Studying the Development of Navigation Using Virtual Environments

Abstract

Research on spatial navigation is essential to understanding how mobile species adapt to their environments. Such research increasingly uses virtual environments (VEs) because, although VE has drawbacks, it allows for standardization of procedures, precision in measuring behaviors, ease in introducing variation, and cross-investigator comparability. Developmental researchers have used a wide range of VE testing methods, including desktop computers, gaming consoles, virtual reality, and phone applications. We survey the paradigms to guide researchers' choices, organizing them by their characteristics using a framework proposed by Girard (2022) in which navigation is reactive or deliberative, and may be tied to sensory input or not. This organization highlights what representations each paradigm indicates. VE tools have enriched our picture of the development of navigation, but much research remains to be done, e.g., determining retest reliability, comparing performance on different paradigms, validating performance against real-world behavior and open sharing. Reliable and valid assessments available on open-science repositories are essential for work on the development of navigation, its neural bases, and its implications for other cognitive domains.

Figure 1.. Schematic of spatial development across the lifespan.

Darker colors in the gradients represent better navigational accuracy and greater integration of systems. Figure adapted from Nguyen & Tansan, 2022/Open science framework, https://doi.org/10.17605/OSF.IO/PCJR3 (CC BY 4.0).

Figure 2.. Mazes with distal landmarks.

A. The children’s 4/8 VM radial maze from Bohbot et al. (2012). Left shows the overhead view of the maze with the 8 arms and surrounding distal cues. Right shows the first-person viewpoint from the center of the maze. Figure adapted from Bohbot et al. (2012) (CC BY 3.0). B. Overhead (left) and first-person (right) views of the adapted Morris Water Maze from Roome et al. (2017). Distal cues were the colorful mountains around the boundary and the proximal landmark remained inside the arena. Figure adapted from Roome et al., 2022/Open science framework, https://doi.org/10.17605/OSF.IO/WKRBZ (CC BY 4.0). C. The virtual object location task from Rodriguez-Andres et al. (2018). Left shows the overhead view with the proximal landmarks and distal landmarks just beyond the blue boundary wall. Right image shows a child participant’s perspective and the VE setup. Figure adapted from Rodriguez-Andres et al. (2018) (CC BY 4.0).

Figure 3.. Natural environments.

A. Virtual Silcton paradigm. Right image shows one of the eight buildings and the city-structure of the VE with general buildings, trees, a road, etc. Figure adapted from Nazareth et al. (2018). B. VE town from Nys et al. (2015), (2018)). The overhead view shows one test route traced in black and a list of landmarks along the route. Figure adapted from Nys et al. (2015) (CC BY 4.0).