**4. Discussion**

The current investigation provides detailed empirical insight into pedestrian visual engagemen<sup>t</sup> with different areas of urban street edges through the use of mobile eye-tracking. The study findings demonstrate, as predicted, that people visually engage with street edge ground floors more than upper floors, that visual engagemen<sup>t</sup> is distributed more towards the street edge on the walked side of non-pedestrianised streets than the opposite side, and that visual engagemen<sup>t</sup> with street edges of pedestrianised streets is balanced across both sides. However, as also anticipated, di ffering everyday activities and streets walked significantly influenced the amount of visual engagemen<sup>t</sup> upon all these street edge areas, except the street edge on the opposite side of non-pedestrianised streets, which was not predicted.

The study insights advance understanding of street edge visual engagemen<sup>t</sup> in a manner that has previously been challenging to attain. This subsequently provides opportunity to evidence existing understandings and assess how street edge design intervention can align to a greater extent with the way in which street edges are engaged with by pedestrians. This is outlined in the following discussion.

### *4.1. The Focus of Visual Engagement upon Street Edge Ground Floors*

The current study shows that pedestrians visually engaged more with street edge ground floors than upper floors along both non-pedestrianised and pedestrianised streets. This insight supports previous work describing the experiential significance of ground floors [19,20], especially the concepts of Gehl [1] and his *city at eye-level* and Glaser et al.'s [2] *street plinths*. It is also notable that the current study findings correspond well with Glaser et al.'s claim that ground floors determine 90% of peoples' experience of buildings that flank streets. The current study highlighted that 90%–92% of people's visual engagemen<sup>t</sup> takes place with the ground floor of street edges. However, ground floor engagemen<sup>t</sup> was shown to be less, at 34%–35%, when taking into account visual engagemen<sup>t</sup> with the entirety of the whole street.

Highlighting the focus of visual engagemen<sup>t</sup> upon ground floors supports the understanding that street edges should not be understood as singular entities with pedestrian experiential engagemen<sup>t</sup> distributed equally across their entirety. Instead, they need to be approached across multiple scales [2,36,37]. Within this, ground floors require consideration as a distinct scale embedded within the overall built morphology of street edges [2,34,38]. Significantly, this mind-set has implications for the way in which design intervention is approached, especially when seeking to create street edges that are more experientially engaging for pedestrians. Delineating ground floors as an experientially salient street edge scale provides better opportunity and scope for decision-making attention to be focused on their specific requirements [38,39]. Ground floors are therefore not solely regarded as the point at which buildings make contact with the ground and the dominant focus of interest is not just the aesthetic qualities and overall form of entire buildings [1,2,53], the totality of which pedestrians rarely visually engage with. Instead, ground floors become a point of decision-making focus in their own right based upon the knowledge that people overwhelmingly engage with them.

### *4.2. Visual Engagement with Street Edges on Di*ff*erent Sides of a Street*

The study findings highlight that pedestrians walking along non-pedestrianised streets visually engaged with the street edge on the walked side significantly more than the opposite street edge. This insight evidences the notion that street edges of the same street should be considered as experientially separate, even though both contribute to the spatial totality of a street [1,34,37]. It also brings into question the understanding that streets are experientially a *place between the edges* [37]. Instead, non-pedestrianised streets required greater consideration as a place between two experientially distinct edges, with pedestrian engagemen<sup>t</sup> focused towards the walked side street edge. Such a mind-set has subsequent implications for street edge design decision-making. Currently, streets and their edges are often approached in their totality. This is especially important when seeking to influence factors such as pedestrian walkability and the overall liveability of urban environments [35,42,43]. Even though, within certain situations, it is appropriate to think of streets in this way, intervention within street edges should also be considered on a side-by-side basis. This is along with the need for design decision-makers to anticipate that pedestrians on di fferent sides of a street will not visually engage with street edge interventions to an equal extent.

The study findings show that there was no significant di fference between the amount of visual engagemen<sup>t</sup> across the street edges on di fferent sides of pedestrianised streets. The significance of this is examined in the section that follows when reflecting upon the impact of pedestrianisation upon street edge visual engagement.

### *4.3. The Influence of Pedestrianisation upon Street Edge Visual Engagement*

There was no di fference highlighted between the dominant amount of visual engagemen<sup>t</sup> with street edge ground floors of non-pedestrianised and pedestrianised streets. This further evidences the need to consider street edge ground floors as experientially significant, regardless of street type [1,2,34]. As examined in the earlier ground floor discussion, such insight highlights the need to focus attention upon ground floors as experientially distinct from the overall built morphology within which they are embedded [2,38].

