baños, & Garcia-Palacios, 2015). Suso-Ribera et al. (2018) also found that in-vivo treatments had the highest rates of success, but that VR treatments were just as effective in patients that reacted well to prior treatment. When looking at arousal, in a different research on the role of arousal in treating participants with phobia for small animals, Diemer, Lohkamp, Mühlberger and Zwanzger (2016) concluded that although a real-world setting elicited a larger increase in physiological arousal, VR simulations could still elicit a mild to moderate arousal response. If VR can be used to simulate and substitute in-vivo
situations, and VR can elicit the similar arousal responses as real-world situations, this indicates that VR could also elicit a comparable arousal response as a real-world situation. Which brings us to the major goal of this research. The excitation transfer process will be compared in a real
8 world setting, going on a real-life rollercoaster, to that of a virtual setting, going on a VR
rollercoaster.
A comparison will be made in how participants that went on a live rollercoaster, and participants that went on a VR rollercoaster rate interpersonal attraction. Based on the research and theories discussed we propose the following hypotheses:
H1A. Participants’ attraction to someone of their preferred sex will increase after experiencing a rollercoaster.
H1B. Participants’ attraction to someone of their non-preferred sex will decrease after experiencing a rollercoaster.
H2. Experiencing a live rollercoaster will lead to a stronger change in attraction than experiencing a virtual rollercoaster.
Immersiveness
Immersiveness describes the type of technology used to display media (and our stimuli) and how technologically immersive these technologies are (Mania & Chalmers, 2001). With this term we can differentiate between the levels of immersiveness that different hardware and technology offers. Technological aspects can dictate how immersive a VR experience is, from the audio delivery system, to how detailed the image is (Butti & Chitarro, 2018; Mania & Chalmers, 2001). Immersiveness is a broadly described term and there are multiple definitions for it (Georgiou & Kyza, 2017). Immersiveness is closely related to presence (Brautovic, John, & Potrebica, 2017; Wei, et al., 2019) and scientist either use them as correlates of each other or as substitutive terms. In this study a distinction is made between presence and technological
immersiveness. Technological immersiveness are features of different display types and different other hardware-related factors.
9 An example of technological immersiveness is how certain VR headsets and experiences offer more “degrees of freedom” than others, allowing more types of movement and thus being able to mimic movement more realistically and offers more interactivity in the virtual
environment (North & North, 2016). The degrees of freedom dictate the number of ways an object can move in 3D space. In a 3 Degrees of Freedom (3DoF) VR experience, your head and head orientation can be tracked. Three axes can be controlled: roll, yaw and pitch. In contrast, in a 6 Degrees of Freedom experience, orientation and position in a 3D space can be tracked. The headset understands not only where you are looking and how your head moves, but also knows where you are positioned in space. Allowing you to also move front and back, left and right and up and down. There is extremely little research on the effects of the levels of degrees of freedom of VR experiences. Although there is very little research on the matter, 6DoF offers more
freedom in movement in a virtual environment. This in turn enables your movement in the environment to better fit (and mimic) your actual movement (in the “real world”) which one would argue makes for a more immersive experience. Because there is little research on the topic, it is hard to make predictions as to how the two levels of degrees of freedom will affect our results. But it will be investigated what the influence is of the different types of VR experiences (3DoF and 6DoF) on the misattribution of arousal, presence and interpersonal attraction.
Spatial presence
Earlier on we made the distinction between technological immersiveness and presence. Presence can be explained as the feeling or even illusion of ‘being there’ (Coxon, Kelly, & Page, 2016; Slater & Sanchez-Vives, 2016). Being in a computer-generated environment but still having the feeling that you were spatially in that environment. It’s because of this spatial presence that VR technology has the ability to create complex narratives with complex viewer
10 interaction (Cummings & Bailenson, 2016). Users can be made to feel like they are present in a VR experience; users can be dropped into a scene and feel like they are part of a story.
