This study examined the effect of manipulating micro-scale physical environmental factors on an environment’s perceived invitingness for transportation cycling in adults. This is the first study investigating the effect of changing micro-scale environmental factors by using manipulated panoramic photographs. Based upon our quantitative and qualitative data, ‘evenness of the cycle path’ appeared to be the most important perceived environmental factor associated with invitingness to cycle for transportation. Limited research has examined this factor as a potential barrier to cycling. One Canadian study using questionnaire data found that when a route had potholes or uneven paving, the likelihood of cycling declined . Because most European research used the NEWS Questionnaire to assess environmental perceptions , where walking/cycling facilities were incorporated together, it was not possible to draw conclusions about the isolating effect of the evenness of the cycle path in these previous studies.
‘General upkeep’ together with ‘separation between cycle path and motorized traffic’ appeared to be the second most important factors to increase the invitingness for transportation cycling. Moreover, both environmental factors interacted with another environmental factor. A well-maintained environment without graffiti on the wall, broken windows, garbage and holes in the road was perceived as more inviting to cycle for transportation compared to a poorly maintained environment. Based on the qualitative data, a large part of the effect of general upkeep is probably explained by the hole in the road surface because it was considered dangerous when a car would avoid it and come closer to the cycle path. The same results were found in a non-European study , indicating the importance of good road pavement for cars: the higher the defects scores were of the road surface for motorized traffic, the lower the proportion of adults who cycled to work. ‘General upkeep’ seems to be especially relevant when it causes dangerous situations for cyclists. Furthermore, the positive effect of an even cycle path was stronger in a well-maintained compared to a poorly maintained environment. A combination of these factors could achieve a larger effect on the invitingness of transportation cycling, than to change them separately. Furthermore, the positive effect of having a separation between cycle path and motorized traffic on transportation cycling, was confirmed by previous studies ,. A recent observational study by Sallis and colleagues  found that implementing measures to improve cyclists’ safety from cars would increase cycling.
The second interaction effect reported in this study suggest that the positive effect of a separation with traffic could be reduced if there was a separation from the sidewalk as well. A possible reason, provided in the qualitative data, was the frightening feeling for cyclists that would be created when two separations are present on both sides of the cycle path. Another reason may be the choice of using bollards in the photographs to separate cyclists from pedestrians because participants are afraid to cycle against these bollards or see this as a disturbing factor that limited evasive options. This may also explain the non-significant main effect of a ‘separation between cycle path and sidewalk’ on the invitingness for transportation cycling. These results should be approached with caution because the provision of separate cycling facilities was the cornerstone of Dutch, Danish and German policies to make cycling safe and attractive . In these countries, city planners did not use bollards to separate cyclists and pedestrians, but grade separation, pavement coloring or surfacing and mentioned that it is important to present visual and physical, to indicate where cyclists and pedestrians are allowed to travel . It is possible that only pavement coloration, as was present on the pictures too, is enough to make a distinction between cyclists and pedestrians.
In both sorting tasks, the absence of traffic was also an important issue, although many participants are realistic about the necessity of cars and make no claim to get all roads traffic free. The impact of traffic danger has also been mentioned in the literature. Perceived and objective traffic danger have been negatively associated with transportation cycling, both the ‘volume’ (e.g., the street has a lot of motorized traffic) as well as the ‘safety’ aspect (e.g., the risk of collision with automobilists) ,. Nevertheless, a study of Foster and colleagues  found no effect of traffic volumes on transportation cycling and appeared to be more strongly related to leisure cycling than to transportation cycling.
The above mentioned significant positive associations of micro-scale modifications like an even cycle path, no obstacles, a separation from motorized traffic and low traffic level with the invitingness for transportation cycling, may have an important effect on safety. Because, safety is shown to be an important determinant regarding whether or not people will cycle , those small and easy changes are important to increase cycling, especially in countries where prevalence rates are still low due to lack of safety . These findings may have important policy implications as they suggest that safety measures may be more effective to promote cycling for transportation than measures to improve the aesthetic appeal of a street. However, further research in real-life settings is warranted to find out whether such modifications could change actual cycling behavior.
A wide cycle path, the presence of vegetation and the presence of a speed bump were important for the invitingness, but to a lesser extent compared to the other environmental factors. The qualitative data confirmed that these environmental factors were not considered as a priority for the participants. Previous research shows  that the minimum width of cycle tracks should be 78 inches (1.98 m) clear to provide safe passing for cyclists while overtaking another cyclist. Another issue is the various opinions of the participants concerning the presence of trees. The trees were mostly seen as an obstacle for cyclists as well as for pedestrians, that could hinder cyclists while avoiding the trees. Other results might be obtained if the trees would be placed somewhere else. Furthermore, it is difficult to draw conclusions about the aesthetics of vegetation because different types of vegetation were manipulated together in this study. These findings, compared to existing literature, indicate the complexity of the environment. The weak relationship between the presence of a speed bump and the increasing invitingness of transportation cycling could be explained with the help of the qualitative data. Many participants could not make the link between the presence of a speed bump and the advantage for cyclists. In the literature, evidence shows that speed bumps improve safety for cyclists . It might be less important for the increasing invitingness for cyclists, but it still remains an important component regarding the traffic safety.
Finally, no moderating effects of the demographic factors on the relationships between the environmental factors and the invitingness for transportation cycling were found. This finding may be encouraging for planning, because improvements of the micro-environment may have the potential to increase the invitingness of transportation cycling in both genders, the age group (45–65 years) and all educational levels. However, before drawing definite conclusions, these findings need to be replicated in a larger group of middle-aged adults recruited from different geographic areas.
The main strength of the present study was the experimental design, because causal conclusions on the effects of modifications in the environment on the invitingness can be drawn. Furthermore, these findings could be used to develop environmental interventions to determine if these findings will actually change the cycling behavior. A second strength was the collection of both quantitative and qualitative data. The qualitative data could help to figure out the underlying reasons why participants sort the pictures in a certain way. Third, there was the use of the manipulated panoramic photographs that have been validated to on-site responses. This allowed the possibility to ask for more items that were combined together at the same time.
This study also has some limitations that should be acknowledged. First, in this study the relationships with invitingness for transportation cycling was assessed and not with actual cycling behavior. Therefore, the present results can only give suggestions towards developing environmental interventions to determine if these findings will actually change the cycling behavior of adults. Environmental interventions in real life settings are needed to find out whether changing the micro-scale environmental factors, identified in this study, will affect actual cycling behavior. Second, in each sorting task, only five environmental variables could be manipulated. Adding an additional environmental factor would exponentially increase the number of photographs and the burden for the participant. Third, ‘general upkeep’ and ‘vegetation’ are environmental factors, consisting of many subcomponents that were manipulated simultaneously. Consequently, it is impossible to say which of the manipulated elements is crucial for changing the invitingness. Fourth, a limitation of using color photographs is the lack of movement . In real life, people notice different things in the environment depending on their speed of travel. The use of computer-generated virtual walk-through environments could be a suitable solution . Fifth, the study sample was relatively small, which might make the results less generalizable. Therefore, the findings need to be confirmed in larger samples. A last limitation of this study is that only one streetscape was used for this experiment. Consequently, it is not possible to generalize these findings to other streetscapes. In further studies it should be investigated whether the effect of micro-scale environmental factors on the invitingness for transportation cycling depends on macro-scale environmental factors. If micro and macro- environmental factors are interacting, future studies should also include different environmental macro settings, e.g., environments with high versus low land use mix diversity.