User Preferences of Perspective and Dimensionality of Tourism Space Geovisualisation Supporting Orientation in a Non-Immersive Virtual Reality

Orientation in the tourism space, where the visitor is often new to the visitor, is still a big problem both in open landscapes and in tourist facilities, especially in large-scale centres of mass tourism. Searching for facilities in spatial orientation, in accordance with the principles of universal design should ensure its intuitive character (Lidwell 2010), and take into account the specificity of the building’s functions and the individual predispositions in the spatial orientation of visitors. Currently, virtual reality (VR), used both directly in tourist facilities and via mobile devices during the entire journey, is one of the more often used tools supporting spatial orientation. VR also has great potential in universal design (taking into consideration the needs of the elderly, with disabilities), i.e. supporting and guaranteeing information and communication accessibility (Act on ensuring accessibility … 2019) in the tourist information system (Zajadacz 2014, 2015). The level of immersion is an important factor in which displays showing, e.g. tourist facilities are classified. Visual displays have been classified as non-immersive (PC monitors), semi-immersive (simulators) and fully immersive (head-mounted displays or computer automated virtual environments) (Slocum et al. 2001, Halik 2018, Lokka, Çöltekin 2019, Çöltekin et al. 2020). The aim of the present research was to find answers to questions about the optimal form of geovisualisation with the use of non-immersive VR in the creation of plans facilitating orientation in the tourism space. The ‘optimal’ form is understood as ‘the best possible under the given conditions’ (Słownik PWN, 2023), regarding the dimensionality and perspective of presentation, as well as the individual characteristics of tourists’ perception.

Measuring user’s preferences is a very important aspect of studies in cartography (Roth et al. 2017, Herman et al. 2018). Knowing what features of building plans are important from the point of view of their users is a crucial factor, and this factor is used to create maps that intuitively support spatial orientation, referring to the ability of organisms to navigate (Chamizo, Rodrigo 2019). Nowadays, VR as a technology enables viewing geovisualisations in an interactive way, which may help support spatial orientation. This aspect is being more and more often used currently and is the focus of many studies (Bruder et al. 2012, Cirio et al. 2013, Carbonell-Carrera, Saorin 2018, Diersch, Wolbers 2019). The effectiveness of VR in relation to the markings in the real world in terms of orientation in space was analysed. Some researchers (including Pastel et al. 2021) found no significant differences in travelling the route between moving in real space and VR. Using VR has been shown to provide similar participant behaviour as that of real-world interaction. On the other hand, the research of the team of Nguyen-Vo et al. (2017) on the impact of different frames of reference of a person moving in VR on the efficiency of movement, including the ease of finding the direction of the transition, target points and the time to cover the distance (Fig. 1), showed differences in perception depending on the adopted frame of reference while moving in VR. The conclusions indicated that the simulated ‘stationary room’ significantly improved user performance in terms of both navigational search time and overall travel distance. It was also highlighted that adding a variant frame of reference to a VR application could be a cost-effective solution for more efficient locomotion in VR. However, according to Nguyen-Vo et al. (2017), visualisation of the virtual environment with the head-mounted display (HMD) differs from reality, and it is still unclear whether visual perception works the same in VR with different levels of immersion.

Fig. 1.

Another current research problem, apart from the reference framework, is the perspective of observation in VR. On a daily basis, we move in three-dimensional (3D) space, observing the world from the perspective of a pedestrian (Gehl 2017). The use of this perspective in the exploration of geovisualisation in VR was presented, among others, by Zagata et al. (2021). It is a natural way of getting to know the surrounding space by humans, but it stands in contrast to the cartographic way in which the arrangement of rooms in large-scale buildings is shown. The dominant way of presentation is a two-dimensional (2D) orthogonal projection from above, which reflects the contours of the rooms and their functions. Most often, such geovisualisation takes the form of a printed plan in a scale smaller than 1:1. Currently, the development of geovisualisation techniques enables the implementation of more interactive solutions indicating the layout of rooms in buildings, which the user can independently traverse in a virtual space. This gives the opportunity to test the developed plans of large-scale buildings in various observation perspectives, i.e. passerby perspective, bird-eye view perspective or top–down view perspective (Medyńska-Gulij 2021) and dimensions (2D/3D information presentation (Halik, Kent 2021). Strategies using 2D and a 2D/3D combination are more effective than those with only 3D (Carbonell-Carrera et al. 2020).