Researchers have found that VR experiences on a head-mounted display enable a higher level of spatial presence compared to experiences on 2D screens and that this higher level of spatial presence resulted in an amplification in the effects of VR on different outcome measures, e.g. aggression (Persky & Blascovich, 2007), arousal and emotional state (Diemer, Alpers,
Peperkorn, Shiban, & Mühlberger, 2015; Riva et al., 2007). Therefore in this study it is argues that spatial presence enables VR to blur the lines between reality and imagination and results in increased arousal and thus an amplification of the effects on interpersonal attraction. Thus the following hypothesis is proposed:
H3. (Serial Mediation) The effect of experiencing a VR rollercoaster on arousal is mediated by presence and arousal..
Method Sample
The study consisted of two experiments: a lab experiment (n = 42) and a field experiment (n = 100). For the field experiment, on the 14th and 15th of June 2018, a total of 46 participants were recruited at the entrance of the Goliath rollercoaster in the Walibi Park and were asked to participate in the study in exchange for 10 Walibi Kix (1 euro gift cards).
For the lab experiment, in May and the first two weeks of July 2018, a total of 115 participants were recruited from the UvA Lab website (used by Communication Science and Psychology students of the UvA), posters in the UvA-buildings, social media sites (Facebook and Whatsapp) and snowball sampling. The only requirements to participate, for both experiment the lab study and the field study, was that participants were between 17 and 30.
11 Of the 115 participants that participated in the lab experiment, 6 participants were
excluded from the dataset: one participant was excluded due to the VR headset crashing and another five were excluded due to either not finishing or answering the survey incorrectly. The final sample of 109 participants for the lab experiment had an average age of 21 and all
participants were between the ages of 18 and 30 (M = 21.46, SD = 2,64).
One participant was excluded from the field experiment due to answering the survey incorrectly. The total number of participants for this study was 154 (N = 154), with an average age between 21 and 22 (M = 21.47, SD = 2.70). Of the participants, 103 (67%) were female, 49 (32%) male and 2 (1%) participants reported belonging to “Other” . The field experiment consisted of one condition (real-life rollercoaster) and had 45 participants. The lab experiment consisted of two conditions; one condition (VR 3DoF) consisted of 50 participants and the other (VR 6DoF) consisted of 59 participants. There was no significant difference between the groups in age (F(2, 151) = 1.23, p = .293) or in gender (F(2, 151) = 1.03, p = .359).
Design and procedure
The lab and field experiments were conducted as part of a larger study and was approved by the UvA Ethics Review Board.. This study on excitation transfer consisted of three
conditions: a real-life rollercoaster, a VR 6-degrees-of-freedom (VR 6DoF) condition and a VR 3-degrees-of-freedom (VR 3DoF) condition.
For the field experiment, participants were recruited by two researchers of the University of Amsterdam on the 14th and 15th of June 2019 and were asked if they wanted to participate in a research centered around the Goliath rollercoaster. Participants age ranged from 18 to 30 (M = 22.07, SD = 3.07); 28 (62%) of the 45 participants were female and 17 (38%) male. The
12 Participants signed the consent form before starting the pre-test. The pre-test consisted of
questions related to participants’ prior experience with and liking of rollercoasters, questions and statements measuring arousal, attraction and more. The pre-test also contained questions related to grammar and math. This was part of the cover story (testing of cognitive skills), but was also employed as a way to help dampen arousal in participants that may have been already in a higher state of arousal. After completing the pre-test, participants sat in the front seat of the rollercoaster and the ride commenced. After the ride finished, participants exited and completed the post-test. The post-test consisted once again of questions and statements measuring attraction, arousal, demographics and more. As part of a larger study, both the pre-test and the post-test contained measures that were not used in this study. Examples being fear of heights and sensation seeking tendencies. After finishing the post-test, participants were given their compensation.
For the lab experiment, participants signed a consent form and physiological measures (torso strap and wristband) were applied. Participants’ age ranged from 18 and 30 (M = 21.30, SD = 2.61); 75 (69%) of the 109 participants were female, 32 (29%) male and 2 (2%)
participants reported belonging to “Other”. Participants were randomly assigned to one of the two conditions and started the pre-test. Light background music and noise was played on repeat to emulate the ambient noise of the Walibi Park. One was.
