Introduction

Banter, humorous yet potentially aggressive interactions for mutual amusement and bonding (Brown et al., 1987), is common among adolescents and emerging adults (Whittle et al., 2019). Such interactions are commonly interpreted as having positive intentions (Budden et al., 2022) of overcoming perceived social distance and increasing group solidarity (Winkler Reid, 2015). Yet, bantering (i.e., the act of engaging in banter) should be taken with precautions as the potential ambiguity and aggressiveness might lead to psychosocial harm or other unwanted consequences, regardless of its actual intention (Hickey & Roderick, 2022; Plester et al., 2022). Also, not all banter has positive intentions. There is maliciously intended banter that masks harmful behaviours such as bullying (Betts & Spenser, 2017) as using humour significantly increases the acceptability of verbal aggression (Gauducheau, 2019). That is, people may use the term banter as an excuse for their unacceptable aggressive behaviours (Buglass et al., 2021). Perhaps due to that reason, it is not always easy for the receiver/target to interpret the meaning behind banter-type interactions. Especially when banter takes place online, the lack of contextual, non-verbal cues may create ambiguity such that misinterpretations and escalations could occur more commonly online than face-to-face (Buglass et al., 2021). A way to overcome such shortcomings in online communication might be the utilisation of digital visual cues such as emojis, emoticons, hashtags, etc. that help highlight the humorous, light-hearted intentions (Tang & Hew, 2019). This review aims to explore and synthesise the literature regarding the use of digital visual cues during banter and how they shape online bantering among adolescents and emerging adults. For the purpose of the review, we followed Arnett’s (2010) descriptors with 13- to 18-year-olds considered adolescents and 19- to 25-year-olds considered emerging adults. In doing that, the overall intent of this review is to explore the opportunities to promote friendly online banter and contribute to minimising banter-related ambiguity and escalations with the help of digital visual cues.

By acting as non-verbal cues in the digital environment, the digital visual cues seem to make virtual interactions less cold and impersonal (Aguert et al., 2016). Studies suggested that using emojis and emoticons along with text helps the receiver to accurately grasp the emotional state of the sender as well as the intention and meaning of the text (Walther & D’addario, 2001). Nowadays, various types of digital cues seem to be widely used among younger generations, such as GIFs, memes, and stickers (Taecharungroj & Nueangjamnong, 2015). From a broader perspective, photos/selfies, initialisms (e.g., lol, jk, etc.), hashtags, and short video content such as Vine are other forms of digital visual cues that adolescents and emerging adults may utilise in creatively communicating their intentions and feelings in the virtual environment. Perhaps due to these digital cues, online bantering has become commonplace. Research has found that online bantering becomes more prevalent among adolescents as they get older (Aguert et al., 2016). Adolescents and emerging adults even create online spaces (i.e., group chats) just for bantering (Abell et al., 2023).

However, although digital cues may help facilitate digital banter by conveying playful and light-hearted intentions, they could be ambiguous and even harmful at times. For instance, emojis are likely to increase the ambiguity in online interactions when there is a lack of textual information available to supplement decoding them (Derks et al., 2008). The ambiguousness of emojis was highlighted in a recent study in that single emoji was used and interpreted in 40 to 78 different ways (Shardlow et al., 2022). This high variation was said to be indicative of the widespread use of emojis without access or reference to a universal dictionary of meaning. Furthermore, a case study showed the potential harms ambiguous memes (i.e., a combination of viral images/videos and catchphrases) may cause to young people’s wellbeing when circulated online in the framing of a joke (Vickery, 2014). In that study, the researcher illustrated how an online meme called ‘Confession Bear’ unexpectedly diverted from its original intention of being humorous with users anonymously confessing their funny secrets and turned into a serious issue as people started sharing their secrets regarding abuse, rape, addiction, etc. using the humorous meme format. Even when the digital cues are not as clearly problematic as the examples given in Vickery’s study, people tend to use, perceive, and interpret digital cues differently depending on various contextual and individual-level factors (Koch et al., 2022).

