Selected Papers of #AoIR2024: The 25th Annual Conference of the Association of Internet Researchers Sheffield, UK / 30 Oct - 2 Nov 2024 Suggested Citation (APA): Valiati, V.; Lupinacci, L.; Soares, F. (2024, October). Wrap your head around it: Brazilian users’ algorithmic imaginaries of Spotify Wrapped. Paper presented at AoIR2024: The 25th Annual Conference of the Association of Internet Researchers. Sheffield, UK: AoIR. Retrieved from http://spir.aoir.org. WRAP YOUR HEAD AROUND IT: BRAZILIAN USERS’ ALGORITHMIC IMAGINARIES OF SPOTIFY WRAPPED Vanessa Valiati Feevale University Ludmila Lupinacci University of Leeds Felipe Soares University of the Arts London Introduction Spotify is one of the world’s most popular music streaming services, operating in over 180 countries. In what is now part of the ‘year-end list’ tradition, Spotify Wrapped summarises each user’s annual listening habits. Although Spotify is generally classified as a streaming service, initiatives such as Wrapped have an explicit component of sociability (Hagen & Lüders 2017) – in this case, not only because they are based on the harvesting of users’ behavioural data but also because they are created to be shared on platforms such as Instagram and Twitter/X. Indeed, Spotify Wrapped has acquired its own role in digital popular culture, inciting anticipation and excitement from users worldwide and becoming an 'algorithmic event' (Annabell & Rasmussen 2023) in and of itself. In this paper, we propose to scrutinise how this algorithmic event is perceived and understood by Brazilian users whilst also identifying and unpacking the platform affordances and algorithmic imaginaries (Bucher, 2017) that inform those interpretations and their associated performances of taste and identity (Airoldi, 2019, Prey, 2018). We explore, in particular, how users negotiate the tensions between algorithmic personalisation and individuation and the possibilities for shared experience to emerge during this event. Through a mixed-method approach, we argue that the 'eventness' (Frosh and Pinchevski, 2018) of Spotify Wrapped is distributed, clustered but sparsely connected, and marked by fleeting, fluid and ephemeral feelings of shared experience and recognition rather than by enduring communities, which in turn reflects and extends previous theorisations of affective publics (Papacharissi, 2014), everyday data cultures (Burgess et al., 2022), and social media liveness (Lupinacci, 2021). Algorithms, performance, and taste In an "algorithmic episteme" (Fisher & Mehozay, 2019), a focal point of extensive research revolves around platforms observing audiences through the collection of data on users' habits. The individual is perceived through a discernible pattern derived from the data they generate, including their tracks on the platform. Cheney-Lippold (2017) encapsulates this notion with "measurable types", through which individuals are temporarily classified as members of fragmented categories. At the same time, the affordances of digital platforms amplify the persistence and visibility of these users' discourses and taste practices (Airoldi, 2019). Therefore, it becomes crucial to assess how these affordances are integrated into the ways individuals reproduce discourses about their tastes and identity performance through their consumption and sharing of musical information (Webster, 2021). In this context, the assumption is made that musical identity undergoes a continual process of "algorithmic individuation," which "should be understood as a dynamic socio-technical process engaged in enacting the individual" (Prey, 2018, p. 1095). As a tool, Spotify Wrapped consolidates the characteristic elements of platformed consumption. It offers the promise of personalisation, leveraging the digital traces users leave throughout the year. Simultaneously, it contributes to the ongoing modulation and individuation of taste and identity, fostering a so-called "shared experience" during this algorithmic 'event.' Then, it becomes relevant to investigate these aspects that cause tension between individual and collective consumption practices: the distinction (how people perform and share their tastes and how they judge or fear being judged by others), whether there is indeed performance management and how this is reflected in their practices, the perceived need to share as