視覺複雜性與景觀偏好之關係探討

複雜性長期以來在景觀、心理學或其他應用科學領域之研究中都是相當重要的偏好影響因子，然而複雜性與偏好關係之相關研究發展至今都面臨一大問題—複雜性與視覺偏好關係並不穩定，在不同研究中往往會觀察到不同結果。有許多實證研究以及理論認為複雜性與偏好之間應為正相關、越複雜的環境越受偏好；有部分研究認為中等程度之複雜性將最受偏好；有些實證研究曾觀察到複雜性與偏好間有負相關或無相關之關係。針對複雜性與偏好關係的不穩定，Stamps III (2004)曾進行整合分析研究，發現複雜性與景觀偏好關係間確實無一致結果，卻並未進一步討論造成此現象的原因。過去研究曾提到不同研究間採用主觀或客觀複雜性測量指標可能是造成關係分歧的因素，其中主觀複雜之判斷受到對於刺激物意義的感知與理解影響、客觀複雜性則不受以上感知過程影響。對於刺激物物理層面之資訊處理會受到意義層面之資訊處理影響的論點，與心理學中處理流暢性理論相似，該理論指出大腦對於刺激物資訊的處理越流暢、越可理解接收到的資訊就會越偏好該刺激物。然而處理流暢性之研究方法未曾應用於景觀領域研究。經過文獻回顧，過往景觀研究與理論中有三項概念—熟悉度、一致性與自然度，定義上具有可使人容易理解環境之意涵，因此可作為可理解性指標之因子探討景觀中與主客觀複雜性及偏好之關係。本研究之研究一以整合分析探討複雜性與視覺偏好間整體關係之穩定性、以及檢視長期以來複雜性與偏好關係之研究結果不穩定的原因，並分為主客觀複雜性指標與研究材料類型差異兩種分類方式進行次群組分析。研究二與研究三中則以照片問卷調查法討論熟悉感、一致性與自然度三項可理解概念對於主觀複雜性、以及複雜性與偏好關係之影響，其中研究二之研究照片以網路蒐集、而研究三照片則使用電腦模擬不同植栽量、複雜度之都市與自然環境，並以迴歸分析探討熟悉感/一致性/自然度分別與主客觀複雜性與偏好之關係。研究一之結果與Stamps III (2004)相同，不同研究間複雜性與偏好之關係無穩定結果，但研究一進一步觀察到複雜性與偏好之整體效果將會是正相關，且造成關係不穩定的原因為使用主觀或客觀指標測量複雜性：主觀複雜性與偏好間較容易觀察到正相關結果，客觀複雜性與偏好之間則較容易觀察到負相關或無相關之結果。研究二與研究三之結果發現熟悉感與自然度確實可作為有助於理解環境之概念，但二者並未如預期地降低複雜性感知、反而會提高主觀複雜性；而三項因子中僅在控制自然度的情況下可同時觀察到主觀複雜性與偏好間有正相關、而客觀複雜性與偏好間無相關之結果。目前之研究結果指出以主觀或客觀指標評估環境複雜性將會導致與偏好關係之模型不同，並建議設計者可透過使人產生熟悉感的元素或自然元素來提高環境的主觀複雜性、進而提高對環境之偏好。

關鍵字

主觀複雜性；客觀複雜性；視覺景觀偏好；熟悉感；一致性；自然度

並列摘要

Complexity has long been seen as an important role in predicting visual preference in the fields of landscape and psychology. By far empirical studies about relationships between complexity and visual preference have reflected contradictive results – some research found a result of positive correlation, while others found negative, zero-correlation, or even an inverted-U relation. Regarding the instability of the relationship between complexity and landscape preference, Stamps III (2004) conducted a meta-analysis and found that the relationships between complexity and landscape preference is not stable indeed. However, the research of Stamps III (2004) did not clarify the reasons which might cause this result. Some studies had mentioned that using perceived or objective complexity measurements might cause the contradictive results. Perceived complexity involves in meaning content, and therefore affects information comprehension, while objective measurement doesn’t. And the argument that information processing at the physical identity of stimulus might be affected by processing meaning of stimulus is similar to the theory of processing fluency in psychology, which characterizes an experience of processing ease of “fluency”. However, the research method of processing fluency has not been applied to landscape research. This study found three factors from past studies that, by definition, can help understanding the environment: familiarity, coherence, and naturalness, which might be able to serve as measurements of “understanding” in landscape research. In order to investigate the overall relationship as well as the causation of the different results of past studies, Study 1 started with a meta-analysis of the relationship between complexity and visual preference. For further discussion of the impact of understanding on perceived complexity and the relationship between complexity and preference, we conducted Study 2 and 3 with the concepts of familiarity, coherence and naturalness. The result of Study 1 was similar to Stamps III (2004), pointing out the relationships between complexity and visual preference are unstable among different studies. Study 1 further observed that the overall effect of complexity and preference will be positively correlated, and the causation of different results was whether the studies used perceived or objective measurements for complexity. A Positive correlation between complexity and preference are more likely to be observed as using subjective complexity measurements, while a negative or zero correlation are more likely to be observed as using objective complexity measurements. The results above answered the long-standing doubts between the relationships of complexity and preference, but new questions came up to our minds: what cause the difference between perceived and objective complexity? Why is the relationship between perceived and objective complexity and preference different? These questions made us conduct the following research of Study 2 and Study 3 with two photo-based surveys. The results of Study 2 and Study 3 found that familiarity and naturalness indeed conduce to understand the environment, but either familiarity or naturalness reduce perceived complexity as expected but increase instead. When controlling naturalness, there is a positive correlation between perceived complexity and preference, meanwhile a zero correlation between objective complexity and preference. The results of the current study point out that evaluating environmental complexity by subjective or objective indicators might lead to different models of the relationship with preference, and suggest that designers can enhance the perceived complexity of the environment through elements which increase familiarity or naturalness, thereby prefer the environment more.