The participants then took a seat on a revolving office chair. The chair was equipped with seatbelts and participants were strapped in, similar to the Goliath rollercoaster experience, the VR headset and over-ear headphones were put on. The ride was started and a strong fan was turned on and put 1 meter in front of the participant to recreate the feeling of wind felt in a rollercoaster. After the ride finished, the headset was removed and the tablet was handed to complete the post-test. The post-test was similar to the one from the field experiment but
13 contained measure specifically related to VR, an example being presence. After completing the post-test, participants were given their compensation.
Stimuli
Participants in the real-life condition rode the actual Goliath ride in the Walibi Park. The Goliath is 46 meters tall and is the tallest rollercoaster in the Benelux (Goliath Achtbaan Walibi Holland, n.d.). With a maximum speed of 106 km/h and length of 1.2 km, it is also the fastest and the longest in the Benelux. It is a steel rollercoaster with multiple drops, helices and bunny hops. The ride lasts 1 minute and 30 seconds.
Participants in the VR 6DoF and VR 3DoF conditions experienced a VR adaptation of the Goliath via a HTC VIVE headset. The HTC VIVE’s head-tracking capabilities enables participants to be able to move forward/backward, up/down, left/right (translation in three perpendicular axes) combined with rotation about three perpendicular axes, often termed pitch, yaw, and roll. Thus enabling both VR 6DoF and VR 3DoF experiences. The VR 6DoF
experience was a reproduction of the Goliath rollercoaster in a popular rollercoaster simulator, No Limits 2. A detailed reconstruction of the Goliath was adopted from the internet and modified with roads, scenery and buildings to recreate the environment of the actual Goliath ride. The VR 6DoF experience lasted for 1 minute and 10 seconds.
In the VR 3DoF condition, participants experienced a VR 3DoF version of the Goliath ride. This was an official video with real-world footage recorded by Walibi Holland with a 360-degree video camera with a first-person view of riding the Goliath in the front seat. Participants had 3 degrees of freedom of movement: roll, yaw and pitch and the experience lasted for 1 minute and 10 seconds.
14 Attraction. Attraction was measured using one modified item from a sexual attraction scale used by Meston and Frohlich (2003): a 5-star rating system in which participants could give scores in increments of half stars on the question “If you were single, how much would you like to date her” for female models and “...him” for male models. The models were presented on a page that resembled the interface of popular dating apps like Tinder or OkCupid, with a fictional name and age shown of the beneath the model’s picture. All participants viewed one picture of two female and two male models in the pretest and viewed a different picture of the same models in the post-test. The pictures in the pre-test and post-test were similar, with the models being photographed from different angles. In a pilot test for this study, two pictures of ten models (five female and five male) were pilot-tested to ensure the models were similar in attractiveness between models and within their pair of pictures. A composite score was made for the pre-test attraction scores for the female models (α =.88, M = 2.77, SD = 1.48), the pre-test scores of the male models (α = .70, M = 2.18, SD = 1.39), the post-test scores of the female models (α = 0.91, M = 2.72, SD = 1.53) and the post-test scores of the male models (α = 0.75, M = 2.20, SD = 1.45). Participants’ score for attraction towards their preferred sex was computed with the scores from participants that like females for the female models and the scored for participants that liked males for the male models. The opposite was done for participants’ score for attraction towards their non-preferred sex.
