Investigating short video consumption practices by individuals with visual impairments in China: questions of how to ‘see’

Abstract

Human visual experience is increasingly mediated. As technology facilitates the sharing of diverse images, understanding the world through photography and videos is crucial. However, visuality affects the digital existence of communities with visual impairments. Focusing on the short-video consumption of individuals with visual impairments, this study examines their media experiences in a visually dominant society. Considering its social and technical setting, China is the ideal research site. Using constructivist grounded theory, we implemented open coding on raw data, highlighting initial categories associated with the usage of short-video applications. Contrary to the perception that those with visual impairments are more likely to use auditory-focused devices, we found that this population uses short video apps for entertainment, work, and social interactions and found design barriers and limited accessibility to be additional challenges. This study explains how digital media promotes inclusion and suggests possibilities for more inclusive digital-media development.

Points of interest

1. In today’s society, where visual information is pervasive, the ways in which people with visual impairments engage with media are often overlooked.

2. This study explores how people with visual impairments in China consume short videos. Despite facing challenges in watching these videos, they are able to engage with short video apps by listening to audio and using touch screen functionalities.

3. People with visual impairments use short videos to socialise, which helps enhance their sense of inclusion within society.

4. While expressing themselves on short video platforms, people with visual impairments may inadvertently contribute to the amplification of public prejudice towards their community.

5. In this research conducted in China, participants with visual impairments shared two main aspirations: to access information easily without barriers and to be accepted and recognised in society without conditions.

Keywords:

• Visual culture
• short video
• visual impairments
• digital technology
• social inclusion

Introduction

Vision is the dominant sense that plays a critical role in every facet of life (World Health Organization 2019). With the increasing prevalence of digital media, digitised photos and other (audio) visuals have become ubiquitous in our daily lives (Wildfeuer et al. 2018). Short video applications have become a popular form of media in China; as of June 2022, 94.6% of internet users (995 million individuals) use these apps (China Internet Network Information Center 2022). Moreover, the amount of time that people spend using short video applications is now even longer than the time they spend using instant messaging, making these applications the ones used for the longest periods of time (China Internet Network Information Center 2022). However, the acceleration of social production and the complexity in obtaining visual information have become challenges for individuals with visual impairments who seek access to, and interact with, short video applications.

Disabilities are essential determinants of digital inequality in various countries and regions (Glumbić, Đorđević, and Brojčin 2022). In China, 90% of individuals with visual impairments believe that Internet is essential in their lives (‘Online Social Report for the Visually Impaired’ 2019), and short videos are an emerging force competing for online time. Meanwhile, interactivity emphasised by digital media has again excluded users with visual impairments. The ocularcentric design of short video applications, such as touch screens and new media designs, has increased the difficulty experienced by individuals with visual impairments in using them (Li 2013). While some assistive tools can help this population use short video applications more easily, the ability of these individuals to completely integrate into a visual culture dominated by short videos remains a challenge. Social inclusion comes at the expense of privacy, independence, and access to timely information (Ravenscroft 2019).

Short video platforms are designed primarily for individuals with intact vision, resulting in systemic visual hegemony that excludes individuals with visual impairments from accessing relevant visual information and entertainment, leisure, work training, community participation, and civic-engagement opportunities (Levin 1993). In China, short video applications, such as Douyin and Kuaishou, are commonly used; they serve as examples of how those with visual impairments find it challenging to navigate a visually dominated digital culture. This study seeks to determine how individuals with visual impairments use short video applications, explore the tensions that exist in a nonmainstream visual culture within the digital society, and elucidate the relationship between individuals with visual impairments, technology usage, and social integration. We conducted localised research in China using grounded theory for coding analysis, based on data gathered through ethnographic observations and semi-structured interviews.

Literature review

Visualising society and ‘disabled’ seeing

With advancements in media technology, humans’ visual experience has become increasingly mediated. Images, including photographs and videos, are used to represent the world (Mirzoeff 2015), resulting in the diffusion of vision, where the rationale for visualisation is extended to non-visual objects (Zeng 2009). The eye has long been recognised as a participant in shaping the truth and is the most fundamental aspect of human cognition (Bolt 2005), not only because knowledge can only be experienced through one’s eyes but also because vision is inherently synesthetic. The act of seeing is mutually permeable to other sense-based activities; thus, listening and reading can also be visualised (Bal 2003). However, research on visual culture has critiqued society’s visual oppression, arguing that breaking free from this oppression requires rediscovering the body’s perceptual value (Foster 1998; Liu 2011; Wu 2006). The development of media technology has facilitated the extension of individuals’ eyes, obscuring other parts of the body and producing a perceptual experience dependent on the denial of the body (Crary 1990). The material processes that underlie a mediated society and meaning production involve platforms and infrastructures (Couldry 2016).