The current study findings show that visual engagemen<sup>t</sup> was more balanced across both street edges of pedestrianised streets than the street edges of non-pedestrianised streets. Previously, there has been limited systematic exploration of the way in which pedestrianisation influences peoples' engagemen<sup>t</sup> with the surrounding street environment, particularly its edges. The current investigation addresses this limitation, with the insights obtained o ffering subsequent opportunity to inform design decision-making. As shown, pedestrianised streets a fford more open and less-restricted distribution of visual engagement. As a result, they provide a spatial setting that encourages visual engagemen<sup>t</sup> with a greater proportion of opportunities within the surrounding street edges. This is significant when considering the fact that pedestrians direct a predominant amount of their visual engagemen<sup>t</sup> towards street edges as they build an understanding of what the surrounding setting o ffers them [1,2,14]. When seeking to establish urban streets that are more engaging, it is therefore beneficial to consider the way in which pedestrianisation significantly influences the extent to which people visually engage with the totality of what the surrounding street edges o ffer them.

### *4.4. Everyday Activities and Di*ff*ering Streets Walked Influence Street Edge Visual Engagement*

Optional activities encouraged a greater amount of visual engagement, in comparison to necessary activities, with the street edge ground floors and street edges on di fferent sides of both non-pedestrianised and pedestrianised streets. This insight adds further detail, from a direct pedestrian perspective, to the observations of Gehl [1] and corresponds with previous mobile eye-tracking research [14]. It also provides new insight into the way that pedestrians can be considered as active perceivers, through the way in which their everyday activities significantly influence and mediate their engagemen<sup>t</sup> with the surrounding environment [26]. The current study findings thus further the argumen<sup>t</sup> that people engage with urban environments in an enactive manner, with combined social influences and spatial factors influencing the urban experience [40,41].

The di ffering streets walked influenced the amount of visual engagemen<sup>t</sup> upon the street edge ground floors of non-pedestrianised and pedestrianised streets. They also influenced the amount of visual engagemen<sup>t</sup> with both the left and right sided street edges of pedestrianised streets and the street edges on the walked side of non-pedestrianised streets. Such insight advances systematic understanding of the extent to which varying streets influence visual engagemen<sup>t</sup> with di fferent street edge areas. Future research, again using mobile eye-tracking, could build upon this foundation through a focused analysis of what specific physical and material attributes of the environment influence street edge visual engagement. This would provide empirical insight into what specific characteristics of the environment visually engage pedestrians, subsequently providing opportunity to inform design decision-making.

The current investigation captured no di fference in the amount of visual engagemen<sup>t</sup> with the street edges on the opposite side of non-pedestrianised streets. This insight was not anticipated. It does, however, provide opportunity to consider how the spatial composition of these streets, along with objects within the street, restricted visual engagemen<sup>t</sup> to the walked side to such an extent that there

was no opportunity for significant variation in visual engagemen<sup>t</sup> with the opposite street edge. This provides further opportunity to evidence the potential experiential benefits of pedestrianisation, which afford more open visual engagemen<sup>t</sup> with the surrounding street edges, as shown during the current investigation. However, further systematic investigation is needed to examine such ideas in detail.

### *4.5. Study Limitations and Future Research*

The current study provides a new empirical understanding of the way in which pedestrians visually engage with urban street edges. Even though this is the case, it is beneficial to acknowledge that there are a number of factors stemming from the current research that require further consideration. There is also opportunity to highlight future research opportunities.

The current investigation did not look at assessing how di fferences in the characteristics of the street edges examined, or the composition of the streets walked, might have influenced pedestrian visual engagement. Instead, the focus was on capturing general principles and clear patterns in the way that people visually engage with street edges. Beyond the fact that the study streets could be categorised as non-pedestrianised or pedestrianised, they clearly have distinct features and attributes that may have influenced where people looked in the street (see Figure 3). This highlights an opportunity for future research, which could more systematically examine the e ffect of specific street and street edge characteristics on street edge visual engagement.

We believe that findings from the current study are transferable to broader situations, with the findings obtained validating and building upon previous insight attained within di ffering urban contexts. It needs to be acknowledged, however, that data collection during the current study took place within a specific urban environment categorised by a certain type of European urbanism (see Figure 3). Further investigation would help to determine if the insights obtained during the current study can be seen when assessing visual engagemen<sup>t</sup> with street edges in other urban settings with contrasting types of urbanism. The current investigation also focused on the examination of pedestrian engagemen<sup>t</sup> with commercial city centre urban street edges. It would be interesting so see how visual engagemen<sup>t</sup> is distributed upon di fferent street edge typologies, particularly residential street edges.

Mobile eye-tracking outdoors provides detailed insight into the way in which people visually engage with their surroundings at the same time as being immersed within the reality of real-world urban settings. However, when people are situated within such environments, it can be di fficult to assess if their attention, along with second-by-second perceptual processing, is actually being directed towards what they are looking at within the often diverse and multi-sensory environment around them. This is due to their being heightened potential for them to be cognitively processing aspects of their wider surroundings or something they have previously engaged with (21). Future research could take this into consideration by attempting to link mobile eye-tracking with wider data collection methods, such as mobile electroencephalogram (EEG) (19,20). This has the potential to provide more robust insight through establishing a stronger link between gaze distribution and cognitive processing. However, the development of such an approach is still at a stage of infancy, with methodological issues needing to be overcome (20).