Arousal. Arousal was measured by using a modified scale from Meston and Frohlich (2003). Three items could be rated on a 7-point Likert Scale. “Which of the following best describes your breathing rate at this moment?” and “Which of the following best describes your heart rate at this moment?” could be rated from “Very slow” (1) to “Very fast” (7) and “Which of the following best describes the degree to which you are sweating at this moment?” ranged
15 from “Not at all” (1) to “A great deal” (7). The items were measured twice, directly before the pre-test rating of dating desirability and directly after the post-test rating of dating desirability.A principal axis factor analysis (PAF) with oblimin rotation indicated that the 3 items for the pre-test arousal measure formed a single uni-dimensional scale, only one component had an
eigenvalue above 1 (eigenvalue 1.90), explaining 63.48% of the variance. The 3-item scale can be considered as indicated by a Cronbach’s alpha of .68 (M = 3.42, SD = .96). The scale’s reliability score (α =.79) would increase (by .11) if the ‘sweating’ item would be omitted. A reason for this could be the way the question can be answered: “Not at all” to “A great deal” versus “Very slow” to “Very fast” in the other questions; this was not deemed an issue and the item was still included.
Similarly the post-test arousal scale formed a single uni-dimensional scale, explaining 66.10% of the variance. The scale proved to be reliable as indicated by a Cronbach’s alpha of .71 (M = 4.05, SD = 1.06). The scales reliability could once again be improved (by .13) by omitting the ‘sweating’ item, but the item was included for the same reasons previously mentioned.
Presence. Presence was assessed by using a modified version of the Self-Location Scale from the Spatial Presence Questionnaire (MEC-SPQ) (Vorderer et al., 2004). The MEC-SPQ proposes two factors that directly measure Spatial Presence: Self-location and Possible Action. The scale consists of 6 items related to feeling physically present in the VR environment, examples being “It was as though my true location had shifted into the environment in the
game.” and “I felt as though I was physically present in the environment of the game.” The items could be rated on a 7-point Likert Scale ranging from “Strongly disagree” (1) to “Strongly agree” (7). Presence was only measured for the participants in the VR conditions (n = 100); asking participants if they felt they were present in the real-world did not seem logical. A principal axis
16 factor analysis (PAF) with oblimin rotation indicated that the 6 items formed a single
uni-dimensional scale, explaining 75.80% of the variance.The 6-item scale proved to be reliable as indicated by a Cronbach’s alpha of .94 (M = 4.79, SD = 1.34) and would not improve when an item is deleted.
Results Attraction
To test Hypothesis 1a, that participants’ attraction to someone of their preferred sex increases after experiencing a rollercoaster, a paired-samples t-test was conducted to compare difference in attraction towards preferred sex before the intervention (rollercoaster ride Real-Life, VR 6DoF or VR 3DoF) and after the intervention. There was no significant difference in the scores before the intervention (M = 3.08, SD = 1.11) and after the intervention (M = 3.15, SD = 1.16), t(141) = 1.68, p = .100. Similarly to H1a, a paired-samples t-test was conducted to test Hypothesis 1b, proposing a decrease in attraction rating for non-preferred sex after experiencing a rollercoaster, compared to before experiencing a rollercoaster. There was no significant
difference in the scores for attraction before the intervention (M = 1.88, SD = 1.53) and after the intervention (M = 1.78, SD = 1.51), t(141) = -1.93, p = 0.056. These results show that
experiencing a rollercoaster, regardless of whether this was in VR 6DoF or VR 3DoF, did not cause a change in participants’ score for attraction. Thus, both Hypothesis 1a and 1b are rejected. To test whether there was a difference in their effect, a paired samples t-test was run for all three conditions. To check if participants’ attraction to someone of their preferred sex increases after experiencing a rollercoaster, per every seperate condition, a paired-samples t-test was conducted to compare difference in attraction towards preferred sex before the intervention and after the intervention for 1) VR 3DoF, 2) VR 6DoF and 3) Real-Life. There was no significant difference
17 in any of the scores for preferred sex before the intervention (M = 3.16, SD = 1.03; M = 3.06; SD = .99; M = 2.99, SD = 1.32) and after the intervention (M = 3.24, SD = 1.14; M = 3.12, SD = 1.03; M = 3.07, SD = 1.34), t(48) = 1.25, p = .218; t(49) = 1.16, p = .252; t(51) = .69, p = .494.
