A gamification based- approach on indoor wayfinding research
The 'indoor_Navigation_Simulator' dataset contains all information regarding the research executed by Dries De Leeuw, Laure De Cock and Philippe De Maeyer. Abstract: Indoor environments can be very complex. Due to the challenges in these environments in combination with the absence of mobile wayfinding aids, a great need exists for innovative research on indoor wayfinding. In this explorative study, a game is developed in Unity to investigate whether the concept of gamification could be used in studies on indoor wayfinding so as to provide useful information regarding the link between wayfinding performance, personal characteristics and building layout. Results show a significant difference between gamers and non-gamers as the complexity of the player movement has an important impact on the navigation velocity in the game. However, further analysis reveals that the architectural layout also has an impact on the navigation velocity and that wrong turns in the game are influenced by the landmarks at the decision points: navigating at deeper decision points in convex spaces is slower and landmarks of the categories ‘pictograms’ and ‘infrastructural’ were more effective in this particular building. Therefore, this explorative study, which provides an approach for the use of gamification in indoor wayfinding research, has shown that serious games could be successfully used as a medium for data acquisition related to indoor wayfinding in a virtual environment. Keywords: gamification; indoor wayfinding; landmarks; space syntax; Unity As for our hypothesis and statistical analysis we would like to refer to our article. Following information is important to interpret our dataset: - Player_ID = ID which is given to each participant after registration to our game. - Level = Our game contains three different levels. - DP = each level contains a number of decision points. - Time = time needed to navigate past a specific decision point. - Distance = distance covered to navigate past a specific decision point. - MVD = mean visual depth values related to space syntax measurements. - DP_Group = categorization based on topology of decision point (1 = Single-turn; 2 = Multiple-turn; 3 = Startend). - DP_Group1 = categorization based on required navigation action (1 = Straightforward; 2 = Turn left; 3 = Turn right). - LM_Group = landmark categorization (0 = No Landmark; 1 = Infrastructural; 2 = Pictograms; 3 = Decorations; 4 = Objects; 5 = Furniture). - Player_Group = categorization of the player according to personal information. - Unique_ID = ID so as to distinguish each decision point separately. - Wrong_turn = indicates if the participant made a wrong turn at a specific decision point (0 = no; 1 = yes). Important to note is that this data is not normally distributed and thus non-parametric test have to be used (for example Mann-Whitney U-test). In addition, the 'Learning_Effect' file represents data on the participants that played the game two times.