Taken together, to understand young people’s experiences of online bantering more comprehensively and help them prevent/overcome harmful interactions, particular attention should be given to the roles digital cues play in banter. Therefore, this review aimed to summarise and synthesise literature evidence on adolescents’ and emerging adults’ use of digital visual cues in producing digital banter, their experiences, views, and interpretations, and factors that might influence these. This review focused on adolescents and emerging adults given that previous studies showed how during these periods both social media use (Shannon et al., 2022) and banter production expanded significantly (Mills, 2018).

In addition to banter, this review included studies on online humour in general, as the term ‘banter’ is more tied to British culture (Buglass et al., 2021) and there are few studies available that have used this term specifically. From the linguistic perspective, humour and banter are related constructs, in which humour is considered more of an umbrella term that encapsulates banter as a type of humour (Dynel, 2009). Moreover, this review included a broad range of digital cues that adolescents and emerging adults might be using, as opposed to solely focusing on particular ones, and thus provides a more comprehensive, multifaceted understanding of digital cues as banter.

The questions addressed in this review are:

  1. 1.

    What are adolescents’ and emerging adults’ use, interpretations, views, and experiences of digital visual cues as digital banter, and what factors contribute to these?

  2. 2.

    What are adolescents’ and emerging adults’ use, interpretations, views, and experiences of digital visual cues as digital humour, and what factors contribute to these?

Methodology

Protocol registration

The review protocol was pre-registered on Open Science Framework in April 2023 (https://osf.io/sfhqt/?view_only=c6c95c6aed724b038e5bd037f34620db).

Search strategy and study selection

As outlined in the review protocol, the database search strategy was limited to peer-reviewed, in-press and published articles and reviews, that were written in English and published since 01/01/1982 when the first set of emoticons was introduced (Raymond, 1996). Keywords were identified using the adapted PICO framework for qualitative reviews (Stern et al., 2014). After that, the synonyms/alternatives were developed by examining the key studies, pilot searching on the databases, and through a series of review team discussions (see Table 1 in Online Resource). Inclusion criteria were developed in accordance (see Table 2 in Online Resource) and studies that did not meet for all the inclusion criteria were excluded.

Quality appraisal

The qualitative studies checklist of the Critical Appraisal Skills Programme (CASP) (Critical Appraisal Skills Programme, 2018), a widely used appraisal tool that is specifically tailored for qualitative studies (Butler et al., 2016), was used for assessing all qualitative studies. Likewise, for mixed-method studies, the Mixed Methods Appraisal Tool (MMAT; (Hong et al., 2018) was used. For other types of studies, such as cross-sectional and experimental, the Joanna Briggs Institute’s (JBI) critical appraisal tools were used (Tufanaru et al., 2017). No study was excluded based on the quality appraisal, instead, the strengths and potential weaknesses of each included study were identified (see Table 3 in Online Resource), following Messina et al. (2017). Judgments on the papers were noted when appraising their quality in line with recommendations by Petticrew and Roberts (2006). The first author carried out the quality appraisal and the review team conducted a complementary quality appraisal. Any differences in the appraisal were resolved through discussion.

Data extraction

A data extraction sheet piloted in similar reviews (Messina et al., 2013, 2017) was utilised. The following information was extracted: study details (including author, year, country, aim of study, and study design), population and settings (including number of participants, gender, age, and recruitment process), methodology (including data collection method and data analysis), and relevant findings (key findings relevant to the review questions were extracted narratively). The first author extracted all information, and the outcomes of this process were checked and verified by the review team (see Table 3 in Online Resource).

Data analysis

The narrative data were analysed using the Reflexive Thematic Analysis, an established method for analysing qualitative datasets and generating patterned meanings/themes that address the research questions; particularly when the research is interested in the interpretations, views, and experiences of participants (Clarke & Braun, 2021). After familiarising with each study through the extensive data extraction and quality appraisal process, relevant data from all included studies were imported into a qualitative analysis software, NVivo, for coding. The inductive coding leaned towards the semantic level to capture the explicit meanings communicated by the original papers. Once codes were developed (and refined) and initial themes were generated by the first author, the review team held a series of reflexive discussions and made iterative changes/refinements to the themes, following Clarke and Braun (2021).