Hypothesis 2 proposes that experiencing a live rollercoaster will lead to a stronger change in attraction than the virtual rollercoasters. A one-way analysis of variance was carried out to assess the influence of the type of rollercoaster (IV) on increase in attraction for someone of the preferred sex (DV). Since Levene’s Test for Homogeneity of Variances was significant (F(2, 139) = 4.68, p = .011), the Welch’s test was employed. However, the test was not significant, (Welch’s F(2, 83.19) = .38, p = .962). A second one-way analysis of variance was carried out to assess the influence of type of rollercoaster (IV) on decrease in attraction for someone of the non-preferred sex (DV). The test was not significant (F(2, 139) = 1.75, p = .410). We can conclude that experiencing a live rollercoaster does not lead to a stronger increase in attraction and that the increase in attraction does not differ significantly between the groups of the three rollercoaster types. Thus Hypothesis 2 is rejected.
Arousal
Hypothesis 3 postulates that experiencing a live rollercoaster will lead to a stronger increase in arousal than the virtual rollercoasters. To test if arousal was significantly different after experiencing a rollercoaster compared to before experiencing a rollercoaster, a paired-samples t-test was conducted. The difference in scores for arousal before the intervention and after the intervention was compared. There was a significant difference in the scores before the intervention (M = 3.42, SD = .96) and after the intervention (M = 4.05, SD = 1.06), t(141) = 7.80, p < .001. So experiencing a rollercoaster lead to an increase in arousal. An analysis of variance was carried out to assess the influence of the type of rollercoaster (IV) on increase in arousal
18 (DV). Since Levene’s Test for Homogeneity of Variances, was significant (F (2, 139) = 9.73, p < .001), the Welch’s test was used. The test revealed a statistically significant effect (Welch’s F(2, 82.37) = 6.20, p = .003) indicating that not all conditions differed in their affect on arousal. The Games-Howell post hoc procedure revealed a statistically significant difference (Mdifference = -.47, p = .010) between subjects in the VR 3DoF video group (M = .31, SD = .62) and the VR 6DoF group (M = .77, SD = .92), with VR 6DoF eliciting a larger increase in arousal. Similarly, a statistically significant difference (Mdifference = -.54, p = .034) was found between subjects in the VR 3DoF group and those in the real-life rollercoaster group (M = .85, SD = 1.24). No
statistically significant difference was found for increase in arousal between the VR 6DoF group and the real-life condition (Mdifference = -.78, p = .939). This indicates that participants’ increase in arousal did not statistically differ between the real-life rollercoaster group and the VR 6DoF group, but that there was a statistically significant difference in increase in arousal between the VR 3DoF group and the VR 6DoF, and between the VR 3DoF group and the real-life group. The largest difference was between the VR 3DoF and the real life condition. Thus, Hypothesis 3 is partially supported.
Presence and Arousal as Mediators
From our previous tests for Hypothesis 1a and 1b, we know that experiencing a rollercoaster does not have an effect on change in attraction. We also previously found, in our tests for Hypothesis 2, that experiencing a rollercoaster had a statistically significant positive effect on arousal. To test if the effect of experiencing a rollercoaster on arousal is mediated by presence a mediation analysis was performed using Model 6 of PROCESS Macro for SPSS (Hayes, 2012). The mediation analysis with 5000 bootstrapped samples was used to test if there was an indirect effect of the two types of VR conditions (VR 3DoF versus 6DofF) on change in
19 attraction towards preferred sex, via presence and arousal. The type of VR rollercoaster had a significant positive effect (B = 1.11, t(98) = 4.54, p <.001) on presence, R = .41, R2 = .17, F(1, 98) = 20.64, p < .001. A significant positive effect (B = .34, t(97) = -.56, p = .049) was found between type of VR rollercoaster and arousal, R = .33, R2 = .11, F(2, 97) = .61, p = .004. Contrary to what was expected, presence had no significant effect on arousal (B = .11, t(97) = 1.73, p = .087). Neither type of VR rollercoaster (B = -.05, t(96) = -.66, p = .509), nor presence (B = .05, t(96) = 1.38, p = .171), nor arousal (B = -.05, t(96) = -.99, p = .322) had a significant effect on attraction towards preferred sex or attraction towards non-preferred sex. The results obtained from this model, coincide with the findings from our previous tests. Eventhough we find a significant effect of condition on presence, an increase in attraction towards preferred sex is not mediated via presence nor arousal. The results are shown in Figure 1.