The corporeal eye is a specific social imagination with a social structure. Individuals with visual impairments face oculocentric scrutiny in which the omnipresence of images represents symbolic violence against individuals with visual impairments (Bolt 2005). The clinical gaze constructs a visual image with a social identity, perpetuating prejudice through the use of terms, such as ‘blindness’ and ‘the blind’ (Calder-Dawe, Witten, and Carroll 2020; Bolt 2004). Vision-based derogatory discourses expose individuals with visual impairments to ignorance (Porkertová 2022; Jessup, Bundy, and Cornell 2013; Steer and Gale 2006). Their visual experience is limited by their physical ability to see and is influenced by social practices, such as the visual understanding of race, gender, age, and other aspects that stem from interpersonal and institutional socialisation (Obasogie 2010). Individuals with visual impairments struggle to fit into digital technology, media platforms, and visual societies designed by the sighted people, magnifying their inability to see and be seen.

Digital disability and social inclusion

Observing individuals with visual impairments has become more ambiguous. The question of whether the development of media technology, especially the widespread use of short video applications, supports or challenges these individuals has been raised. Sweet et al. (2020) concluded that the relationship between individuals with disabilities and social media is centred on six themes: community, cyberbullying, self-esteem, selfdetermination, access to technology, and accessibility. The impact of digital use on individuals with disabilities generally remains in both the digital dividend and divide. Regarding digital dividends, the spread of digital media facilitates social interactions among individuals with disabilities, further facilitating the development of social relationships (Batey and Waine 2015; D’Aubin 2007; Dobransky and Hargittai 2006; Jaeger 2011). ICT developments have led to increased educational accessibility (Almeida et al. 2020; Della Líbera and Jurberg 2017), a greater sense of well-being (Duplaga and Szulc 2019), and more opportunities to obtain jobs, (Cahyadi and Setiawan 2020) and building relationships (Darragh et al. 2017) with other disabled individuals. Moreover, the media infrastructure has facilitated their personal and group interests, progressively and comprehensively increasing their prosperity and accessibility. Individuals can control how and when they disclose information about their disabilities (Furr, Carreiro, and McArthur 2016) and use the Internet to discuss issues in managing their impairments and rights (Qu and Watson 2019).

The development of an information society is perceived as an opportunity for greater inclusion of individuals with disabilities in social activities; however, it can be another source of frustration and exclusion. The mass media that existed before the proliferation of the Internet and smartphones did not adequately accommodate individual differences among users as the mass media used a uniform technology. Typically, it was presumed that individuals with visual impairments could only listen to the radio.

Therefore, media activities that heavily relied on visual engagement, such as watching television and films, were deemed entirely inaccessible to them. This occurred due to traditional stereotypes of visual impairment (Adam 2018), the requirement of assistive technologies and services, such as audio descriptions (Walczak 2017), and the belief that objectively generated descriptions could never be equivalent to the experience of sighted persons (Udo and Fels 2009). Even as technology has rapidly evolved, individuals with disabilities continue to have limited access to Internet-based social media, which reflects their experiences of social exclusion and devaluation (Batey and Waine 2015). Various studies have demonstrated that individuals with disabilities experience a significant digital divide (Duplaga 2017), meaning that gaps between those with disabilities and those who are able-bodied are persistent. Individuals with disabilities have acquired knowledge and have established interpersonal relationships on the Internet. However, these are challenging to effectively leverage as resources that can break through the barriers that disabled communities encounter in their offline lives (Bates and Davis 2004; Gregorius 2016). Fear of technology, dependence on digital devices, and a lack of knowhow hinder the digital media use of these communities, and increasingly complex websites and applications create additional barriers.

Regarding the use of digital media by individuals with visual impairments, existing research often treats digital media as a monolithic entity, exploring how it either provides visually individuals with, or deprives them of, access to information, and how it constructs either positive or negative representations of individuals with visual impairments (MacLeod et al. 2017; Morris et al. 2016; Voykinska et al. 2016; Wu and Adamic 2014). However, there is limited research on the aspect of visibility, addressing questions such as what individuals with visual impairments can or cannot ‘see’, the reasons behind this, and the efforts they undertake to ensure visibility or being seen.

Moreover, several studies have focused on adolescents with visual impairments using smartphone photography—taking photos to edit and share them on visual social media, such as Instagram and Snapchat (Bennett et al. 2018), and on a community of video bloggers (vloggers) with visual impairments who actively produce multiple video content categories on YouTube (Seo and Jung 2021) and SELFIE Helper, a mobile app for individuals with visual impairments to take and manage selfies (Auer and Tsiatsos 2022).

Recent research on digital disabilities has increasingly focused on mobile applications that are closely related to users’ visual experiences. However, most of these studies primarily approached the topic from an engineering perspective, concentrating on aspects, such as the reasons for using these applications, their objectives, and how to improve their accessibility. Thus, these studies often overlook a critical premise: the exclusion of individuals with visual impairments from the visual culture. Namely, substituting the participants of these studies with individuals with different types of disabilities would not result in significant variation in the findings. In visual culture, as the contradiction between the importance of sight and the inability to see becomes increasingly prominent, competition for visibility between sighted individuals and those with visual impairments intensifies. Consequently, researchers should explore what individuals with visual impairments can or cannot see at different levels; how this relates to visual impairment itself or to technological design; and what cultural traditions or social norms underlie these phenomena.