Results

Identification of studies

The search yielded 12,372 records after duplicated records were removed. The screening process against the inclusion criteria consisted of two stages and was conducted on Rayyan (Ouzzani et al., 2016), a systematic review tool. In the first stage, titles and abstracts of all identified studies were reviewed by the first author, and the second author screened around 20% of them to check for consistency in the approach. The studies that clearly did not meet the inclusion criteria were excluded, but those that were unclear were moved to the second stage, full-text screening, along with those that appeared to meet the criteria. The interrater agreement was 98.6% or very high (McHugh, 2012) and the disagreements were resolved through discussion. 94 papers were identified for the full-text review and the first and second authors independently reviewed all of them against the inclusion criteria. Both authors agreed to include 17 papers in this review. Figure 1 summarises the review process.

Fig. 1
figure 1

Study selection process

Full size image

Summary of included studies

All studies included in this review covered at least one type of digital cue in the context of banter or humour, although in most of the cases, it was not the main research interest of the papers. That is, the primary focus and aim of the included studies varied a lot, and therefore confirms that the use of digital cues in online banter has received much less research attention. The studies were conducted in the UK (n = 4), US (n = 4), China (n = 3), Australia (n = 1), Canada (n = 1), Norway (n = 1), Sweden (n = 1), Portugal (n = 1), and France (n = 1). Qualitative studies were most prevalent (n = 8), followed by mixed-methods studies (n = 5), and quantitative studies (n = 4). The majority of the included studies focused on emerging adults / mostly university student samples (n = 12), a few of them were focused on adolescents (n = 4), and one study included both samples. Most studies (n = 10) explored the text contextualisation effect of digital cues such as emojis, emoticons, hashtags, three dots (…), initialisms (lol, jk, etc.), and exclamation marks. Among them, emojis and emoticons were mostly investigated. The rest of the studies explored personal photos/videos (n = 4), online memes (n = 3), and Vine (a short video-based media type; n = 1) as banter cues. For more detailed information, please see Table 3 in Online Resource.

Study quality

The included studies were appraised for both their strengths and weaknesses. Overall, the studies demonstrated clear aims and appropriate methodologies to address them. Qualitative studies provided sufficient extracts to support their arguments, and most quantitative studies were complemented by qualitative data analysis to ensure that their interpretations reflected the participants’ perspectives. However, potential weaknesses included the use of convenience sampling with university students (mostly without reasonable justification for excluding adolescents), insufficient consideration of cultural effects on study outcomes (i.e., the unique culture of the country where the study was conducted (Schermer et al., 2019), and lack of information on the authors’ influence or relationship with the participants. Details on the study quality can be found in Table 3 in Online Resource.

Reflexive thematic analysis

Four distinct themes were generated, including (1) Digital cues are an essential part of digital banter, (2) Digital cues are ambiguous, (3) Harmful online banter containing digital cues, and (4) Personal and contextual effects. Selected extracts from the included papers are utilised as quotes to supplement the analysis.

Digital cues are an essential part of digital banter

The most prevalent patterned meaning observed across the included studies was that digital cues play an essential role in digital banter and humour as they contextualise the text content, highlight the light-hearted intention and therefore help mitigate the risks associated with misinterpretation (Aguert et al., 2016; Goodyear & Quennerstedt, 2020; Prada et al., 2018; Steer et al., 2020; Zhao, 2020). For example, based on a series of focus group discussions with UK adolescents, Steer et al. (2020) suggested that “by using emojis, adolescents are attempting to overcome the lack of social indicators and context [in the online environment] and reduce the likelihood of their banter interactions being misinterpreted” (p. 5). Putting laughing face emojis at the end of mock aggressive texts might be a common way of contextualising digital banter (Keane & Hammond, 2022; Zhao, 2020). One of Steer et al.’s participants said:

“Like, you can tell it’s a joke, like say, it says something funny but then like, you reply to them saying “you’re such an idiot” but you can put emojis on them, say you put a laughing face, you could just mean it as a joke but if you didn’t do anything erm, it could mean that you were, mean it in a bad way. (Year 7)” (p. 5).