Figure 1. Model ‘Presence and Arousal as Mediators
Conclusion and Discussion
This study had two main aims: replicate previous studies that empirically tested the misattribution of residual arousal on emotional valence and to examine whether an experience in VR can elicit the same effect of misattribution of residual excitement, resulting in an
20 amplification of subsequent responses or emotions, as is seen with arousing stimulus in the “real world”. This “spilling over” of residual excitement onto a subsequent situation or response, thus intensifying a later emotional state, is what is posited by the Excitation Transfer Theory (Cantor et al., 1975; Zillmann, 1971). Though often cited and used, few studies have empirically tested the assertions of the Excitation Transfer Theory. In order to achieve both of these goals, it was investigated whether excitement from riding different rollercoasters, both real-life and VR rollercoasters, could result in a misattribution of emotional valence. Specifically, if residual excitement from riding real-life and VR rollercoasters would result in an amplification of someone’s experience of interpersonal attraction, and whether this effect is mediated by arousal and spatial presence. Contrary to previous studies (Dutton & Aron, 1974; Meston & Frohlich, 2003), experiencing an arousing stimulus or experience (a rollercoaster) did not have an effect on interpersonal attraction. Subjects who experienced a rollercoaster did not show an increase in attraction towards someone of the sex they preffered or a decrease in attraction towards someone of the sex they did not prefer. These results were not in line with expectations; reasons for these unexpected results are discussed later on in this Conclusion and Discussion section.
Second, this study strived to explore the difference not only between real-life
rollercoaster and VR rollercoasters and how this affects the misattribution of emotional valence, but also how different types of VR experiences may influence this effect. Specifically VR experiences that offer 3 degrees of freedom and 6 degrees of freedom of movement. When looking at the misattribution of arousal, resulting in an effect on interpersonal attraction, contrary to what was expected, no difference was found between the real-life rollercoaster, VR 3DoF rollercoaster and VR 6 DoF rollercoaster experiences. No significant effect was found on interpersonal attraction for any of the rollercoaster experiences.
21 Although no effect was found for experiencing a rollercoaster on attraction, a positive effect was found for experiencing a rollercoaster on arousal. In concordance with our
expectations, experiencing a rollercoaster lead to an increase in arousal in participants; this agrees with findings from previous research on the Excitation Tranfer Theory and the misattribution of arousal on emotional valence (Dutton & Aron, 1974; Meston & Frohlich, 2003). Furthermore, results showed that the increase in self-reported arousal was the same for participants who experienced the real-life rollercoasters as participants who experienced the VR 6DoF rollercoaster. This supports the idea that VR can provide an experience that elicits the same arousal response as a real-life experience. The results also revealed that participants experiencing a rollercoaster in a VR 3DoF experience showed a weaker increase in arousal than participants that experienced a real-life rollercoaster and a VR 6DoF rollercoaster experience. This indicates that the degrees of freedom, the different ways of moving in a VR experience, is consequential to people’s experience in VR. These results have exciting implications for VR: a virtual environment can replicate reality and processes we see in reality, to as a significant and comparable degree as in real-world settings.
Lastly, we found that the level of degrees of freedom in a VR experience also has an effect on the amount of spatial presence participants felt in the VR experience. The freedom to move around in more ways in 3D space had a positive effect on the level of spatial presence. Thus participants feel more as if they were actually in the VR environment, as if they were physically there, in the VR experience where more ways of moving around was possible. Although, we found no effect of presence on arousal, nor did it directly or indirectly affect attraction. Spatial presence could still have an effect on other aspects of VR. Examples being flow/involvement and consumer satisfaction as was found in a study by Kim and Ko (2019).
