Stereoscopic Rendering via Goggles Elicits Higher Functional Connectivity During Virtual Reality Gaming

Abstract

Virtual reality (VR) simulates real-world scenarios by creating a sense of presence in its users. Such immersive scenarios lead to behavior that is more similar to that displayed in real world settings, which may facilitate the transfer of knowledge and skills acquired in VR to similar real world situations. VR has already been used in education, psychotherapy, rehabilitation and it comes as an appealing choice for training intervention purposes. The aim of the present study was to investigate to what extent VR technology for games presented via goggles can be used in a magnetic resonance imaging scanner (MRI), addressing the question of whether brain connectivity differs between VR stimulation via goggles and a presentation from a screen via mirror projection. Moreover, we wanted to investigate whether stereoscopic goggle stimulation, where both eyes receive different visual input, would elicit stronger brain connectivity than a stimulation in which both eyes receive the same visual input (monoscopic). To our knowledge, there is no previous research using games and functional connectivity (FC) in MRI to address this question. Multiple analyses approaches were taken so that different aspects of brain connectivity could be covered: fractional low-frequency fluctuation, independent component analysis (ICA), seed-based FC (SeedFC) and graph analysis. In goggle presentation (mono and stereoscopic) as contrasted to screen, we found differences in brain activation in left cerebellum and postcentral gyrus as well as differences in connectivity in the visual cortex and frontal inferior cortex [when focusing on the visual and default mode network (DMN)]. When considering connectivity in specific areas of interest, we found higher connectivity between bilateral superior frontal cortex and the temporal lobe, as well as bilateral inferior parietal cortex with right calcarine and right lingual cortex. Furthermore, we found superior frontal cortex and insula/putamen to be more strongly connected in goggle stereoscopic vs. goggle monoscopic, in line with our hypothesis. We assume that the condition that elicits higher brain connectivity values should be most suited for long-term brain training interventions given that, extended training under these conditions could permanently improve brain connectivity on a functional as well as on a structural level.

Keywords: ICA; fMRI; fractional amplitude of low-frequency fluctuations; graph analysis; resting-state networks; seed-based functional connectivity; stereoscopic and monoscopic goggles; virtual reality.

Figure 1

A screenshot of the player’s view when flying in the virtual landscape. The colorful halos indicate the positions of the next flowers to be “pollinated.”

Figure 2

1—Difference of the mean fALFF maps, 1A goggles > screen and 1B goggles < screen. 2—Group differences in fALFF. Higher fALFF (in red) in left cerebellum and left postcentral gyrus in the goggles (monoscopic + stereoscopic) condition compared with the screen condition. In the reverse contrast (in blue) there was higher fALFF in right superior orbital frontal cortex.

Figure 3

Network spatial maps and group differences: default mode network (DMN) and visual. In the primary visual network, there was an increase in connectivity (in red) in the left calcarine and in the higher visual network in the left lingual in the goggles (monoscopic + stereoscopic) condition as compared to the screen condition. A decreased in connectivity (in blue) was seen in the left middle occipital in the primary visual network and in the bilateral cuneus in the higher visual network. In the DMN, we found an increase (in yellow) in the connectivity in the inferior frontal and bilateral lingual for goggles (monoscopic + stereoscopic) as compared to screen condition.

Figure 4

(A) Mean seed-based functional connectivity (SeedFC) maps per condition, seed located in the bilateral frontal superior cortex. (B) Left—Group differences in SeedFC in goggles vs. screen condition. There was stronger connectivity between a seed in the bilateral superior frontal cortex and the left superior temporal lobe for the magnetic resonance imaging (MRI) goggle contrast goggles (monoscopic + stereoscopic) as compared to the screen. (B) Right—Group differences in stereoscopic vs. monoscopic condition. The stereoscopic view elicited stronger connectivity between the bilateral frontal cortex and left insula and putamen.

Figure 5

(A) Mean seed-based functional connectivity (SeedFC) maps per condition, seed located in the bilateral inferior parietal cortex. (B) Group differences in SeedFC in goggles vs. screen condition. There was stronger connectivity between a seed in the bilateral parietal inferior cortex to right calcarine cortex and to right lingual cortex for the MRI goggle contrast goggles (monoscopic + stereoscopic) as compared to the screen.

Figure 6

Graph analysis. Betweenneess, transitivity, global efficiency, local efficiency and characteristic path length were calculated. *Means group differences before multiple comparison correction (p < 0.05 uncorrected). Group differences were not significant after multiple comparison correction.