In addition to signalling banter, digital cues were also utilised in expressing how well the banter was taken (Vandergriff, 2013; Zhao, 2020). For example, in Chinese hudui (banter), laughing and winking face emojis and emoticons at the end of the reciprocation signalled jocular abuse that both parties took as non-serious play, as the author interpreted (Zhao, 2020). In some cases, reciprocation cues reflected the banter was taken seriously as Zhao interpreted “he says “fuck off” and adds angry emojis at the end of the utterance. In doing so, this indicates a shift from a humorous frame to a serious frame” (p. 55). Their analysis shows how the valence of reciprocating emojis or other digital cues determines the success of banter. Yet, it should be noted that the receiver might fake a positive reciprocation to avoid awkwardness. In those instances, digital cues are used as face-saving behaviour that attempts to end the interaction on good terms. When asked why they use GIFs (graphics interchange format) as a reciprocation, one of Church et al. (2023) participants said:

“Sometimes people say things that are awkward or in poor taste. You don’t really want to respond, but you have the obligation to. So, a GIF is a funny, semi-sarcastic way of responding to it. And it can help divert the situation.” (p. 739).

The included studies showed that adolescents and emerging adults have a variety of choices when it comes to deciding which digital cues to use in conveying banter and humour. Digital cues can be tailored to the context and create nuances and subtlety in virtual communication to some extent. By observing interactions between adolescents, Aguert et al. (2016) found that around 80% of banter-type humorous online interactions are accompanied by digital cues. Some of them more explicitly expressed banter/humour such as the laughing face emojis or ‘jk’ (just kidding) initialism, while others were more implicit, such as the “…” (three dots), “!!!” (exclamation marks), or writing in capital letters (Aguert et al., 2016; Vandergriff, 2013). All such cues can be used in conveying humour but given the higher chances of misinterpretations in the online context, adolescents and emerging adults might prefer more explicit digital cues, as Aguert et al. concluded “adolescents used markers that left no doubt as to how their utterances should be interpreted” (p. 212). Yet, having choices available to express humour means adolescents and emerging adults can use them creatively, as a participant of Church et al. (2023) commented:

“It’s fun to find something that fits the situation and have everyone “haha” it in the group chat.” (p. 738).

In addition to utilising the pre-existing cues, the so-called digital natives (younger generations who have spent most of their lives surrounded by digital devices and social media (Prensky, 2001) could create a new version of digital cues for themselves, using the various tools the internet provides. For instance, creating user-generated memes using combinations of text (e.g., adding catchphrases), images or even someone’s photos (including their own) to use as banter tokens (Ask & Abidin, 2018; Lu & Fan, 2018; Vigmo & Lantz-Andersson, 2014). In addition to these photo and text-based digital cues, a short video formats such as GIFs (Church et al., 2023) and Vines (Kunze, 2014) have become a popular humour token among adolescents and emerging adults. In that sense, the virtual environment gives adolescents and emerging adults an unusual power to produce/orchestrate banter and convey humour that is hard to achieve in a face-to-face environment.

Digital cues are ambiguous

Another main theme generated from the included studies was that the digital cues were likely to be ambiguous as misinterpretations still occur despite the common use of them in banter-type exchanges (Buglass et al., 2021; Cui et al., 2023; Goodyear & Quennerstedt, 2020; Keane & Hammond, 2022; Steer et al., 2020). While the multiple ways of using and interpreting the same cue already make them ambiguous and open to interpretation (for example, according to Shardlow et al. (2022), the fire emoji could mean heat, actual fire, physical attractiveness, etc.), their lack of ability to express complex emotions in the same way non-verbal cues do in face-to-face communication could be an additional reason leading to ambiguity. That is, in face-to-face interactions, multimodal non-verbal expressions (simultaneous changes in tone of voice, facial expressions, body language, etc.) convey deep layers of emotions in a way that the instigator wants them to be. Whereas digital cues such as emoticons and emojis could be far too simple (or unimodal) to express such nuanced emotions. One of Steer et al.’s participants said:

“You can’t input your own, you can only type but can’t say it in the voice you want to say it in sort of, anything to make it funny or to really like, make it mean, that’s (.) the whole point, that’s one of the bad things about social media is (.) people can take it wrong ways” (p. 5).

