Findings that access to natural environments is associated with psychological well-being and physical health are often attributed in part to the restorative qualities of natural environments. Experimental studies focused on restoration have often compared a natural environment with an urban environment, and they have with some consistency found a restorative advantage of the natural environment. Comparisons among different built urban environments are however lacking, as is knowledge of how the physical attributes of the built urban environment might affect possibilities for restoration. This kind of knowledge is needed. Some estimates suggest that 70% of the world’s population will be urban dwellers in 2050. Many of those people may be living in an environment that is harmful to health and well-being. Yet, even with limited access to wild areas and parks, the urban environment need not be so deleterious. The present study is based on the premise that the physical environment of the city can also promote restorative experiences. The focus of the study is on urban streetscapes. A substantial part of people´s daily activities takes place on or beside streets, so streetscapes may come to have a noticeable effect on people´s health and well-being. In focusing on streetscapes, we chose to inspect two attributes, complexity and enclosure, as both have been shown to affect environmental preferences, a variable known from previous research to be sensitive to possibilities for restoration. The study involved computergenerated simulations of streetscapes (city blocks) without vegetation. Each block consisted of 12 buildings, each depicted as 8 m wide. The buildings on the two sides of the street were separated by 10 m, a distance taken up by the street (6 m wide) and sidewalks (each 2 m wide). All buildings had a doorstep. Complexity was manipulated in terms of the silhouette complexity and surface complexity of the buildings lining the street. There were two levels of both silhouette complexity (peaked versus flat roofs) and surface complexity (building facades with few details versus many details, such as door and window trims, a cornice, ornamentation, and a decorated handrail at the doorstep). Level of enclosure was manipulated in terms of the height of the buildings (1-3 stories). An image pool of 148 virtual streetscapes was created. The images were prepared in Google SketchUp 7 and rendered in VRay v1.5.1 for more photorealistic purpose. A screenshot was taken of each of the streetscapes, which included street lamps and traffic signs as well as the buildings, street, and sidewalks. No cars or people were included. The individual streetscapes are the units of analysis in the study. In addition to the physical variables described above, each of the streetscapes has a value for several psychological variables, including perceived complexity, perceived enclosure, preference, the likelihood of restoration, and two constructs in attention restoration theory, being away and fascination. These values are the means of ratings assigned by groups of participants. The rating data were collected from 263 Icelanders, 38% males and 62% females with mean age of 40 years (SD±10.9). Each of them rated only half of the images in the total pool (i.e., 74 images) on only one of the psychological variables. Those participants who provided a rating for the images on being away, fascination, or the likelihood of restoration were asked to bear in mind a scenario that specified a need for restoration. The primary focus of analysis is on the associations that silhouette complexity, surface complexity, and enclosure (i.e., the manipulated image characteristics) have with ratings of restoration likelihood, and on the possibility that those associations are mediated by the restorative quality variables being away and fascination. The presentation will cover the results of these analyses, which are currently underway.