This study aimed to explore the specific practices of individuals with visual impairments who use short video applications in the context of visual culture and social structures that shape these practices and their effects on social inclusion. Specifically, this study sought to address the following research questions: (1) What characteristics can be attributed to the short video consuming practices of individuals with visual impairments in a highly mediated and ocularcentric society? (2) Within the short video watching practices of individuals with visual impairments, what opportunities and obstacles do digital technologies present? (3) Why do individuals with visual impairment engage in viewing short videos? By addressing these questions, this study explored how the short video practices of individuals with visual impairments challenge ocularcentrism and the potential for media technology equity.

Method

This study was initiated by an observation that it is often considered more appropriate for individuals with visual impairments to use media devices that prioritise auditory input. However, with ocularcentric media becoming increasingly prevalent in the modern digital society, individuals with visual impairments are now actively or passively watching short videos for entertainment, work, and socialisation purposes. This phenomenon highlights the critical role of digital media in promoting the inclusion of individuals with disabilities and necessitates further investigation. Thus, we adopted a constructivist grounded theory to explore the use of short video applications by individuals with visual impairments. The following sections outline the research design and the methods for data collection and analysis.

Research design

Advocates of constructivist grounded theory maintain an open and critical perspective when engaging with the existing literature, data, and analysis (Charmaz and Thornberg 2021). Constructivist grounded theorists assume that researchers cannot separate themselves from their preconceptions; instead, they co-construct meanings with their participants by listening to their feelings and experiences through dialectical processes and then rendering these collected empirical realities (Charmaz 2006). By adopting this approach, we aimed to explore the use of short video applications by individuals with visual impairments to facilitate comparisons between cases that exhibited both similarities and heterogeneities (Oktay 2012). Our study aimed to gain a comprehensive and in-depth understanding of the experiences of individuals with visual impairments in a society saturated with visual culture. This study was approved by the relevant university Human Research Ethics Committee.

We adopted an ethnographic research methodology to implement the abovementioned theoretical approach. Specifically, we observed the engagement of individuals with visual impairments with short video content through on-site observations and participation in online communities. We conducted in-depth interviews to further our understanding of the topic. During fieldwork, we interviewed 10 individuals with congenital or acquired visual impairment. With their consent, their interviews were audiorecorded, and the transcribed interviews were translated into English.

NVIVO12 software was used to organise, code, and analyse the data to ensure rigour and transparency in the data analysis. The collected data were continuously analysed through online communities until theoretical saturation was reached. According to Hayhoe (2012), grounded theory-based disability research is significant because it can investigate personal influences and analyse the development of experiential knowledge beyond the details presented in texts. Considering the participants’ limited educational level, we were greatly involved in theoretical coding and supplemented our observations with online and in-person fieldwork notes.

Data collection

We combined field visits with online observations for the data collection. Ethnographic observations were conducted at various field sites in Beijing, Xi’an, and Chongqing from August to December 2022. These included two blind massage parlours and two moviescreening events specifically organised for individuals with visual impairments, totalling eight sessions with each session lasting 2–2.5 hours. Opting for massage parlours aided in a concentrated observation of the video consumption practices of the visually impaired, as massage therapy constitutes a significant employment avenue for these individuals in China. Additionally, movie-screening events specifically organised for individuals with visual impairments proved to be an efficient strategy for data collection as it provided participants with diverse range of occupations. Furthermore, their behaviour in watching films provided deeper insights into the methods utilised by visually impaired individuals to process visual information. During these sessions, we observed interactions between individuals with visual impairments and short video applications, including browsing, liking, commenting, and sharing with peers. Concurrently, with the participants’ consent, we joined four Douyin group chats and performed continuous digital ethnographic observations for a period of four hours per week. To avoid interfering with the individuals’ engagement with short videos, we did not actively communicate with them during our online and on-site observations; instead, we compiled our observations into fieldwork notes.

To deeply analyse the experiences of individuals with visual impairments using short video applications and their interactions with the visual culture, we conducted semistructured interviews. We recruited 10 participants with either congenital or acquired visual impairment through purposive and snowball sampling from our research sites. We considered demographic characteristics, such as gender, age, education level, and occupation, to provide a comprehensive understanding of the effects of visual impairment on the participants’ consumption of short videos. The participants were aged between 18 to 50 years old, with four individuals aged between 20 and 30 years (eight mens and two women). The participants’ occupations included masseur, entrepreneur, student, Internet operator, and unemployed. According to the Practical Assessment Criteria for Persons with Disabilities in China (2006), visual disabilities are classified into blindness or low vision. After this criterion was applied, seven of the ten respondents in this study were classified as having low vision. This reflects a lack of practical concern for individuals with visual impairments in contemporary Chinese society, where they are often equated with the blind. Our study relied mainly on participants’ self-reports regarding the level of their visual impairment, as most participants’ visual impairments were not professionally diagnosed. Examples of self-reported statements include the following:

I could see the words clearly when I was younger, but now I can see shapes clearly when they are very close to me.