Based on their participants’ views, such as the ones above, Steer et al. conceptualised using emojis during banter as “a low-level attempt to clarify intent” (pp. 7–8) and that “although emojis can have a useful interpretative purpose for young people they cannot be relied upon to provide full interpretation” (p. 7).

Due to such insufficiency of digital cues, multiple contextual evidence could be simultaneously activated and processed with them for a ‘full interpretation’ (Cui et al., 2023; Keane & Hammond, 2022). In Cui et al.’s experiments, the receivers relied on whatever additional contextual cues were available, such as the relationship with the sender, when interpreting banter-type ambiguous messages. Without such additional information, emojis alone were likely to be too vague to rely on. They described emojis as “potentially ambiguous even when surrounded by texts or embedded in contexts, let alone neutral contexts devoid of contextual cues” (p. 9).

The ambiguity of digital cues extended further when adolescents and emerging adults added novel meanings to existing cues that had no relation to their original meaning. For example, according to Cui et al. (2023), a smiling emoji that originally conveyed positive emotions is now used with a repurposed meaning among younger generations, often implying sarcasm and negative attitudes. However, since there are several different smiling emojis, such as ‘smiling face with smiling eyes’Footnote 1 (😊) and ‘slightly smiling face’ (🙂), and Cui et al. did not specify which smiling emoji they were referring to in Unicode or provide its official name, it is difficult to determine which emoji may carry this negative connotation. Nevertheless, such repurposed usage is more likely to circulate within in-groups (such as a group of friends or certain communities) and therefore makes it difficult to comprehend for those outside the groups. Another potentially repurposed meaning was observed regarding ‘lol’ (laughing out loud) initialism among university students (Keane & Hammond, 2022). The ‘lol’ initialism has been used widely over the years and generally has a positive valence that conveys laughter and humour (Vogler et al., 2019). However, contrary to this expectation, ambiguous texts with ‘lol’ were perceived largely negatively in Keane and Hammond’s study. This suggests a potential hidden or repurposed meaning of ‘lol’ that might contain a negative connotation. This shift in the meaning of the cues might not always be apparent to the wider public and could lead to misunderstandings and escalations.

In addition to assigning novel meanings to existing cues, adolescents and emerging adults can create their own in-group cues, further adding to the complexity of interpreting digital cues. For example, adolescent boys tended to use certain hashtags as a humour token in their social media posts, yet only certain viewers could comprehend their humorous/ironic connotation (Goodyear & Quennerstedt, 2020). As the authors said, “irony and humour were present within the boys uses of hashtags, such as, #bitofagymlad #gains #Iamagymlad […] these hashtags and meanings represented a certain kind of genre of communication amongst the boys, that those external to this media culture may not understand” (p. 28).

Harmful online banter containing digital cues

Rather than being ambiguous and open to misinterpretation, digital cues can also be intentionally utilised in harmful ways in the context of banter or humour (Allison et al., 2019Footnote 2; Buglass et al., 2021; Kellerman et al., 2013; Steer et al., 2020; Walker et al., 2021; Zhao, 2020). A particular way of doing this was apparently the effect of the so-called online roasting culture (insulting others humorously (Dynel & Poppi, 2019). This can range from leaving a jocular insult on a friend’s social media photos (Zhao, 2020) to engaging in an online community dedicated to roasting each other’s personal photos for fun (Allison et al., 2019). In that sense, personal photos were being used as online banter/humour initiator tokens. Although this behaviour can be perceived as a harmless joke, the risk of insulting another’s media was acknowledged and on average Allison et al.’s participants considered the mean comments on RoastMe subreddit as harmful and unacceptable despite the potential humorous elements:

“My natural reaction was to laugh but I felt bad laughing at a mean comment” (p. 16).

Sharing private photos and videos of others as banter without permission was another way of utilising digital cues in harmful ways. For example, in one of the studies, emerging adults used embarrassing photos of others (their friends) as a joke/humour and uploaded them to the internet without permission (Kellerman et al., 2013). Emerging adults may assume that others will find the joke amusing and may not realise that offense can be taken, perhaps due to the developing social awareness and banter skill (Whittle et al., 2019). One of Kellerman et al.’s participants said:

“I thought the embarrassing photo was funny, and hopefully they would feel the same way rather than insulted or embarrassed” (p. 296).