Numerous studies have highlighted the differences in spatial orientation due to individual characteristics, such as gender (Chamizo, Rodrigo 2019). It has been shown that in both rats and humans, males seem more likely to rely on geometrical information to reach a goal, while females are more likely to use landmarks (Ward et al. 1986, Williams et al. 1990, Silverman, Eals 1992, Sandstrom et al. 1998, Jones et al. 2003, Pacheco-Cobos et al. 2010, Vashro, Cashdan 2015, Chamizo, Rodrigo 2019). Ward et al. (1986) observed that when giving directions, women tend to use landmarks as points of reference, while men tend to use distance or cardinal directions (like North, South, East, and West). Rodríguez et al. (2013) suggested young female rats behave like males on a spatial task, while a sex difference appears only after puberty is consistent with the suggestion that sex differences in spatial cognition in humans appears due to the hormonal and cognitive changes associated with puberty.

Spatial thinking is according to Nielsen et al. (2011, p. 60) ‘the ability to visualise and solve problems spatially’. It depends, as Maier (1996) points out, on individual spatial orientation skills and is defined as the ability to physically or mentally orientate oneself in space. There are numerous classifications of spatial skills (Smith 1964, Linn, Petersen 1985, Sorby 1999, Marunic, Glažar 2014, Weckbacher, Okamoto 2014) and some of them, such as those by Bodner and Guay (1997) and Tartre (1990), consider spatial orientation as the main component of spatial skills. It is measured i.a. using the spatial orientation tests – the Santa Barbara Sense of Direction (SBSOD) is a self-report scale of environmental spatial ability. It contains 15 questions regarding, e.g. the cardinal directions, reading maps and showing directions to other people (Hegarty et al. 2002) or spatial rotation. – The other test is the mental rotation test (MRT) (Vandenbert, Kuse 1978), which contains in our case a set of five rotated sets of 3D figures and the user’s task is to find the same 3D figures in a given set.

Individual predispositions related to spatial orientation can be developed through education, using also VR. Through the use of behavioural modelling methods (Bailenson 2019), the creation of attitudes of VR participants can be influenced. This also applies to eliminating the identified differences in spatial orientation between men and women. According to Chamizo and Rodrigo (2019, p. 9) ‘no doubt, cognitive intervention programs at schools (for example, different types of games geometric for girls and spatial memory ones for boys) could help to reduce the sex differences observed in adulthood’. Both the indicated possibilities that lie in VR in the field of supporting spatial orientation, and invariably the many difficulties in finding locations on the part of tourism space users, have become an impulse to undertake research focused on the search for the optimal form of geovisualisation of tourism space in non-immersive VR.

The main purpose of the research was the search for parameters for tools supporting spatial orientation in large-scale objects by selecting the optimal (1) observation perspective (passerby perspective, bird-eye view perspective, top– down view perspective) in the non-immersive VR and (2) dimensionality presentation (2D/3D), considering such characteristics of visitors as gender and individual spatial orientation.

Until now, 2D plans are most often created due to the lower level of complexity involved in the process of recreating reality. However, an open question pertains to whether creating building plans in 3D could improve the level of spatial orientation of users since we live in 3D space, and the technology of spatial information processing allows for the quick creation of 3D studies (Biljecki et al. 2016, Halik 2018, Halik, Smaczyński 2018). What should be the optimal parameters of non-immersive VR geovisualisation, considering the perception of the addressees of this type of tools? The research was based on the following hypotheses:

H₁: Users exploring a 2D plan or a 3D plan of a high-volume building prefer a specific geovision perspective (passerby perspective, birdeye view perspective, top–down view perspective).

The detailed tasks included identifying which of the presented plans (Fig. 2) facilitates spatial orientation in the building to the greatest extent. The participants of the experiment assessed the plans twice, i.e. before and after confronting the plan with reality, the preferences regarding the plans were analysed regarding such independent variables as gender and the individual level of spatial orientation of the study participants, determined on the basis of SBSOD and MRT tests.

Fig. 2.