I have some light perception, and I can see some outlines blurred.

It’s blurry, like high myopia.

I can see a little clearer when the light is strong.

I can see larger images but cannot clearly see faces.

I can’t see any words clearly, and my sensitivity to light is decreasing.

I can’t see anything at all.

My vision is less than 0.1, and I can only see the outline. I rely on my hearing when using a smartphone.

I have colour blindness and myopia with astigmatism. Apart from having difficulty recognising colour cards, there is no other subjective impact on me after using corrective glasses.

We conducted semi-structured interviews to collect data. Owing to the participants’ different locations, two interviews were conducted face-to-face; seven interviews were conducted online; and one was conducted through written responses.

Each interview lasted approximately 60 minutes and the entire interview was recorded with the participants’ consent according to the principles of informed consent, voluntary participation, and confidentiality. The interviews were transcribed and translated into English. Finally, we continuously compared and analysed the collected data in the Douyin group chats until theoretical saturation was achieved.

Data analysis

This study employed procedural grounding theory to guide the coding process of raw data in three stages. To ensure the reliability of the study, two researchers independently coded the interview data. Then, the results were analysed and compared with NVIVO12 to derive the final coding results.

The first stage of data analysis involved open coding. Researchers read the collected raw data word-by-word around the research questions and performed initial category extractions. This stage was divided into labelling and conceptualisation, depending on the level of abstraction. After the data were analysed and summarised, 20 concepts were identified. Along with the concepts of visual, auditory, and tactile experiences, the following concepts were found: memory storage; operational habits; manual assistance; richness of information; ease of operation; level of hardware support; efficiency of image-text conversion; cross-platform operation; accessibility modes; acceptance and resistance to algorithms; integration into communities; self-image expression; educational learning; work necessities; leisure activities; awareness of accessibility; and universal acceptance. The second stage of data analysis involved axial coding, in which highly relevant concepts related to research questions were re-extracted. This stage focused on generating seven valuable categories to interpret conclusions: sensory adaptation, experiential transplantation, infrastructure, technical negotiation, identity construction, identity needs, and disability rights.

During the third stage of selective coding, this study identified three core categories—diffuse compensation of vision, reconstruction of technical prostheses, and participatory social inclusion in mainstream society—and systematically analysed relationships between the core and main categories. We suggest that the concepts of sensory adaptation and experiential transplantation encapsulate specific processes that reflect the visual experiences generated by individuals with visual impairments when they use short video applications. These processes address the question of why certain elements can or cannot be observed. The concepts of infrastructure and technical negotiation represent concrete actions taken by users with visual impairments while engaging with short video applications, clarifying how negotiations can shift experiences from ‘not being able to see’ to ‘being able to see’. Furthermore, the concepts of identity construction, identity needs, and disability rights demonstrate the goals and challenges faced by individuals with visual impairments in the pursuit of social inclusion through short video applications. These concepts reflect broader issues of discrimination stemming from the perspective of ocularcentrism.

To test the theoretical saturation, the study interviewed two visually impaired individuals and repeated the above process, incorporating observations of their behaviour when using short video applications. The results showed that the sample coding generated categories that could all be classified into the previously coded results, and no new explanatory elements or related relationships were produced. Therefore, the study concluded that theoretical saturation had been reached.

Findings

The results showed that sample coding generated categories that could be classified into previously coded results, and no new explanatory elements or related relationships were produced. Therefore, the study concluded that theoretical saturation was reached. This study divides the analysis of relationships among the three core categories into two parts. First, we explain the fundamental nature of ‘diffuse compensation of vision’ in the practice of short video consuming by individuals with visual impairments, especially how the ‘reconstruction of technical prostheses’ presents both opportunities and challenges for this group. Second, we explore reasons and motivations behind the use of short videos by individuals with visual impairments and examine the difficulties they experience in achieving inclusion within mainstream visual culture and society and the available space for disability rights.

Foundation of inclusion: vision or technology?

Disability-related research has long focused on the contradiction of technology because it simultaneously offers opportunities for individuals with disabilities and limits their inclusion (Roulstone 2016). Different combinations of disabilities and technology types produce different results. Specifically, regarding visual technology, Crary (1990) posited that disciplinary technologies have emerged to recode and regimen eye activities to increase productivity and reduce distractions. This perspective highlights how the visual experiences of individuals with visual impairments are also subject to media practices that reconstruct their pathways for social inclusion. With their intersection, do upgraded technology and visual impairments support or hinder each other or is it a zero-sum game?

Diffuse compensation of vision

Despite the long-standing bias that individuals with visual impairments rely solely on hearing, our findings indicate that vision is not merely an ‘absent feature’, but rather a complex interaction for individuals with visual impairments, who have diverse needs. The social construction of visuality is pervasive; despite being conveyed through sounds or other forms, a great deal of information is still permeated by the logic of visual production. Thus, even many congenitally blind individuals often use words highly related to vision, such as ‘see’, ‘watch’, and ‘view’. This phenomenon is known as the diffuse compensation of vision and is manifested in three aspects.