22 Other studies have also showed spatial presence to have an effect on how effective VR can be used as a medium to affect people’s emotions and levels of anxiety and relaxation (Diemer, Alpers, Peperkorn, Shiban, & Mühlberger, 2015; Riva et al., 2007).
One reason for not finding no significant effect on interpersonal attraction could be that participants could still be very aware of the change in arousal when they have just finished the rollercoaster ride. Cantor, Bryant, Zillmann (1974) concluded that there was a "sweet spot" when it comes to the transfer of residual arousal: if the change to the subsequent response happens within too small of a timeframe fast, the switch to the subsequent response occurs too jarringly and participants become aware of the switch. If the switch to the subsequent response happens too late after the initial arousing stimulus, most of the arousal may have already decayed and there is little residual arousal to transfer.
Second, the use of a pre-test and post-test measure for attraction might have alerted participants to the aim of the study. Even if participants were not aware of the aim of the study, participants could remember seeing the models before the rollercoaster intervention and
participants might be actively adhering to the ratings they previously gave. One measure that could help with this problem, is the use of a Solomon 4 Group design, where there is one group that does not get a pre-test. During the experiment some participants brought up the fact that they just saw the same models and were being asked to once again score them for how attractive they were. This could lead to participants opting to choose the same score as in their previous
response.
Third, it is possible that some participants interpreted the items concerning attraction, measured by asking if participants would like to date the models depicted, more in a way of giving the models a rating for physical attractiveness. The star rating system is not only used on
23 dating sites but also on app stores or shopping websites. On these websites it is often used as a way of giving a rating of likability or approval of a product or service. Thus some participants could have interpreted this as a way of giving a score for the attractiveness of the model and not of how much they would like to date the model. There seems to be some evidence for this in the data: paticipants gave sometimes favorable scores to both the models of the sex they preferred and the models of the sex they did not prefer.
Future research should focus on utilizing physiological measures in investigating the Excitation Transfer Theory. It is useful to be able to compare the perceived, self-reported measurement of arousal with those obtained with physiological measurements. If there is a discrepancy in these two measures, it opens up research into the differences between perceived arousal and actual, physiological arousal. Is one better than the other, or are they equal, do they affect other factors differently? Special consideration must be taken into account regarding the collection of physiological data in field experiments. Current hardware and systems can be troublesome in setting up and are not always reliable, as was experienced in this study during the field experiment. With the rise of VR and VR research, a reliable, non-invasive way of
accurately collecting physiological measures is crucial.
Lastly, as there are different ways of affecting the processes and effects inherent to VR, future research should examine how other factors can affect a VR experience. In this research we looked at how the difference in degrees of freedom affected the VR experience and its effects. Research could be done into the use of haptic feedback, or VR chairs, or different ways of interacting with an environment (different controllers), the use of avatars, the use of audio or the inclusion of a social factor in VR experiences.
24 References
Cantor, J. R., Bryant, J., & Zillmann, D. (1974). Enhancement of humor appreciation by transferred excitation. Journal of Personality and Social Psychology, 30(6), 812.
Cantor, J. R., Zillmann, D., & Bryant, J. (1975). Enhancement of experienced sexual arousal in response to erotic stimuli through misattribution of unrelated residual excitation. Journal of personality and social psychology, 32(1), 69.
Coxon, M., Kelly, N., & Page, S. (2016). Individual differences in virtual reality: Are spatial presence and spatial ability linked?. Virtual Reality, 20, 203-212.
Cummings, J. J., & Bailenson, J. N. (2016). How immersive is enough? A meta-analysis of the effect of immersive technology on user presence. Media Psychology, 19(2), 272-309.
Diemer, J., Alpers, G. W., Peperkorn, H. M., Shiban, Y., & Mühlberger, A. (2015). The impact of perception and presence on emotional reactions: a review of research in virtual reality. Frontiers in psychology, 6, 26.
Diemer, J., Lohkamp, N., Mühlberger, A., & Zwanzger, P. (2016). Fear and physiological arousal during a virtual height challenge—effects in patients with acrophobia and healthy controls. Journal of anxiety disorders, 37, 30-39.
