環境體驗中感知危險架構之探討

Abstract

感知危險為景觀環境及休閒遊憩領域中的重要影響因子之一，環境若引發危險感受，可能降低環境品質，驅使人們迴避該環境，甚至對健康產生負面影響。儘管感知危險有此重要性，過去研究對其定義卻缺乏一致性，導致測量問項多樣化，研究結果難以比較與討論，進而阻礙理論概念的推進。 此外，現有研究在研究設計上也存在高度的差異。在討論範疇上，可分為不限於特定環境事件和限於特定環境事件兩類；評估方法則包含同時評估多個變項與僅針對單一變項進行評估兩類；而刺激呈現工具方面，現地調查與照片評估雖為常見做法，但仍存在侷限性。 本研究旨在釐清感知危險的核心構面與評估方法，並建構具操作性的理論架構與測量工具。論文分共分為三部份研究：第一，藉由比較不同呈現刺激之工具，檢驗沉浸式虛擬實境（Immersive Virtual Reality, IVR）技術應用於感知危險評估之適用性；第二，透過專家諮詢及照片問卷調查，探討感知危險的定義與核心構面，並建立量表架構；第三，藉由質性訪談與照片問卷調查，探討在具體環境與事件中進行構面評估的可行性與效益。 第一部份結果顯示，IVR之日夜感知危險評估皆與現地評估高度一致，而照片僅於日間與現地無顯著差異，夜間則出現明顯落差。因此，IVR技術不論日夜均可作為感知危險評估中有效的刺激呈現工具，照片則較適用於日間環境。 第二部份結果顯示，「可能性」和「後果」為感知危險概念的必要組成；「可控性」和「迴避情緒行為」則分別為其前因與結果。研究亦證實，同時評估「可能性」和「後果」不至於造成月暈效應及共同方法變異，顯示在感知危險評估上的適用性。 第三部份結果顯示，在刺激中包含非特定負面事件或特定情境擔心事件時，受測者對於「可能性」與「後果」兩者之判斷更為清晰明確。此結果支持於特定環境與事件脈絡下，能更細緻地探討感知危險各構面之關聯性。 本研究結果可釐清感知危險的定義與構面，感知危險研究適用之研究設計，並驗證IVR技術於感知危險評估之適用性。結果除了有助感知危險理論之深化與發展，也未來研究發展與設計提供理論依據，並有助於實務環境管理與設計之參考，協助提升公共空間之品質及管理效能。

Perceived danger is a critical factor in landscape environments and recreational studies. When an environment evokes a sense of danger, it may lead to decreased environmental quality, discourage users from engaging with the space, and even have adverse effects on health. Despite its importance, the definition of perceived danger has lacked consistency in previous studies, resulting in a wide range of measurement items. This definitional inconsistency may hinder cross-study comparisons and impede theoretical advancement. Moreover, substantial difference exists in research designs, including variation in discussion scope (general vs. event-specific assessments), evaluation methods (multi-variable vs. single-variable approaches), and the tools used to present stimuli (e.g., on-site, photo and IVR), each with their own methodological limitations. This study aims to clarify the core dimensions and appropriate research methods of perceived danger, and to develop a practical theoretical framework and measurement tool. The research comprises three parts: (1) evaluating the applicability of immersive virtual reality (IVR) as a stimulus presentation for perceived danger assessments, (2) identifying and validating the core components and definition of perceived danger through expert consultation and photo-based surveys, and (3) examining the feasibility and effectiveness of conducting dimensional evaluations within specific environmental-event contexts using qualitative interviews and photo questionnaires. The first part results showed that IVR were highly consistent with on-site assessments in both time conditions, whereas photo matched on-site evaluations only during daytime, with notable discrepancies at night. This suggests that IVR is a valid method for assessing perceived danger both day and night, while photo may be more suitable for daytime evaluations. The second part findings confirmed that these two dimensions are essential to the conceptualization of perceived danger. "Perceived control" and "avoidant emotional response" function as antecedents and outcomes, respectively. Furthermore, results showed that jointly measuring likelihood and outcome does not produce halo effects or common method bias, supporting methodological suitability. The third part findings revealed that when including negative events, participants made more distinct and accurate judgments of both dimensions. This supports the feasibility and value of applying dimensional assessments within defined contexts for a more nuanced understanding of perceived danger. Overall, this study clarifies the definition and structure of perceived danger, proposes suitable research designs, and empirically validates the applicability of IVR technology for perceived danger assessments. The results contribute not only to the theoretical advancement of the field but also offer practical implications for enhancing the quality and safety management of public spaces.