First, at the physical level, the visual sense remains dominant. Li’s (2013) survey of individuals with visual impairments found that more participants obtained their news information from the television than from the radio. Although compensatory effects using other senses still enrich perception, the authority of vision has been universally established. The participants frequently mentioned their experiences using their eyes, even when they were browsing short videos through screen readers. For example, participant Lisa spent most of her time browsing short videos through screen readers, but ‘sometimes, when some influencers (Wang Hong) were live streaming, I could not help but watch for a while’. Participant Jack also said, ‘When my eyes could still see relatively well, I was still used to using my eyes to look, and listening was too slow’. Previously, the ‘one glance ten lines’ visual-search process was a linear operational process, even if the speech rate was faster. Though there are more sound effects are more prosperous, these effects cannot truly satisfy the desire of individuals with visual impairments to ‘see’. Therefore, individuals who have impaired vision, but have some vision remaining, continue to use their eyes even if they must constantly overcome physical discomfort. However, this socially constructed state of visual dominance conflicts with the functional defects of the organ itself. Individuals with visual impairments need to expend more physical energy to obtain the same information as individuals with full vision, creating a new form of violence in the diffusion of vision.

Second, at the symbolic level, visual information is rich. Visual abilities of individuals with visual impairments cannot be simply described as ‘able to see’ or ‘unable to see’ because there are different degrees and dimensions in visual impairment, leading to highly diverse visual information needs. These requirements include screen brightness, font size, light sensitivity, colour intensity, and movement speed; visually impaired individuals may have unique needs when they swipe short videos. Visual elements still play an essential role and can potentially convey information in the media practices of individuals with visual impairments. For example, even though Lewis’s vision was already fragile, visual elements still played an essential role in the way he selected videos to watch. ‘The main thing that attracts me is the movement’, he said, adding, ‘some people are splitting and jumping on (short videos), which attracts me’. Unfortunately, most of the visual information in short videos is presented in a uniform technical format. Attempts to use dark modes or modes designed to enlarge fonts or increase volume are not compatible with existing screen-reading systems.

Third, at the experiential level, visual storytelling is constantly being diffused. Visual information often offers a unique narrative structure, which includes modalities, grammar, and meaning (Cohn 2020). We found that individuals with visual impairments can profoundly understand the visual narrative of short videos, either through the memory and habituation of visual media practices or through the constant supplementation and improvement of interactions with social groups, such as family, friends, and colleagues, ultimately adapting to the production and consumption logic of short videos. When testing comprehensive screen-reading tools for individuals with visual impairments, which included specific visual descriptions, Choi et al. (2019) found participants could establish mental images of visual information. Thus, some individuals who acquired visual impairments during the television era transferred their viewing habits to the era of short videos and were more receptive to film and television works with richer visual information. Meanwhile, some individuals who had not experienced visual impairment retain their touchscreen habits using smartphones and even construct new visual experiences based on their minimal visual abilities. For example, respondent John said, ‘some interface layouts are imprinted in my mind. As long as the software doesn’t undergo major changes, based on some subconscious things, I’m clear about it’. Another participant, Jack, said, ‘This feeling is strange. Even though it’s blurry, based on my previous experience, I might know what it looks like’.

Based on the analysis of the body, symbols, and experience, the issue of visuality for individuals with visual impairments continues to be highlighted. Although the ‘eye’ affects the ‘invisible’ with physiological, temporal and spatial limitations, and, more importantly, visuality (Zeng 2009), the ‘eye’ has been embedded in individuals’ imagination of knowledge/intelligence, emotions/attitudes, social/things relations, shapes/time, and other aspects, and also reshapes us through the relationship between the medium and the body.

Reconstruction of technical prostheses

Based on Stiegler’s (1998) reconceptualization of prostheses in which he argued that external technological tools were parts of a human body rather than a substitutive supplement, this study aimed to explore how short video platforms embed themselves into the bodies of individuals with visual impairments and become extensions of their bodies. To fully participate in a media society constructed of short videos, individuals with visual impairments need to understand the infrastructure (hardware, software, text, etc.) of short videos and explore consciousness and behaviour in a more diverse way.

Short videos can regenerate technological prostheses in such a way that they become a means by which individuals with visual impairments can be included in society. More individuals with visual impairments are absorbed into this technological system through the continuous improvement of hardware support, ease of operation, information richness, and image–text conversion efficiency. Namely, these individuals can incorporate these features into their bodies. After several years of development, short video platforms have accumulated considerable amounts of information in their databases. With the constant upgrades to recommendation algorithms, most videos with distinctive themes and full sound (including commentary, music, and sound effects) can be accurately pushed to individuals with visual impairments. A participant, Jack, said, ‘Douyin is convenient for me, and the content is also comprehensive. After a video emerges, you can immediately know what type of content you want’. Short videos are often criticised for lacking depth, but this has become an important reason that they bridge the digital divide. Despite being visually dominant, short videos can penetrate all aspects of the lives of individuals with visual impairments because these videos rely more on collaborative filtering and other methods to personalise recommendations, thus meeting users’ needs for the acquisition of information flow. This approach rarely requires users to perform complex operations, at least in this regard, making it friendly for individuals with visual impairments. For example, Alex said, ‘I spend an average of at least four hours a day on Douyin, even more sometimes. It is essential software in my life’.