Similar behaviour was observed among adolescents (Steer et al., 2020). Screenshotting an other’s private media without permission and sharing it with an offensive caption could be considered humorous among adolescents, as one of Steer et al.’s participants described:

“Screen shotting something off of someone’s social media and then pasting it on your own, with like a harsh caption or something, I think that’s (.) quite embarrassing for the person… Like some people might think “oh, it’s a joke” I’ll do it, we do it, that to each other and that’s where it’s like banter, whereas if it’s just constantly, or even just once it can be extremely embarrassing.” (p. 4).

This sharing of others’ personal media ‘as a joke’ behaviour was exacerbated when emerging adults were found to be sharing inappropriate content of others without permission (Walker et al., 2021). Such actions, aside from being inappropriate, have the potential to be illegal. For example, in the UK, non-consensual sharing of sexually explicit images of others is illegal (McGlynn & Rackley, 2017). However, some emerging adults might use the terms ‘banter’ or ‘joke’ to mask this illegal and problematic behaviour. In Walker et al.’s study among anonymous, university students, about one-third of those who had shared someone’s sexually explicit photos without permission stated that their intention was to be humorous. As Walker et al. said: “although framing nonconsensual sharing as a “joke” and humorous, the motivation remains harmful and possibly with an intention to cause others distress” (p. NP9099). Indeed, regardless of the intention, invoking humour or banter can easily justify harmful and problematic behaviours as evidenced by the other included studies (Buglass et al., 2021; Steer et al., 2020). One of Buglass et al.’s participants said:

“Banter is like a term you use to cover yourself when you say something offensive, like an offensive joke, but you’re not sure if other people will take it bad. It’s like you say ‘oh, it’s just banter’ at the end as if that diffuses the whole thing.” (p. 296).

Personal and contextual effects

The last theme generated was about the importance of considering personal and contextual effects in using, perceiving, and interpreting digital cues as or during banter/humour (Aguert et al., 2016; Allison et al., 2019; Church et al., 2023; Cui et al., 2023; Keane & Hammond, 2022; Prada et al., 2018; Vandergriff, 2013). The effects of age and gender were investigated most frequently among the included studies. Particularly, while online bantering was more prevalent among older adolescents, the utilisation of digital cues during online banter and humour was more popular among younger adolescents, or 100% of their banter-type comments were accompanied by contextualisation cues (Aguert et al., 2016). Whereas for older adolescents, around one-third of banter-type humorous exchanges were free from digital cues. According to Aguert et al., this may be attributed to the enhanced verbal abilities that develop as adolescents grow older, allowing them to convey humour more subtly and with greater sophistication, relying less on overt cues. They stated that “the younger adolescents used markers to signal their ironic intention more frequently than the 16-year-olds did. If there is any developmental trend, it is that these markers are used less automatically and are more tailored to the context, but always with a view to being understood.” (p. 213). In terms of gender, young women tended to have a more positive attitude toward digital cues such as emojis and therefore utilised them more frequently (Prada et al., 2018). There was an interaction between gender and age that younger women were found to enjoy using emojis the most in the context of online humour (Prada et al., 2018).

When it comes to interpreting digital cues, the receivers tended to rely on social context such as the relationship with the sender and the sender’s social status, and this was more likely to be the case for young women. Specifically, digital cues following banter-type humorous comments were perceived and interpreted more positively when received from a high social status individual (Cui et al., 2023) and a close friend (Keane & Hammond, 2022). When similar ambiguous comments with digital cues were received from low social status people and acquaintances, the receivers were more likely to take offense. It shows how subjective the interpretation of ambiguous online comments can be when the influence of social context is added to the equation and the mere presence of digital cues might not be convincing enough to come to a positive conclusion. As Cui et al. highlighted, interpreting ambiguous messages involves not only bottom-up processing of the message content and features (e.g., emojis) but also top-down processing influenced by the receiver’s personal biases about the sender and broader social norms (e.g., sender’s occupation). These top-down factors have a role in shaping how the message and emojis are perceived.

