Selected Papers of #AoIR2023: The 24th Annual Conference of the Association of Internet Researchers Philadelphia, PA, USA / 18-21 Oct 2023 Suggested Citation (APA): Bhandari, A. & Bimo, S. (2023, October). Algorithms, Aesthetics, and the Changing Nature of Cultural Consumption Online. Paper presented at AoIR2023: The 24th Annual Conference of the Association of Internet Researchers. Philadelphia, PA, USA: AoIR. Retrieved from http://spir.aoir.org. ALGORITHMS, AESTHETICS AND THE CHANGING NATURE OF CULTURAL CONSUMPTION ONLINE Sara Bimo York University Aparajita Bhandari University of Waterloo Introduction This paper examines the development of digital subcultures and microtrends in a social media landscape increasingly driven by algorithms. Whereas in the past people may have identified as members of subcultures such as “hippies” or “punks” (groups which were typically united by shared experiences, interests, and/or environments) (Hebdige, 2012), in the modern social media landscape increasingly niche subcultures are proliferating in the form of “microtrends.” Niche aesthetic movements such as the “coastal grandma,” “goblin academia,” and “chic modernist,” (Aesthetics Wiki, 2023) appear seemingly out of nowhere, enjoy a rapid rise in popularity, and often die out just as quickly. Such trends call on users to partake in a highly specific aesthetic performances which are often tied to a particular lifestyle philosophy. While these trends often borrow elements from “real world” style movements, they are typically born entirely on the internet and are intimately shaped by the data-driven recommendation systems which have come to structure modern social media environments. Whereas pre-digital subcultures spread values and aesthetic sensibilities through word-of-mouth or pre-digital media, the algorithmic recommendation systems of the Web 2.0 age allow users to instantly connect with both like-minded aesthetes and similarly aligned commodities; thus, trends are both formed and propagated through the help of algorithms which connect similar data-groupings (Bondy & Burgess, 2021; Seaver, 2019). In this project, we aim to extend scholarship on digital subcultures by examining the intersection of identity formation, algorithmic capitalism and user practices surrounding microtrends. Whereas past literature has emphasized the environmental impact and labour costs of microtrends (Reilly & Hawley, 2019, Lamm, 2022), we shift focus to instead examine this phenomenon from the perspective of user engagement and self- identity. We are guided by three related research questions: (1) What tactics and practices constitute user participation in microtrends? (2) How does user engagement with microtrends function as an act of relational self-expression? (3) What are user discourses surrounding microtrend participation? Methods We employed a qualitative mixed-methods analysis that combined visual discourse analysis and Critical Techno-cultural Discourse Analysis (CTDA) (Brock, 2018). We examined the content shared within the popular hashtag #aesthetic across three algorithmically driven platforms: TikTok, Instagram and Youtube. Due to its widespread usage across platforms and trends, #aesthetic provided a useful access point through which we could examine a large variety of microtrends. Visual discourse analysis was chosen as it is especially useful as a method for examining the meaning and rhetorics embedded in the circulation of visual messages (such as videos and images), which were the primary forms of content we were interrogating across platforms (Rose, 2022). We paired this visual discourse analysis with a CTDA which allowed for an interrogation of the material and semiotic nature of digital artifacts within the specific context of users’ offline social and cultural practices. Brock (2018) defines CTDA as “[an] analyses of information technology material and virtual design with an inquiry into the production of meaning through information technology practice and the articulations of information technology users in situ.” (Brock, 2018, p. 2). CTDA highlights the power relations embedded in technological artifacts, allowing for the in-depth examination of the relationships between technology, culture, and user practices. By combining these two methods we were able to analyze both the visual and textual content shared within the hashtag #aesthetic, paying attention both to the visual messaging and the broader platform infrastructures in addition to audience engagements and interactions. Findings and Analysis Aesthetic Consistency. We found that the visual style of posts belonging to a particular microtrend was remarkably consistent; microtrends appeared to be largely understood (and reproduced by users) as a set of recurrent symbolic markers and motifs. Posts tended to