an opportunity to participate in a flow of content, and how the platform shapes all these aspects. Methods To explore Brazilian users’ perception of Spotify Wrapped, we adopted a mixed-method approach. We combined Social Network Analysis of X/Twitter data, a manual review of the most shared messages, and contrasted those results with a preliminary thematic analysis of semi-structured interviews conducted with 25 Brazilian Spotify users. We used Brandwatch to collect X/Twitter data. For data collection, our search query included the combination of the words “Spotify” and “Wrapped”, and we filtered for messages in Portuguese only, as we focused on Brazilian users in this study. We collected messages posted in November and December 2023 to encompass the period before Spotify launched the Wrapped to its users and the time after that. We collected 22,283 messages, of which 12,745 were shares/retweets used for the Social Network Analysis. After generating a network graph, we manually reviewed the users with the highest indegree and their most shared messages. We used an arbitrary threshold of 50 weighted indegree to select the most shared messages. We then conducted an exploratory thematic analysis of the 41 most shared messages in the dataset. This dataset will undergo further examination through a content analysis based on the categories identified in the exploratory analysis. Additionally, we conducted 25 semi-structured interviews with Brazilian Spotify users to understand their engagement with Spotify Wrapped and how it affects their online self- presentations on social media. Interviews were complemented with elicitation techniques Preliminary findings and next steps The network graph based on X/Twitter data shows a variety of small clusters sparsely connected (Figure 1). Each of these small clusters is centred around a specific user who had their message highly shared on X/Twitter. Few users (less than 8%) shared more than one message, and only 26 (0.21%) shared five or more messages. Consequently, the average degree of the network is close to 1. This means that the network around Spotify Wrapped on X/Twitter in Brazil was mainly based on users sharing one or a few specific messages rather than engaging in-depth in community- building conversations. Fig 1. Network of X/Twitter users around Spotify Wrapped We identified three overarching categories in the preliminary thematic analysis of the interviews and the most shared messages on X/Twitter: Cultivating your Wrapped: users recognise their active role in ‘nurturing’ their annual listening habits to ensure their Wrapped is accurate, exciting, ‘beautiful’, or good. They recognise the labour of ‘cultivating’ (Siles et al., 2019) their media consumption to obtain affective gratifications; in this case, they acknowledge the happiness and success of getting the ‘right’ data stories. This often involves the incorporation of Spotify listening into existing routines (such as working out to the same playlist every single day) and the creation of new rituals to ensure the desired results. Managing contexts and algorithmic sensitivity: we identified strategies to manage and manipulate music consumption practices throughout the year to ensure that only what really ‘represents you’ (or, actually, what you want to represent you) is shown on Spotify Wrapped. We interpret this as a widespread hope that the platform’s algorithmic systems would somehow be able to identify the contexts of listening that are indeed ‘significant’ (and therefore worthy of being on your Wrapped) and those that are trivial or considered noise. Individual personalisation and the shared experience: participants acknowledge the social aspect of Wrapped, including the anticipation that others will be seeing their annual retrospective and will have opinions on it and the potential these data stories have for generating conversation and sparking connections. Our findings reflect previous theorisations of social media liveness (Lupinacci 2022), which argued that the emergent sense of 'shared experience' is often fleeting, fluid, and ephemeral and does not necessarily translate into a more enduring or complex sense of community. In this context of algorithmically-mediated 'affective publics' (Papacharissi 2014), being part of the content stream becomes an end in and of itself. When it comes