Second, short videos constantly activate the awareness of individuals with visual impairments about self-efficacy as part of their bodies. Specifically, their understanding of technological updates can be categorised into three types. The first type is technology of which they are aware and proficient. For instance, many individuals with visual impairments deliberately increase text-to-speech speed to improve efficiency, sacrificing some sound features. The second type is technology with which they are less familiar. Owing to the lack of systematic learning and training, many potentially helpful and accessible designs cannot effectively enter their lives. For example, the ‘Magnifier’ feature developed in iOS systems, which can dynamically recognise and describe images and videos, was rarely used by the individuals with visual impairments we interviewed.

The third type is technology of which they are aware but which they cannot use at that time. For example, many individuals with visual impairments are aware of animation effects, beauty tools, and emojis in short videos and express a desire to use them, but they are restricted by their complexity or poor user experience. Further, individuals with visual impairments perceive algorithm recommendations. They can strategically express their identities as individuals with disabilities online to adjust to the influence of the algorithm. In our research, most individuals with visual impairments also have a clear perception of algorithms. For instance, David stated, ‘I feel as if I were “monitored.” But, currently, all internet products conduct user profiling’. Individuals with visual impairments can reverse the effect through proactive searching, using multiple platforms, and continuous liking, thus obtaining more accurate recommendations for the information they want. For example, Jack stated, ‘When I don’t understand something, I can search on Douyin, and if I think it can help me, I just save it for repeated viewing in the future’.

Furthermore, short videos benefit individuals with visual impairments by forming a set of accessible paradigms through repeated actions. Through a critical analysis of the screen reader function that almost every individual with visual impairment uses, we found that VoiceOver on IOS devices or TalkBack on Android devices could assist them in proficiently using most smartphone software. Although most individuals with visual impairments can quickly adapt to the interaction mode between fingers and screens (Smaradottir, Håland, and Martinez 2018), the compatibility between screen readers and short video applications is insufficient, resulting in the adoption of more diverse operational strategies when individuals with visual impairments browse short videos. The main reasons are as follows. First, the information output of the screen reader mainly relies on sound, which conflicts with the audio-visual system of short videos. For instance, Alex stated, ‘The video, how do I say it, on Douyin still can’t read it out. It cannot read some homophonic characters or other pages either’. Second, the screen reader, sometimes, becomes a burden on information reception. If the individual only wants to browse short videos, they tend to turn off the screen reader and switch to swiping with their fingers, similar to sighted individuals. However, they must reenable the assistive function if they need to participate in interactions, such as liking and commenting. Moreover, many individuals with visual impairments with acquired-vision loss do not suddenly become blind but experience varying degrees of degeneration over several days to decades. For them, the most suitable way to watch short videos is not to listen directly to the screen reader but to gradually transition by adjusting the font size, brightness, and darkness.

Besides adjusting for conflicts between modes, individuals with visual impairments frequently engage in cross-platform operations. For example, owing to differences in technology, content, and social ecology among different short video applications, individuals with visual impairments make more diverse choices based on the accuracy of algorithm recommendations, the efficiency of the target audience’s acceptance, and the difficulty operating production tools. Therefore, the accessible practices of individuals with visual impairments, which are full of obstacles, form a memory of the stereotype described by Stiegler (1998), which preserves traces, experiences, and lessons. Their consciousness and demands have also become externalised technology.

Purpose of inclusion: self or society?

In contemporary society, a vast amount of information is presented visually, making access to this information challenging for individuals with visual impairments. As vision loss occurs, individuals may become excluded from society. However, for many individuals, such as Jack, short videos are more than just a means of passing time out of boredom; it is an essential component of remaining connected to the world. Jack explained, ‘Listening to [visual content] can help me understand what’s happening now, like news or something else, so I will not be out of touch with this era’. This statement highlights the critical role played by inclusion in the life of individuals with visual impairments. Because many individuals with visual impairments have limited access to education in China, short videos and knowledge-based content can serve as powerful tools for their empowerment and integration when they find other information resources difficult to access.

Self: falling into the mainstream path

The 50th Statistical Report on China’s Internet Development (2022) reports that the user base for short video applications has reached 960 million, with 54.8% of deep users watching or listening to related content daily, making it the leading audio-visual Internet application. Despite the traditional belief that individuals with visual impairments may not benefit from video applications because of their integrated sound information, these applications have evolved into comprehensive platforms that integrate search engines, social media, streaming media, and shopping websites, surpassing other platforms, such as Google, Twitter, and Weibo.

