Problem. As technology advances, architectural design methodology changes in response; most significantly the design process in recent years has been bolstered by computer technology, and digital fabrication. As this process often gives rise to non-rectilinear buildings, scrutiny continues over the appropriateness of the resultant architectural forms. In turn, this gives impetus to study the new forms generated for their effects on the inhabitants' wellbeing. Neuroscientific studies show more positive responses to curved contoured objects than to sharp contoured objects (Bar and Neta 2006; Leder, Tinio and Bar, 2011), and more activation in the amygdala when subjects view images of sharp objects versus curved objects (Bar and Neta 2007). Nanda and Pati (2009) explored the relevance of the aforementioned studies to architecture and made a case for further research. A subsequent architectonic study suggests participants viewed curved architecture as more beautiful than its rectilinear counterpart (Vartanian et al., 2013). Method. The present study tested preference levels, using six sets of bipolar adjectives, of four different architectural geometries and building forms in an innovative semi-immersive virtual environment ("CAVEtte"), designed and built by the author and a collaborator. All four buildings chosen in the study were built "Maggie's Centres" (cancer care centers) around the United Kingdom designed by different architects, all of whom were given the same architectural brief. The designs were digitally modeled in Rhino based on available plans, sections, elevations and images, and presented to 65 students and employees (19 females and 46 males) at NewSchool of Architecture + Design (NSAD). Each participant watched four walk-through videos, one of each building, in a randomly selected order. After each video, participants filled out a subjective survey. Due to limitations in the immersive experiment, a second similar study was conducted to ensure the limitations found did not affect the results. In the follow-up study, 20 new students and employees of NSAD were tested. In addition to both the immersive experiments, two 3-Second Surveys were conducted (one of abstract shapes, and the other of black and white images of the selected Maggie's Centres) in an effort to triangulate the data. Results. An Analysis of Variance (ANOVA) was conducted to compare results between both immersive experiments and found there was no statistical difference among the curved, mixed and angled walk-through videos. However, there was a statistical difference found between Experiment 1 and 2 with regards to the rectilinear walk-through video. Due to the ANOVA results, the data presented is from Experiment 1 only. Overall, the immersive experiment findings suggest people prefer curved buildings to other geometric architectural interior environments. While these findings are consistent with the aforementioned contour-focused studies, there were some intriguing novel results when the data were parsed demographically along the lines of age, gender, education level, designers versus non-designers and years in the profession. These results demonstrate the need for future studies in a highly-immersive virtual environment to continue examining the effects of architectural contours at a physiological level. Additionally, while the 3-Second Black and White Survey slightly supported the hypothesis (the image of the mixed building was selected most, followed by the curved building), results from the 3-Second Shape Survey did not support the hypothesis, and in turn led to inconclusive results with room to examine potential limitations.