本研究以《牛頓》為析論核心，透過社會與歷史條件的爬梳、深度訪談以及視覺框架分析，從中探究1980年代臺灣科學視覺化現象，研究發現：一、就科學視覺化成因而言：1980年代，國民所得提升、加諸官方宣導「書香社會」，高單價的《牛頓》打破時人對於科普雜誌訴諸文字的理解，形成「嘆為觀止」的視覺感受，可稱之為視覺創新「表現」之一環，促成精美印刷、圖片彩印的科普出版風潮。二、就科學視覺化認知而言：《牛頓》認為圖解科學較照片真切；《科學眼》則堅持攝影比圖片更加真實，兩者皆旨在「求真」，也就是視圖片與照片為客觀、中立超然的載體，旨在佐證科學理論。三、就科學視覺化實踐而言：《牛頓》自製專欄的圖片依循日文版視覺框架，建構科學視覺化內容，無論就圖像配色、表徵形式以及符號象徵，皆與日本一致。但在意識形態的表徵上，試圖加入自身所處情境的元素，也讓讀者看見與日文版不同的視覺意象。

Compared to recent specific and micro-focused research on scientific visualization in European and American academic circles, before 2019 there have been few systematic studies in the field of science communication in Taiwan. By 2020, the East Asian Science, Technology and Society: An International Journal had published a special issue titled “Thinking and Acting with Diagrams” to discuss scientific visualization. Yin-Chung Au (區曣中) (2022) subsequently analyzed the visual representation of European and American biology through a literature review, but neither of these two studies closely relate to how scientific visualization is presented in Taiwan. This study accordingly takes Newton magazine as the object of analysis and includes popular science magazines from the same period, particularly Science Eye. Through exploring historical and social conditions and conducting in-depth interviews and text analysis, the following three research questions are understood. (1) What historical and social conditions contributed to the scientific visualization phenomenon of Newton in the 1980s? (2) How did the editors of Newton and other popular science magazines from the same period perceive and practice scientific visualization? (3) What are the practical outcomes of scientific visualization in popular science magazines in Taiwan? From the literature review, this study mainly refers to the following inquiries. (1) Through Mithcell (1994) and Boehm (1994, cited in Bertolini, 2015), their advocacy of “the turn to Pictorial/Iconic” is not only concerned with “pictorial representation”, let alone “Pictorial presence”, but also the rediscovery of post-linguistics and post-semiotics related to vision, installation, discourse, and even body intertextuality. Three orientations of visualization research are sorted out from this: visual representation/representation, visual thinking, and knowledge visualization. (2) In contrast to visual cultural studies that advocate for “the turn to Pictorial/Iconic”, scientific visualization research has only gradually received attention since the late 1980s. This study mainly refers to Daston & Galison’s (2007) analysis of “objectivity” behind scientific visualization, including: the capture of the “truth-to-nature” through artistic representation of nature, the depiction of nature through mechanical objectivity, and the cultivation of “trained judgment” with interpretive expertise. (3) In the 1960s, under the influence of the “strong program”, researchers began to focus on laboratory data recording, interpersonal interaction, equipment operation, and other aspects involved in scientific visualization. This means treating scientific images as a “process of inscription” (Latour, 1990) involving representational device (RD) and representational object (RO). Following the perspective of “inscription”, Pauwels (2006c) points out that the exploration of scientific visualization involves not only concrete materials or abstract concepts but also social and technological conditions, as well as conditions related to the functioning of media and user feedback. (4) The methodology of scientific visualization can be divided into the following three aspects. First, from visual language to immutable mobiles: Rudwick (1976) notes that the emergence of “visual language” means that followers of the geological community must learn the rules of this language through professional training to accept and understand the norms, thereby achieving scientific communication through representation. Continuing this concept, Lynch (1990) discusses how scientists manipulate biological images and classify photographs as diagrams, stating that diagrams are a “simplification” and also an “idealization” of photographs. Latour (1990) proposes the concept of “immutable mobiles”, which refers to the ability of “images” to bring different things together in time and space - that is, on the one hand, the image defines objects, and on the other hand, it can be carried and moved to anywhere. Second, the classification of visualized scientific knowledge: Eppler & Burkhard (2006) classify knowledge visualization into six forms: heuristic sketches, conceptual diagrams, visual metaphors, knowledge animations, knowledge maps, and scientific charts. Third, a new attempt at analyzing scientific visual frameworks: Rodriguez & Dimitrova (2011) apply Goffman’s (1974) framework theory to visual exploration and construct four analysis systems for visual frameworks, including (1) denotative systems: descriptions made through picture titles, captions, and related texts; (2) stylistic-semiotic systems: the presentation of stylistic conventions and social meanings; (3) connotative systems: examining the presentation of icons, indices, and symbols, with symbols being the focus of analysis due to their social significance; and (4) ideological representation: examining cultural conditions and social factors to explore the economic and political interests in the visual framework. In terms of research methodology, this study considers the “analytical framework influencing the production of scientific visualization,” as Pauwels (2006c) outlined. Since the concepts and aspects involved are relatively concrete, this framework forms the basis of analysis. It incorporates Latour’s (1990) interpretation of drawing and Burri & Dumit’s (2008) suggestions on exploring scientific visualization. Eppler & Burkhard’s (2006) classification of knowledge visualization is also integrated, along with the scientific visualization framework elements suggested and compiled by Rodriguez & Dimitrova (2011) and Wardekker & Lorenz (2019). Three aspects are analyzed separately: (1) social and historical conditions (Pauwels - social and technological production process); (2) media operation (Pauwels - media, visual representation); and (3) execution of scientific visualization style (Pauwels - referent, form/execution). The study presents the following findings. (1) The causes of scientific visualization: In the 1980s, the increase in Taiwan’s national income and the official promotion of a “reading society” led to the high-priced magazine Newton breaking people’s understanding of popular science magazines as relying solely on text and thus forming a “breathtaking” visual experience. This is one aspect of visual innovation “representation”, which promoted the trend of high-quality printing and color printing of images in popular science publishing. (2) The cognition of scientific visualization: Newton believes that graphic illustrations are more realistic than photographs, while Science Eye insists that photography is more truthful than graphics. Both aim for the pursuit of truth and consider graphics and photographs as objective, neutral carriers to support scientific theories. (3) The practice of scientific visualization: Taiwan’s editors made the images in columns of Newton follow the visual framework of the Japanese version to construct scientific visualization content. Whether it is the color matching, representational form, or symbolic representation, they are all consistent with Japan’s Newton. However, in terms of ideological representation, an attempt was made to incorporate elements of their own context, which also allow readers to see visual imagery different from the Japanese version.