What drives individuals with visual impairments to use short video applications? Through a coding analysis of all materials, we found that their main motivations included personal demand for entertainment, learning, work, and shopping. Regarding entertainment, no significant differences in interest were observed between individuals with visual impairments and sighted individuals; both groups consume a wide variety of content. For example, John, who experienced deteriorating vision, used Douyin for entertainment and emotional support: ‘Sometimes when I’m really in a bad mood, I use Douyin to watch some funny videos’. Short videos have also become a primary source of news, as Ian explains, ‘I usually watch content related to work, and sometimes I also follow international news. For example, I would actively search for the recent hot topic of the Ukraine war’.

In terms of learning, although some individuals acquire a diverse range of job skills before they experience vision loss, many individuals with visual impairments choose to become masseurs because of societal perceptions and prejudices regarding their abilities (Xiong and Liu 2023). Despite challenges, masseurs with visual impairments often work long hours, typically from early morning to late evening. To facilitate efficient learning in a limited amount of time, Khlaif and Salha (2021) noted that short video platforms offer units of instruction that are often less than 60 s. Therefore, short videos have become a crucial resource for learning and connecting online and offline videos. For instance, John highlights that for more in-depth theories, such as understanding the meridians and acupoints of the human body, learning from Douyin is preferable because many offline teachers know only the names of a few conventional muscles.

At the work level, masseurs with visual impairments often operate within confined spaces, such as their homes or blind massage parlours. However, they have begun to engage with existing cultural narratives and public discourses (Ellcessor and Kirkpatrick 2017). For example, Lewis began using Douyin after being asked by the manager to post videos, which helped him earn 50 yuan every 500 views. Currently, his account contains tens of thousands of views. Short videos have become a crucial channel for attracting customers in a consumer society, and the capital-accumulation logic associated with them can embed and train more individuals. However, the participation of individuals with visual impairments is often mixed, with the individuals being swept along and forced to participate. While seeking self-realisation and satisfaction through short videos, they ultimately desire deeper social recognition.

Society: detached from the mainstream network

According to Bühler and Pelka (2014), digital media can empower individuals with disabilities by giving them access to social media, peer support, and public space. This analytical framework allowed us to reconsider the importance of short video platforms as a medium of inclusion. We observed that individuals with visual impairments engaged in basic media behaviours, such as watching, liking, and commenting, on short video platforms. They also became professional creators, built communication communities, participated in public activities, and shared their lives and knowledge on Douyin. Some masseurs with visual impairments have become bloggers with many followers, and the interaction data for each video they publish are quite impressive. Lisa, whom we have been continuously observing, originally shared only a few seconds of life footage without editing, such as music or lyrics, on her Douyin account in August 2022. However, by December 2022, she had developed the ability to produce themed videos with animations, voiceovers, subtitles, and music, exhibiting a certain level of professionalism. She also established her fan group on Douyin, and after reaching a particular level, expanded her fan group to WeChat. Some of the participants in this study used the social attributes of short video platforms to help them ‘see’, letting viewers serve as their eyes. For example, Richard, who had acquired visual impairment, posted videos shot from different angles and environments, used equipment, and requested feedback from followers, repeatedly adjusting frames until they confirmed that they were watching a short video with qualified audio-visual expression (including appropriate characters, scenery, focus, exposure, and colour).

When individuals with visual impairments reveal their identities on short video platforms, those who participate in interaction with them are mostly other individuals with visual impairments. However, this relationship is weak and limited. On the one hand, fundamental social relationships are only a digital reorganisation of offline social circles.

Tony, an accessible design expert, says, ‘We are more likely to see other individuals with visual impairments because their friendship will be imported, and the circle of contact will be within this circle’. However, weak relationships established through social media platforms may provide diverse information. Moreover, interpersonal trust and emotional support are relatively weak (Xiong and Liu 2023), failing to provide support to individuals with visual impairments and generating adverse effects. We joined the fan groups of two visually impaired masseurs; one group focused on sharing job-seeking experiences while the other was designed to chat and make friends. Both groups had rapid declines in activity within a month. The latter was eventually disbanded because some members frequently sent voice messages and emojis that caused disputes.

Furthermore, under the dominance of the visual culture, factors such as appearance and body shape, have become necessary standards for measuring people. Individuals with visual impairments do not escape the logic for mainstream aesthetic evaluation because of their visual disadvantages. They also play the roles of the subject and the object of gaze within this framework. For example, Lisa often watches live broadcasts of hosts and says, ‘I can only see a rough outline of their faces because of my impairment, but if they dress nicely, I will follow them’. However, she also becomes the object of the gaze: ‘I lost several followers when I posted a video of myself’. Gaze theory holds that the desiring subject actively takes possession of a passive object. In a world where order, such as that between men and women or able-bodied and disabled individuals, those in the group with less power are more likely to be subject to bias and lose opportunities to speak out. Individuals with visual impairments also find themselves excluded from mainstream society through active sacrificial self-expression.