Other factors that might influence the use and interpretation of digital cues as banter and humour were the effects of personal characteristics and platform norms. One of the included studies showed how the receivers’ personal characteristics such as rejection sensitivity may affect interpreting banter-type comments and digital cues, with people high on rejection sensitivity tending to interpret the meaning of digital cues following banter-type texts as more hostile (Keane & Hammond, 2022). As opposed to that, a factor that might reduce online ambiguity was the explicit norms of online communities. Where community norms explicitly allowed aggressive humour, the mean/roasting comments were taken as less offensive, and using the targets’ personal photos as a humour token was perceived as more acceptable within the online community space (Allison et al., 2019).

Discussion

Given the higher prevalence of online bantering among adolescents and emerging adults and its tendency to lead to miscommunications and escalations, there is a growing demand to explore online bantering experiences among these age groups (Betts & Spenser, 2017; Buglass et al., 2021). This review investigated how digital visual cues such as emojis, GIFs, etc. were being used, perceived, and interpreted among adolescents and emerging adults in the context of digital banter and humour and the factors that influence these, to understand the advantages and disadvantages of such cues in reducing banter-related ambiguity and perceived aggression. Seventeen papers were included in the review.

As expected, there was a satisfactory level of evidence supporting the roles digital cues play during digital banter and humour. Particularly, digital cues tended to be used either as a contextualisation marker that disambiguates banter-type texts or an individual token that initiates and reciprocates digital banter/humour. The former supports the existing notion that digital cues are the digital equivalent of face-to-face non-verbal cues that make online interactions emotionally intense and natural (Erle et al., 2021). This functional equivalence of digital cues to non-verbal cues was evident in this review in that even the way receivers faked a positive reciprocation with laughing GIFs to ‘an awkward’ online humour was identical to fake laughing during face-to-face humour. Laughing was the most explicit non-verbal cue that signals humour and banter and therefore seemed easier to compensate with laughing emojis, emoticons, GIFs, etc. Whereas the more implicit non-verbal cues such as changing the tone of voice and body movements were harder to compensate digitally yet there evidently were less explicit options such as three dots, capital letters, exclamation marks, etc. That is, adolescents and emerging adults seemed to find a way around making online bantering similar to face-to-face interactions. Yet, young people felt more comfortable using more explicit cues such as laughing emojis than three dots or exclamation marks in this review, as it was evident in a non-banter context as well (Forbes & Buchanan, 2019).

However, the perceptions and views of digital cues as banter/humour tokens were not homogeneous in that adolescents and emerging adults described how these cues can increase online ambiguity and misinterpretation at times. This was indicative of issues such as the lack of ability of digital cues to convey complex meanings, multi-ways of using the same cues, lack of access and reference to a universal dictionary of meanings, and in-group or novel meanings assigned to cues (Shardlow et al., 2022; Steer et al., 2020). In addition, existing literature suggests that digital cues such as emojis are not only open to interpretation due to the lack of consensus on the meanings they represent, but also, they could be perceived as a ‘display rule’, not an indicator of genuine emotions (Liu, 2023). Future studies may explore whether this finding is consistent with the views of adolescents and emerging adults. The inappropriateness of the intensity of emotions conveyed through the digital cues was also linked with misinterpretations and ambiguity (e.g., repetitive emojis could be perceived as more intensive than a single emoji) (Stein, 2023). Again, this notion has not yet been explored among young people and within a banter context, showing the need for future research.

Moreover, there were cases where digital cues were intentionally utilised in harmful ways and the most common way of doing it was sharing private media of others without permission. Although adolescents and emerging adults tended to interpret this behaviour as a joke, it did not conceal the potential malicious intention of cyberbullying and even a criminal offense. The ambiguous nature of visual cues may allow individuals with malicious intent to disguise their behaviour as humour. While the masking of harmful behaviours through banter has been described in studies exploring the nature of banter (Abell et al., 2023; Buglass et al., 2021), little attention has so far been given to the role digital visual cues have in such “banter”. This is particularly important because when humour is involved, it becomes harder for bystanders to intervene in problematic behaviours (Scott et al., 2019) and targets to standup for themselves (Douglass et al., 2016). Without appropriate intervention, the targets of such harmful “bantering” are at risk of suffering from mental health issues such as social anxiety (Douglass et al., 2016). Future research should consider the intent behind the use of digital visual cues, with the goal of raising awareness among young people about banter that may intentionally cross the line of acceptability.

