Essas descobertas podem ter implicaçaµes interessantes para pesquisas relacionadas a madia. Por exemplo, eles podem, em última análise, inspirar o desenvolvimento de teorias mais especaficas, baseadas em evidaªncias, sobre as tendaªncias...
![](data:image/jpeg;base64,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)
Um exemplo de meme de imagem. O manãtodo dos pesquisadores agrupa automaticamente memes com imagens e textos semelhantes, levando em conta as diferenças na codificação, resolução, marca d'a¡gua, filtro visual ou proporção. Crédito: Morina & Bernstein.
Memes, imagens ou vadeos com sobreposições de texto que transmitem mensagens engraçadas para paºblicos específicos são um fena´meno global bem conhecido. Como eles tendem a circular rapidamente on-line e se tornar virais rapidamente, os memes a s vezes também são incluados em estratanãgias de marketing e engajamento social.
Devido a velocidade com que circulam online, rastrear a origem dos memes nem sempre éfa¡cil. Alguns pesquisadores de madia acreditam que eles são difundidos primeiro em comunidades marginais da Internet e depois introduzidos em plataformas convencionais, enquanto cientistas comportamentais sentem que muitas vezes derivam de redes intermedia¡rias que conectam comunidades perifanãricas e convencionais.
Pesquisadores da Universidade de Stanford realizaram recentemente uma análise em grande escala da web com o objetivo de testar essas duas hipa³teses diferentes e determinar as fontes tipicas de memes de imagem online. Suas descobertas sera£o apresentadas na Conferência ACM sobre Trabalho Cooperativo Suportado por Computador e Computação Social, um evento internacional de primeira linha que apresenta as mais recentes pesquisas sobre interação humano-computador.
"Nosso artigo surgiu da pergunta: de onde vem a cultura da internet?" Michael S. Bernstein, um dos pesquisadores que realizaram o estudo, disse. "Estamos cercados por memes e outras formas de cultura da internet; mas quais comunidades estãooriginando o conteaºdo que vemos ao nosso redor todos os dias? A narrativa da pesquisa - incluindo minha própria pesquisa anterior - sugere que muitos memes influentes vão de comunidades de fora o mainstream, enquanto a pesquisa sociola³gica sugere que as comunidades que fazem a ponte entre o mainstream e a periferia são as mais influentes culturalmente."
Embora muitos artigos anteriores tenham discutido as possaveis origens do conteaºdo viral na internet, atéagora ninguanãm havia conduzido uma análise de memes em escala da web para rastrear claramente sua fonte. Bernstein e seu colega Durim Morina se propuseram a preencher essa lacuna na literatura, examinando especificamente a origem e disseminação de memes de imagem.
Em primeiro lugar, os pesquisadores começam a navegar na web para identificar comunidades online que postam memes de imagens em inglês. Posteriormente, eles usaram técnicas de visão computacional incorporadas em APIs modernas de visão em nuvem para rastrear a disseminação de cada imagem para trás, para determinar por quais comunidades elas circulavam e de onde se originavam.
"Isso acabou nos dando um grande conjunto de comunidades que poderaamos rastrear", explicou Bernstein. "Escrevemos manualmente scripts que monitoravam toda a atividade em qualquer uma dessas comunidades na web, extraaam ca³pias de cada meme postado nele e usamos a API de visão em nuvem para verificar se ele nunca havia sido visto antes no web pública. Isso nos deu o que precisa¡vamos: a capacidade de rastrear cada ca³pia de um meme de imagem de volta a comunidade que o postou pela primeira vez na web pública e identificar todas as vezes que foi espalhado em outro lugar."
De modo geral, as descobertas reunidas por Bernstein e Morina sugerem que os memes da internet não se originam de comunidades perifanãricas, mas que são extremamente centralizados. Mais especificamente, eles descobriram que a maioria dos memes de imagem são publicados primeiro no Reddit e depois compartilhados em outras plataformas.
Essas descobertas podem ter implicações interessantes para pesquisas relacionadas a madia. Por exemplo, eles podem, em última análise, inspirar o desenvolvimento de teorias mais especaficas, baseadas em evidaªncias, sobre as tendaªncias com as quais os memes de imagem são disseminados online.
“Nossa análise mostrou que raros memes transformadores ainda podem vir da periferia, mas no dia-a-dia, a maior parte do conteaºdo que vocêvaª épostado pela primeira vez em uma plataforma mainstreamâ€, acrescentou Bernstein. "Nosso pra³ximo trabalho envolvera¡ a criação de memes, mas acho que também éimportante considerarmos se estamos satisfeitos com a centralização da cultura da internet. Vocaª obtanãm um grande paºblico postando no Reddit, TikTok e assim por diante, mas as subcomunidades desempenham um papel importante papanãis em nossa cultura e não estãoclaro se nossos designs de madia social estãoapoiando isso tão bem quanto poderiam atéagora."