Overall, the integration of individuals with visual impairments into society is not smooth. While they have developed a unique and effective set of practices through positive physical adaptation and technological negotiation, they are still confined to small groups and are only seen as tools for discourse, floating outside the daily lives of the general public. Regardless of its limitations or external pressures, the process of social inclusion has contradictions.

Conclusion and discussions

Re-examining ocularcentrism

Why were individuals with visual impairments chosen as research participants to examine ocularcentrism? From a broader perspective, at least 2.2 billion individuals worldwide have visual impairments (‘Blindness and Vision Impairment’ 2023). The eye is not just a sensory organ; it represents the mainstream social context of offering narratives, interactions, and power structure. It ubiquitously influences individuals’ desires and boundaries for autonomous exploration. Visuality has long been considered synonymous with rationality and truth. When the eye becomes sight-impaired, other co-opted organs exhibit the characteristics of automatic replication in visual culture, and symbolic violence not only appears in visual representations but also permeates other sensory and technological systems. The visual culture supported by Internet technology is not just about making people ‘able to see’ or ‘unable to be seen’, but about forming a rigorous technical system of ‘visuality’, which excludes different groups, including individuals with visual impairments, through a merging illusion. Ocularcentrism is continuously strengthened through the diffusion of vision.

In this study, we present three main features of the digital-media era: seizing and regulating people’s time through dominant media, hiding hegemony through the negotiation of body and technology, and consolidating central status by reconstructing connotations and metaphors. In other words, individuals with visual impairments are excluded from the ubiquitous visual media. From the perspective of organs, a singular evaluation of the eyes obscures richer perceptions of the world. Under standardised disciplines, we are not only attracted by our gaze, but we can also control visual perception and its relationship with other senses more freely and consider the relationship between humans and technology at multiple levels. From a knowledge perspective, the exclusive worship of visual rationality leads us away from reason in the frenzy of desire. Therefore, the social inclusion of individuals with visual impairments is meant not simply to emphasise how they should adapt to the current visual culture or what efforts should be made for them, but to discuss how society should make holistic adjustments to be compatible for individuals with visual impairments and reflect on the deprivation of human subjectivity caused by this trend of media visualisation. By expanding our understanding of visual and multisensory knowledge and deconstructing the separation between visual and tactile perceptions, we can challenge ocularcentrism and the way in which blind individuals are seen as knowers and political actors in the world of others. Therefore, this visual disability may contain some possibility of resistance.

Possibility for media technology equity

Our study found that individuals with visual impairments have two main expectations regarding disability rights. First, they hoped for equal access to information without barriers. Second, they hoped that their group would gain social recognition without discrimination.

Based on the findings of the study, the following recommendations are offered. First, digital accessibility must be fully integrated into the design and upgrade processes of platforms (Ellis 2016). More channels should be provided for individuals with visual impairments to participate, voice their opinions, and become technology designers. Although an increasing number of Internet companies have recruited individuals with visual impairments for accessibility development, it is still an individualised path and lacks expressions of group consciousness. Therefore, besides formal complaints and suggestion channels, two more approaches should be considered. First, based on privacy protection, platforms should use new technologies to establish behaviour databases, dynamically understand the limitations of individuals with visual impairments in using short videos through digital trajectory analysis, and provide more precise feedback. Second, individuals with visual impairments should be empowered to produce more visual content through the low operational threshold of short videos and analyse the videos they make. For example, Tony said, ‘Many individuals may be unable to write or speak, but they can film. Thus, inclusivity facilitated by short videos is broader, which also applies to communities with visual impairments’.

From another perspective, the target of media-technology equity should not be individuals with disabilities but the general public. The fundamental issue of disability rights is that new technology should enable society to reach a consensus on people with disabilities and their demands. In the case of individuals with visual impairments and their short videos, the more significant factor is that an active media platform can generate a set of technological democratic practices, in which universal infrastructure and differentiated usage demands can coexist; the expression of identity by individuals with visual impairments and the acceptance of information by the public will no longer be locked in conflict, and the visual-dominated media can be more diverse through negotiation with different body organs. For technology designers, the goal should not be to highlight the ‘disability’ but to treat it equally with all other ‘differences’. As a participant, Tony, said, ‘accessibility does not simply mean that individuals with disabilities can use digital technology. Its broader meaning is to enable anyone to obtain information they need in different scenarios’.

However, this study has some limitations as an exploratory investigation. The internal mechanisms of specific processes have yet to be fully explored: how social capital develops in visual relationships; how algorithmic perception impacts the social integration of individuals with visual impairments; and how consensus is formed regarding one’s body through short videos. Further, the study was limited by its exclusive focus on short videos without an in-depth examination of other visual forms, such as streaming videos and live broadcasts, and their potential differences.

Acknowledgements

The authors highly appreciate those who participated in the interviews. The authors would like to thank Professor Graeme Kirkpatrick for his insightful and helpful comments on earlier drafts of this article.

Disclosure statement

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Additional information

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.