Dutton, D. G., & Aron, A. P. (1974). Some evidence for heightened sexual attraction under conditions of high anxiety. Journal of personality and social psychology, 30(4), 510.
Georgiou, Y., & Kyza, E. A. (2017). The development and validation of the ARI questionnaire: An instrument for measuring immersion in location-based augmented reality settings. International Journal of Human-Computer Studies, 98, 24-37.
25 Goliath Achtbaan Walibi Holland. (n.d.). Retrieved 28 July, 2018, from
https://www.walibi.nl/nl/rides/thrills/goliath
Kim, D., & Ko, Y. J. (2019). The impact of virtual reality (VR) technology on sport spectators' flow experience and satisfaction. Computers in Human Behavior, 93, 346-356. Lau, K. W., & Lee, P. Y. (2015). The use of virtual reality for creating unusual
environmental stimulation to motivate students to explore creative ideas. Interactive Learning Environments, 23(1), 3-18.
Mania, K., & Chalmers, A. (2001). The effects of levels of immersion on memory and presence in virtual environments: A reality centered approach. CyberPsychology &
Behavior, 4(2), 247-264.
Meston, C. M., & Frohlich, P. F. (2003). Love at first fright: Partner salience moderates roller-coaster-induced excitation transfer. Archives of Sexual Behavior, 32(6), 537-544. Milk, C. (2015). How virtual reality can create the ultimate empathy machine. Lecture,
TED.
Neumann, D. L., Moffitt, R. L., Thomas, P. R., Loveday, K., Watling, D. P., Lombard, C. L., ... & Tremeer, M. A. (2018). A systematic review of the application of interactive virtual reality to sport. Virtual Reality, 22(3), 183-198.
North, M. M., & North, S. M. (2016). A comparative study of sense of presence of traditional virtual reality and immersive environments. Australasian Journal of
Information Systems, 20.
Persky, S., & Blascovich, J. (2007). Immersive Virtual Environments Versus Traditional Platforms: Effects of Violent and Nonviolent Video Game Play. Media Psychology, 10(1), 135-156.
26 Riva, G., Mantovani, F., Capideville, C. S., Preziosa, A., Morganti, F., Villani, D., ... & Alcañiz, M. (2007). Affective interactions using virtual reality: the link between presence and emotions. CyberPsychology & Behavior, 10(1), 45-56.
Roettl, J., & Terlutter, R. (2018). The same video game in 2D, 3D or virtual reality–How does technology impact game evaluation and brand placements?. PloS one, 13(7),
e0200724.
Rubo, M., & Gamer, M. (2018, June). Virtual reality as a proxy for real-life social attention?. In Proceedings of the 2018 ACM Symposium on Eye Tracking Research &
Applications (p. 81). ACM.
Slater, M., & Sanchez-Vives, M. V. (2016). Enhancing our lives with immersive virtual reality. Frontiers in Robotics and AI, 3, 74.
Suso-Ribera, C., Fernández-Álvarez, J., García-Palacios, A., Hoffman, H. G., Bretón-López, J., Banos, R. M., ... & Botella, C. (2018). Virtual Reality, Augmented Reality, and In Vivo Exposure Therapy: A Preliminary Comparison of Treatment Efficacy in Small Animal Phobia. Cyberpsychology, Behavior, and Social Networking, 22(1), 31-38. Wei, W., Qi, R., & Zhang, L. (2019). Effects of virtual reality on theme park visitors' experience and behaviors: A presence perspective. Tourism Management, 71, 282-293. White, G. L., Fishbein, S., & Rutsein, J. (1981). Passionate love and the misattribution of arousal. Journal of Personality and Social Psychology, 41(1), 56.
White, G. L., & Kight, T. D. (1984). Misattribution of arousal and attraction: Effects of salience of explanations for arousal. Journal of Experimental Social Psychology, 20(1), 55-64.
27 Zillmann, D. (1971). Excitation transfer in communication-mediated aggressive behavior.
