Are two different systems comparable regarding the sense of presence? I know that using only questionnaires it is not indicated to do so, but considering that I will study presence through questionnaires, behaviour responses and physiological responses, does this give more validity?
I'd like some more context for your question. You can certainly try to measure presence in several different ways, in different VR systems, and compare and discuss these measurements. There are many limitations and potential problems but which of these are important for you depends on the details of your question.
What kind of differences are you thinking of, for your systems? What aspect of or effect of presence are you primarily interested in? There are many ways to turn your initial question, but my basic response would be: yes.
Presence will be an independent variable. I want to see if the sense of presence influence trust in Virtual Reality. My hypothesis is that higher is the sense of presence higher is the trust people are going to have in the system. My idea is to compare a full immersive system (HMD) with a less immersive one (desktop VR).
Ahh. In truth I can't really say that I approve of using presence as an independent variable. There are a lot of factors that affect the sense of presence, and many relevant aspect of the sense of presence. Assuming that you have presence at one level in one system and at another level in another system is definitely problematic. There will probably be a consistent difference in presence between your systems but the degree and nature of this difference can vary significantly
Sure, you can say that you have "designs for different levels of presence" as an independent variable but in my opinion this doesn't really give you anything over more objective descriptions of the differences between your systems, like "I'll investigate the effect of full immersion compared to a desktop version of the same virtual environment".
I would prefer using such objective factors as independent variables. Then it would be fine to relate to presence in a discussion of your results, possibly together with measures that you relate to presence. In this case presence would be a dependent variable though and you'll be looking for correlations, not causation. I think that's what you'll really be doing anyway, but with this approach you'll do it deliberately. You could base a lot of interesting discussion on such correlations, e.g., what are the common factors underlying presence and trust, or how might they interact with each other?
provided that you make all the other variables the same you can compare presence across platforms. but I agree with Daniel on this if there are too many variables you will not be able to effectively isolate the efface of presence across platforms.
A good way to measure the effect of presence is by evaluation of physical symptoms, like heart rate or skin conductance, for example. One can use the same VR simulation in distinct platforms (like HMD and desktop), with a focus on situations that would stimulate strong emotions (like fear, or stress) and register this data. The higher the heart rate or the conductance, means the subjects where better immersed.
To be clear, my answer above was not really meant to discourage you from investigating the issue you're suggesting. I would recommend that you rephrase it and change some aspects of your approach, but I think the essence of it is workable, interesting and valuable.
Feel free to contact me privately if you want to discuss this further.
Thank you all for the answers. I am thinking deeply on how to measure this relationship. By now, I am more interested in the correlation between presence and trust. The causality may be the next step. I am generally against the comparison of the sense of presence between two different systems, because the people will weight their answers depending on the system they are interacting with. But probably the physiologic measures could be helpful in this, being an objective measure. Like you said the variable are several and difficult to control. I am trying to limit the number of them (i.e. by using the same virtual environment, same setting, same control system and by having some pre-test measures like tendency of immersion) but Presence is influenced by several factors, and taking all of them in consideration is quite hard.
Many studies have shown that the level of immersion is related to the degree of presence reported by participants, if that helps? I haven't got the references to hand though I'm afraid
We don't know enough about the experiment, and what you'll be taking into consideration, to say anything very precise or helpful. Nevertheless, I'll venture to say that you should be sure to take into account other important aspects apart from the interface, such as the nature of the task and the context of use. You may well find that the relationship between sense of presence and trust varies for different tasks (for some it might be that a greater sense of presence leads to stronger privacy concerns, and thus to a loss of trust).
Also, consider using conditional indirect effect analysis. I have found it useful to study the relationships between subjective variables, and between these and objective ones. It's even a relatively simple step toward speaking of causation (even though not everyone agrees!).
Thank you for your answers. David: my plan is to study the relationship for various tasks and samples. Starting from a simple task with students, to a complex task with a big firm employees. My aim is to validate the relationship regardless of the context or the participant. I cannot tell you more about the experiment because it is still in development, I am trying to design the best experiment and these answers are very helpful to me.
George: You are right and Immersion is for sure one of the fundamental factor I am going to assess in my experiments.
It is obviously possible to compare the sense of presence of two different VR systems. But this can be done through a set of measurable criteria, encompassing factors thought to be important.
- Badges
- Science topic
- Similar topics
- Correction