In the experiment performed, six plans of a large-area building were tested on a non-immersive PC monitor screen (Fig. 2). The developed plans were divided into two groups, which were to indicate user preferences in terms of (1) dimensionality: 2D (A, B, C) vs 3D (D, E, F) and (2) presentation perspective: passerby view (A, D), bird-eye view (B, E) and top–down view (C, F). In order to determine preferences in relation to individual features (gender, level of spatial orientation), the participants of the study completed the SBSOD and MRT tests at the beginning.

As part of the research, the respondents were asked twice about their preferences related to the presented plans in order to check whether the confrontation of the studied plans with reality would change their preferences. For the first time, participants’ preferences were asked directly after viewing all six plans on the monitor screen. Then, assuming after Bailenson (2019, p. 125) that ‘when people experience something directly they begin to see it in a different light’, the respondents were asked to walk for a short while (max. 5 min) along the floor of the building shown on the plans and after returning they had to re-select the optimal plan. During the second study of preferences, in order to remind the respondents of the appearance of individual plans, they were presented on the A4 sheet with screenshots of all six analysed plans.

Apparatus

The research was carried out in a specially dedicated room in a large-scale building, the plans of which were presented to users. The analysed six interactive plans were displayed on the screen of a 27” monitor (Fig. 3). Users could move freely on them: forward, backward, left and right using the directional arrows placed on the keyboard. They could also use the computer mouse to rotate the camera view relative to the height of the starting plane representing a given perspective: passerby perspective, bird-eye view perspective and top–down view perspective. The questionnaire with the six tested geovisualisations used in the research is available from the authors.

Fig. 3.

The results of the analyses concerning the opinions of the surveyed persons before and after the verification of plans with reality were discussed taking into account (1) perspective and (2) dimensionality in connection with independent features of research participants (i.e. gender, individual level of spatial orientation). On this basis, the data synthesis was carried out, indicating the plan which, in the opinion of the participants of the experiment, facilitated the spatial orientation in the visited building to the greatest extent possible.

The preferred perspective plans

With regard to the perspective of the perception of the plans, preferences were taken into account due to such characteristics of the participants of the experiment as gender and the level of individual spatial orientation as measured by the SBSOD and the MRT tests (Fig. 4).

Fig. 4.

Considering drafting the plans taking into consideration the gender variable before walking through the building, due to the preferred presentation perspective (Fig. 4), the most popular of them is the top–down view (see Fig. 2) chosen by 84% of the women and by 59% of the men. After a walk inside the building, the top–down view is still the most popular presentation perspective (66% women and 49% men), although its popularity has decreased in favour of the bird-eye view by women and the passerby view by men.

Linking the preferences of the respondents with the results of the SBSOD spatial orientation test (Fig. 4) showed that before walking around the building, the most frequently chosen perspective was the top–down view indicated by 76% of the people with a low indicator SBSOD and 72% of the people with a high SBSOD rate. After a walk around the building’s interior, the top–down view was still the most popular presentation perspective (62% SBSOD ≤ 4.667; 56% SBSOD > 4.667).

Data analysis for the results from the MRT spatial rotation test (Fig. 4) showed that the top–down view chosen by 82% of the people with a low MRT index is the most frequently chosen perspective before walking around the building and 63% of the people with high MRT. After a walk inside the building, the top–down view is still the most popular presentation perspective (65% MRT < 5; 52% MRT = 5).

Preferred dimension plans

Similarly, as in the case of perspective, with regard to the dimensionality of the perception of plans, preferences were taken into account due to such characteristics of the experiment participants as gender and the level of individual spatial orientation as measured by the SBSOD and MRT tests (Fig. 5).

Fig. 5.

When examining gender preferences (Fig. 5) in terms of presentation dimensionality, it was found that 2D plans (A + B + C answers, see Fig. 2) were preferred before walking through the building, which were chosen by 65% of the women and 54% of the men. After a walk in the building, women still preferred the 2D plans (66% women), but men changed their preferences to 3D plans (57%).

Taking into account the level of spatial orientation (Fig. 5) of users before walking around the building in terms of presentation dimensionality, it can be concluded that 2D plans were preferred, which were chosen by 56% of the people with a low SBSOD index and 65% of the people with a high SBSOD index. After a walk in the building, 2D plans are still preferred (54% SBSOD ≤ 4.667 and 61% SBSOD > 4.667).

