Viewing Maps Does Not Improve Spatial Learning From Navigation

Researchers from The University of Hong Kong have determined that viewing a top-down map before or during navigation in a virtual environment does not significantly improve spatial learning. Published in Cognitive Research: Principles and Implications, the study tested whether map viewing could facilitate the formation of a cognitive map through first-person navigation. The findings indicate that performance on tests of spatial knowledge was equivalent whether participants viewed maps or relied solely on navigation, even when maps were presented repeatedly during the exploration process.

The investigation addressed a fundamental question in spatial cognition: whether the allocentric information provided by a map can effectively integrate with the egocentric information gained from moving through an environment. While it is well established that people can acquire spatial knowledge by studying maps, it has remained unclear whether this map-based knowledge can make the process of learning via direct navigation easier. The study aimed to resolve this ambiguity by experimentally comparing spatial knowledge acquired with and without map viewing in controlled virtual settings.

The research comprised three distinct experiments designed to test different types of environments and varying degrees of map detail. Experiment 1 utilized a city environment characterized by rectangular intersections and a grid-like structure. Participants navigated these virtual cities to find a sequence of targets. In the map viewing condition, participants viewed a structural map of the environment before performing exploration trials. The structural maps showed the layout of roads and main buildings but did not include the locations of the target kiosks, preventing direct learning of target positions from the map itself.

Experiment 2 shifted to a village-like environment that lacked a rectangular grid, featuring a main path and irregular side paths. This design tested whether the regularity of the city grid in the first experiment had masked potential benefits of map viewing. The maps in this experiment were also structural, showing only the paths, and were presented before navigation. The researchers dissociated the axes of the map from the intrinsic axes of the environment to test whether the map influenced the internal orientation of the cognitive map formed by the participants.

Experiment 3 introduced further modifications to maximize the potential utility of map viewing. This experiment used a village environment with more distinctive landmarks and presented a detailed map that included all environmental features, including the target buildings, though the targets remained unlabeled. Unlike the previous experiments where maps were shown only before navigation, the detailed map in Experiment 3 was presented repeatedly throughout the exploration, specifically after each exploration trial. This experiment also utilized an immersive virtual reality system with a head-mounted display to provide a more realistic navigation experience.

Across all three experiments, participants performed three tasks to assess their spatial knowledge: scene and orientation-dependent pointing (SOP), judgments of relative direction (JRD), and map reconstruction. The SOP task is thought to depend on egocentric representations, where participants judge the direction of unseen targets from their current position. The JRD task requires participants to imagine themselves at a specific location facing another location and point to a third target, a measure associated with allocentric or survey knowledge. Map reconstruction involved drawing the locations of targets on paper, providing another measure of allocentric spatial knowledge.

The results consistently showed no evidence that map viewing improved overall accuracy of spatial representations learned from navigation. In Experiment 1, an analysis of variance found no main effect of map preview on JRD errors, nor an interaction between map preview and learning blocks. Similarly, there was no effect of map preview on the accuracy of map sketches. The same pattern of null results was observed in Experiment 2, where map preview did not improve JRD or SOP performance, nor did it affect the accuracy of map reconstructions. In Experiment 3, despite the increased detail and repeated exposure to the maps, there was still no significant difference in JRD or SOP performance between the map and no-map conditions.

Bayesian analyses of the combined data from Experiments 1 and 2 indicated that any overall benefits from map viewing were likely to be small. The 95% credible intervals for effect sizes suggested that the maximum potential benefit for the JRD task corresponded to a 2.5-degree reduction in angular errors, which is a relatively minor improvement. For the SOP task, the maximum potential benefit was a 6.0-degree reduction in angular errors. These statistical estimates support the conclusion that map viewing did not provide a substantial advantage in spatial learning under the tested conditions.

One notable finding from Experiment 2 was that while map viewing did not improve overall accuracy, it did influence the orientation of the spatial representation. Participants who viewed the map showed alignment effects in their JRD performance that were consistent with the axes of the map, whereas those who did not view the map showed alignment effects consistent with the intrinsic axes of the environment. This suggests that map viewing can affect how spatial information is oriented in memory, even if it does not enhance the overall accuracy of that information. However, this alignment effect was not replicated in Experiment 3, where no interaction between map viewing and alignment was observed.

The researchers speculate that the lack of benefit from map viewing may stem from the difficulty in translating between egocentric and allocentric representations. Integrating information from a map, which is in an allocentric frame, with information from navigation, which is initially egocentric, requires a cognitive translation process that may be resource-intensive. The study suggests that this translation cost may prevent participants from effectively using map information to facilitate learning from direct experience. This finding contrasts with some previous studies that reported benefits from map viewing, but those studies often used fully labeled maps that allowed participants to learn the layout directly from the map rather than integrating it with navigation experience.

The study has several limitations that should be considered. The participants were a relatively homogeneous group of college students, which may limit the generalizability of the findings to other populations. The researchers acknowledge that individual differences in visual-spatial ability might play a role in how people utilize map information, but the sample sizes were not sufficient to test for such individual differences. Further research with larger and more diverse samples might reveal that map viewing is more beneficial for certain subsets of people.

Despite these limitations, the study provides valuable insights into the processes of spatial learning and the integration of different sources of spatial information. The findings suggest that while maps are useful tools for navigation, they may not necessarily enhance the cognitive map formed through direct experience. The study highlights the importance of understanding the distinct roles of egocentric and allocentric representations in spatial cognition and the challenges involved in integrating them.

In conclusion, the research demonstrates that viewing a top-down map does not facilitate spatial learning from navigation in virtual environments. The consistent null results across three experiments with varying designs and map details suggest that the integration of map and navigation information is not straightforward. The study contributes to the broader understanding of spatial cognition by highlighting the limitations of using maps to enhance learning from direct experience and the potential cognitive costs associated with integrating different spatial reference frames.

Subject of Research: Does viewing a top-down map facilitate spatial learning from navigation in a virtual environment?

Article Title: Does viewing a top-down map facilitate spatial learning from navigation in a virtual environment?

Article References: Ding, J., Chan, H. M., & Saunders, J. A. (2026). Does viewing a top-down map facilitate spatial learning from navigation in a virtual environment?. Cognitive Research: Principles and Implications, 11(1), Article 43. https://doi.org/10.1186/s41235-026-00732-y

Image Credits: AI Generated

DOI: 10.1186/s41235-026-00732-y

Keywords: Does, viewing, top-down, facilitate, spatial, learning, navigation, virtual, environment, scientific research