The spatial configuration of local climate zones reveals effects on wayfinding in human walking

behavior over vehicle navigation because people find it easier to position themselves in urban environment [3]. The exclusion of wayfinding from spatial planning increases the likelihood of people being lost in complex urban settings, resulting in psychological discomfort. There- fore, improving wayfinding could address a question of ‘where am I?’, which contributes to a more comfortable movement experience by reconfiguring urban spaces. Wayfinding has been frequently investigated in environmental psychology using a spatial cognition perspective. Spatial cognition is broadly defined as the understanding of the rela- tionships between the self and their surrounding environments and objects [4]. Moving through space, the various city images are created in the human mind due to different spatial knowledge acquired from the process of spatial cognition [5]. These city images can be the cognitive representations of the external environment to guide human wayfinding in move- ment. The more spatial information people can cognize from the surroundings while navigat- ing, the easier it is for them to achieve wayfinding. Based on the links between spatial cognition and wayfinding, environmental psychologists pay close attention to the physical characteristics of urban spaces for facilitating easy wayfinding [4]. Kevin Lynch was one of the pioneers who incorporated spatial cognition into urban planning, based on people-space inter- actions [6]. He proposed a concept of ‘legibility’ in the book—the image of the city [7]—to sug-

gest the effects of urban spatial characteristics on wayfinding. Following Kevin Lynch, increasingly researchers combined the wayfinding related to spatial cognition and urban spa- tial planning with the development of advanced analytical tools. Space syntax is a technique for quantifying wayfinding by modelling urban spatial charac- teristics, based on a premise that urban spaces are not the background of society, but rather that spatial and social attributes are intrinsically linked [8]. It implies the validity of measuring wayfinding in human movement using physically spatial characteristics. Moreover, space syn- tax also highlighted the ‘spatial configuration’ concept that describes the relationships between two spaces by taking into account their relationships with others [9]. It means that space syn- tax, compared with Kevin Lynch’s works, further provides a network analysis perspective. The roles of spatial configurations in the human movement have long been recognized [10, 11]. For example [12, 13], found that street spatial structure shaped human moving behavior. [14] even directly identified spatial configuration as the determinant of wayfinding. Spatial configu- rations tend to be heterogeneous across locations, for example, shown as either dispersed or agglomerated patterns. Such heterogeneity may result in incoherent spatial characteristics as well as poor wayfinding ability in human movement. We thus claim that space syntax has the potential to extend the conventionally spatial cognition-based wayfinding in environmental

psychology to spatial characteristics-based wayfinding in the urban planning field. However, the space syntax technique describes urban spatial characteristics by simplifying physical spa- tial relationships into street networks, without taking into account urban data, such as land use and built-up structure [15]. In this way, wayfinding measurement based on space syntax tech- nique has limited implications for and relevance to the implementation of spatial planning strategies. One approach that may increase the relevance of space syntax-based wayfinding in spatial planning is to incorporate urban form considerations. This could be achieved by measuring the spatial configuration of urban spaces using landscape metrics. Landscape metrics are algorithms that use numeric values to quantify urban spatial characteristics, such as the dis- tribution of land-cover types relative to others [16]. Land-cover type is the most common category for landscape metrics quantification in spatial configuration analysis [17, 18]. How- ever, other than land-cover type, land surface structure—the spatial configuration of various buildings—is often ignored in spatial analysis of urban form. This limitation can be addressed by using the local climate zone (LCZ) classification system that provides ten PLOS ONE The spatial configuration for wayfinding PLOS ONE | https://doi.org/10.1371/journal.pone.0289780 September 8, 2023 2 / 18