contain a discrete set of common elements which were then described through hashtags. For instance, the “coastal grandmother” aesthetic features cream- hued linen, Hamptons-esque scenery, and woven sun hats as its hallmarks. This suggests that existing microtrend categories are relatively inflexible and standardized. Users appeared to be largely aware of the particular motifs that characterize a particular aesthetic and were unlikely to experiment within a given trend. The specificity of these trends was somewhat at odds with their affective intentions. For example, popular trends such as the “clean girl aesthetic,” and the “coconut girl aesthetic” displayed a tension between the formulaic nature of user participation and the ideals of effortlessness and ephemerality that were promoted by the trend. Aesthetic Anxiety. Users displayed anxiety over matters of categorization and identification. User discourses surrounding microtrend participation indicated that they are invested in “correctly” ascribing and associating with microtrends. For example, users often policed the boundaries of a given trend, arguing over what the hallmarks of the aesthetic were and debating whether a given post fit into the tagged aesthetic well enough. In the case of “boundary” content–that is, content that attempted to align with a particular micro trend, but in some way departed from the norm–users would often identify the content as an entirely new aesthetic rather than allow for an existing trend to develop and mutate. In addition, users frequently responded to more ambiguous posts by asking questions such as “what aesthetic is this?” and “how do you get this aesthetic?” These common refrains revealed a preoccupation with categorizing content according to a pre-set identifier, rather than accepting ambiguity. Aesthetic Creation. Users were often actively involved in the creation of new microtrends, often based on particular elements of their own identities. This creative process often involved the abstraction of a particular element of their behaviors or personalities; for instance, one user attempted to start a “museum girl” trend which aestheticized their own penchant for visiting museums via the fragmentation of this experience into a set of discrete markers (such as framed oil paintings, sketch pads, and exterior shots of aesthetically pleasing buildings). Such posts are often meticulously tagged and indicate that users have a high degree of expertise in both the aesthetic and technical logics of microtrend construction. Users do not simply wish to create a new trend–they also aim to make the trend highly visible. Users furnish recommendation algorithms with easily identifiable and marketable aesthetic markers and make posts easy to share and recommend to others via accurate tagging. Discussion When brought into dialogue with the underlying processes of data-fication and abstraction that structure social media environments, these user practices demonstrate a fascinating reflection of algorithmic techno-logics in human behavior. Just as algorithmic architectures abstract and discretize human behavior into data-elements, users appear to be discretizing their own behaviors and identities through their participation in microtrends. We theorize these user practices (the flattening of lifestyles into discrete aesthetic markers, the preoccupation with categorizing and organizing, and the fragmentation of identity into easily packaged trends) as an emergent process of “self-discretization,” wherein users “do the work” of abstracting and fragmenting their identities for the sake of attaining visibility within a data-fied digital environment. In this way, humans internalize not just the messaging, but also the logics of algorithmic capitalism. References Kaye, B. & Burgess, J. (2021) Algorithmic recommender systems and everyday data cultures: the view from jazz TikTok. In Selected Papers in Internet Research 2021: Research from the Annual Conference of the Association of Internet Researchers. AoIR - Association of Internet Researchers, United States of America Brock, A. (2018). Critical technocultural discourse analysis. New Media & Society, 20(3), 1012-1030. Hebdige, D. (2012). Subculture: The meaning of style. Routledge. Lamm, A. (2022). You Need This: Global Fast Fashion, Environment, and Consumption Culture. Reilly, A., & Hawley, J. (2019). Attention deficit fashion. Fashion, Style & Popular Culture, 6(1), 85-98. Rose, G. (2022). Visual methodologies: An introduction to researching with visual materials. Visual methodologies, 1-100. Seaver, N. (2019). Captivating algorithms: Recommender systems as traps. Journal of material culture, 24(4), 421-436. List of aesthetics. Aesthetics Wiki. (n.d.). Retrieved February 2, 2023, from https://aesthetics.fandom.com/wiki/List_of_Aesthetics