Preferences regarding the level of users’ spatial rotation before walking around the building in terms of the dimensionality of the presentation showed that people with a spatial orientation level lower than 5 (the maximum possible result) prefer 2D plans (71%), while people who obtained the maximum result in the MRT test preferred 3D plans (54%). However, after walking inside the building, 2D plans were preferred in both study groups (62% MRT < 5; 51% MRT = 5).

By analysing the totality of the presented results (Fig. 6), it can be concluded that the plan that facilitates orientation in space the most is plan C (2D top–down view perspective). This conclusion is in line with the results presented by Carbonell-Carrera et al. (2020), who showed that strategies using 2D and a 2D/3D combination are more effective than those with only 3D.

Fig. 6.

Plan C was indicated by 50% of the experiment participants after viewing all six proposals (immediately before walking around the building) and by 41% of the respondents immediately (after walking through the analysed building). Considering the plans in terms of the preferred presentation perspective, the most popular of them is the top–down view (A + C answers), chosen by 74% of the respondents before the walk and 59% after the walk. When examining the preferences in terms of presentation dimensionality, 2D plans (A + B + C answers) were preferred, which were chosen by 60% of the people before the walk and 57% of the respondents after the walk.

The research showed that the preferences of the participants of the experiment did not change statistically significantly after walking through the analysed building, in the case of choosing an individual plan; p-value = 0.73 and grouping plans in terms of dimensionality, p-value = 0.66. However, there is a statistically significant difference between the plans grouped in terms of p-value = 0.0051 perspective. This means that the confrontation of plans with reality during the walk had an influence on the change of preferences of the respondents in terms of the preferred perspective, i.e. people who previously chose the top–down view perspective changed their preferences to bird-eye view or passerby view.

The analysis of the respondents’ preferences in terms of gender (Fig. 6) before walking around the building showed that both in the opinion of women (56%) and men (41%) the plan that best presents the arrangement of the rooms is plan C. The most popular plan was plan C (49% of indications of women and 28% of indications of men) although its popularity decreased, especially in men in favour of plan D. The differences were not statistically significant; therefore, the differences in spatial orientation presented in the literature were not confirmed due to gender (Ward et al. 1986, Williams et al. 1990, Silverman, Eals 1992, Sandstrom et al. 1998, Jones et al. 2003, Pacheco-Cobos et al. 2010, Rodríguez et al. 2013, Vashro, Cashdan 2015, Chamizo, Rodrigo 2019).

Generalisation of the conclusions in terms of individual spatial orientation measured by the SBSOD (Hegarty et al. 2002) and the MRT (Vandenbert, Kuse 1978) tests in the study group indicates the agreement of views (Fig. 6). The results of the SBSOD spatial orientation test before walking around the building showed that, in the opinion of both people with low and high rates (48% and 52% of the responses, respectively), the layout of the rooms was best presented in plan C. Also, after a walk in the building, plan C was still the most popular plan (40% SBSOD > 4.667; 41% SBSOD ≤ 4.667). Also, the analysis of the preferences of the respondents in terms of the result obtained from the MRT spatial rotation test before walking around the building showed that both people with a low MRT index (62%) and people with a high MRT index (34%) preferred plan C. Plan C was still the most popular walk in the building (47% MRT < 5; 32% MRT = 5).

The results of the research showed a significant polarisation of the preferences of plans in terms of optimal spatial orientation in terms of room arrangement. Undoubtedly, the plan that, according to the participants of the experiment, best presents the layout of the rooms is plan C (top–down view and 2D objects) as indicated in all the analysed categories (i.e. gender, SBSOD, MRT). Therefore, the adopted hypotheses were confirmed:

Users exploring a 2D plan or a 3D plan of a large-scale building prefer a specific geovision perspective (passerby perspective, bird-eye view perspective, top–down view perspective);

It should be emphasised that the presented research results refer to geovisualisation with a low level of immersion; hence, the question is still open as to whether visual perception works the same in VR with different levels of immersion. Moreover, the participants of the experiment represented one age group (20–30 years old) and were non-disabled people. The research needs to be continued in terms of the diversity of the given sample in terms of age, being mindful of all generations of adults. Their results will broaden inference and allow for the identification of intergenerational differences. On the other hand, taking into account the principles of designing universal tourism space, it is also important to recognise the perception and preferences of plans supporting spatial orientation, considering the opinions of people with disabilities.