to algorithmic imaginaries, then, we highlight the participants’ desire for ‘the algorithm’ to have a much more sophisticated sensibility to context and intention of listening, as the platform should be able to identify the listening sessions that are indeed “Wrapped-worthy” automatically. We also provide further evidence to the previously theorised idea that Spotify Wrapped is an emblematic example of algorithmic eventness (Annabell and Rasmussen, 2023) and of everyday data intimacy and data publics (Burgess et al., 2022), as it foregrounds the ambivalent belief and imagination of the powers endowed to algorithmic systems at the same time it demonstrates the perceived vulnerability of imagined surveillance surrounding the sharing of personal results. Funding This research is supported by the National Council for Scientific and Technological Development, Brazil (CNPq/MCTI Nº 10/2023 - 420387/2023-4). References Annabell, T.; Rasmussen, N.V. (2023). Spotify Unwrapped: How to Critically Examine Your Repackaged Data Stories. Presentation. Algorithms For Her 2, Sheffield, UK. Airoldi, M. (2019). Digital traces of taste: Methodological pathways for consumer research. Consumption Markets & Culture. https://doi.org/10.1080/10253866.2019.1690998 Burgess, J., Albury, K., McCosker, A., Wilken, R. (2022). Everyday data cultures. Polity. Bucher, T. (2017). The algorithmic imaginary: Exploring the ordinary affects of Facebook algorithms. Information. Communication & Society, 20(1), 30–44. Cheney-Lippold, J. (2017). We are data: Algorithms and the making of our digital selves. New York University Press. Fisher, E., & Mehozay, Y. (2019). How algorithms see their audience: media epistemes and the changing conception of the individual. Media, Culture & Society, 41(8), 1176- 1191. https://doi.org/10.1177/0163443719831598 https://url6.mailanyone.net/scanner?m=1rQprD-0004w5-3X&d=4%7Cmail%2F90%2F1705674000%2F1rQprD-0004w5-3X%7Cin6e%7C57e1b682%7C10917358%7C13772167%7C65AA870FDD03D7592441C1317111A0F6&o=%2Fphto%3A%2Fdts0ri.1%2F1.og5%2F08802301.2666191.089909&s=XgM-yoAulmviwMurAp3g4-T1Qu4 https://url6.mailanyone.net/scanner?m=1rQprD-0004w5-3X&d=4%7Cmail%2F90%2F1705674000%2F1rQprD-0004w5-3X%7Cin6e%7C57e1b682%7C10917358%7C13772167%7C65AA870FDD03D7592441C1317111A0F6&o=%2Fphto%3A%2Fdts0ri.1%2F1.og3%2F174164705937831918&s=THTSALw9yEphu1gn1Zuhejq2fRc Frosh, P., & Pinchevski, A. (2018). Media and events after Media Events. Media, Culture & Society, 40(1), 135-138. https://doi.org/10.1177/0163443717726007 Guest, G., K. M. MacQueen, and E. E. Namey. 2012. Applied thematic analysis. New York: SAGE. Hagen, A. N., & Lüders, M. (2017). Social streaming? Navigating music as personal and social. Convergence, 23(6), 643-659. Jørgensen, K.M. (2016) The media go-along. Researching mobilities with media at hand. MedieKultur, 60, 32-49. Lupinacci, L. (2021). ‘Absentmindedly scrolling through nothing’: liveness and compulsory continuous connectedness in social media. Media, Culture & Society, 43(2), 273-290. https://doi.org/10.1177/0163443720939454 Papacharissi, Z. (2014). Affective Publics. Sentiment, technology, and politics. Oxford University Press. Prey, R. (2018). Nothing personal: Algorithmic individuation on music streaming platforms. Media, Culture & Society, 40(7), 1086-1100. Siles, I., Segura-Castillo, A., Sancho, M., & Solís-Quesada, R. (2019). Genres as Social Affect: Cultivating Moods and Emotions through Playlists on Spotify. Social Media + Society, 5(2). https://doi.org/10.1177/2056305119847514 Webster, J. (2021). The promise of personalisation: Exploring how music streaming platforms are shaping the performance of class identities and distinction. New Media and Society, 25(8), 2140-2162. https://doi.org/10.1177/14614448211027863 https://url6.mailanyone.net/scanner?m=1rQprD-0004w5-3X&d=4%7Cmail%2F90%2F1705674000%2F1rQprD-0004w5-3X%7Cin6e%7C57e1b682%7C10917358%7C13772167%7C65AA870FDD03D7592441C1317111A0F6&o=%2Fphto%3A%2Fdts0ri.1%2F1.og3%2F174164700737726017&s=Lf7Pjpdkio0HTr38cKqItl5y4sU https://url6.mailanyone.net/scanner?m=1rQprD-0004w5-3X&d=4%7Cmail%2F90%2F1705674000%2F1rQprD-0004w5-3X%7Cin6e%7C57e1b682%7C10917358%7C13772167%7C65AA870FDD03D7592441C1317111A0F6&o=%2Fphto%3A%2Fdts0ri.1%2F1.og3%2F174164704037439529&s=-esZVVguXP7ztXilGxD7BhGSkL8 https://doi.org/10.1177/2056305119847514 https://doi.org/10.1177/14614448211027863