Lastly, the included studies showed, personal and contextual factors such as age, gender, personality, social context, and platform norm could affect the use, perception, and interpretation of digital cues as banter/humour. These factors tended to explain the variances in interpretations and utilisations to some extent in line with existing literature in the non-humour context (Koch et al., 2022; López-Rúa, 2021).

Limitations of included studies

Although these findings constitute a relatively comprehensive picture of how digital cues are used during online banter and humour along with the multifaceted views and interpretations of adolescents and emerging adults, there were some common limitations in the included studies. Particularly, as investigating the role of digital cues as banter/humour tokens were not the main research question of most of the included studies, relevant findings, extracts, and discussion were limited. Initial literature scoping showed how creating ambiguous online memes and circulating them in the framing of a joke could be harmful to young people’s wellbeing (Salcudean, 2020; Vickery, 2014), yet such studies were not included in this review as they did not meet the age range criteria. This shows that despite the wide range of studies exploring digital cues in the context of banter and humour, most of them have not specifically targeted adolescent and emerging adult populations and investigated their opinions and perspectives yet.

This limitation was also evident in how included studies could not explain whether ‘lol’ initialism holds a different meaning than laughter among young people and if so then what the meaning is and how it is being used. Also, the evidence of assigning novel meanings and creating in-group cues was limited to a few hashtags and a smiley emoji in this review, while real-life experiences of alternative usages of digital cues among these age groups could be more extensive in both breadth and depth. These concerns were doubled when some of the included studies were based on authors’ interpretative analysis of young people’s online interactions without verifying whether their interpretations were actually grounded in young people’s interpretations and views (Harcourt & Einarsdottir, 2011). Thus, without exploring adolescents’ and emerging adults’ real-life experiences, perspectives, and interpretations in their own words, understanding their usage of digital cues as banter/humour tokens is partial and limited.

Another main limitation of the included studies was the tendency to highlight and explore the use and interpretations of emojis and emoticons more than other cues such as GIFs, stickers, memes, etc. This might be because emojis and emoticons are used more frequently than other cues; however, given the rapid development of using and creating digital cues among young people, there is a need for studies that do not limit the role of digital cues during online banter to a single/chosen cue.

Limitations, strengths, and implications of the review

The first limitation of this review was the exclusion of non-English and grey literature. Although excluding grey literature could enhance the quality of the findings (Hopewell et al., 2007), exploratory reviews on this relatively novel phenomenon may benefit from the views and interpretations that come across in non-peer-reviewed publications. Also, due to the relatively fast-paced development of digital communication culture, the ways young people used digital cues in the included studies may have shifted, potentially making the findings less applicable to current practices. To address this temporal nature of digital culture, continuous research is necessary to keep pace with how young people engage with digital cues.

To the best of our knowledge, this is the first systematic review that has synthesised literature findings on the use, perspectives, and interpretation of digital cues in the context of digital banter and humour. This review systematically showed the ways digital cues were viewed to help reduce perceived ambiguity and aggression along with how they were insufficient and used inappropriately as banter/humour in the current state.

In terms of the implications, the findings of this review collectively highlight the areas that future research may need to focus on, particularly (1) the links between using the term banter and inappropriate behaviours and how digital cues are involved in (e.g., the sharing without permission behaviour), (2) the ways using digital cues affect the receiver’s interpretation of the banter and potentially their wellbeing, (3) variabilities in young people’s uses and interpretation of digital cues as banter token and factors associated with these, etc. The findings of this review may presumably facilitate further debate, discussion, and other research output in this area. Meanwhile, as Steer et al. (2020) suggested, raising awareness among adolescents and emerging adults on how people use and perceive the same cues differently due to various psychosocial factors could help reduce miscommunications and escalations. In addition to that, awareness raising on the harmful consequences of sharing other’s private media without permission could be crucial in reducing banter-related conflicts and perceived cyberbullying. Lastly, practitioners may need to explain to young people how in-group cues and novel meanings assigned to existing cues may create exclusions and negatively affect the wellbeing of those outside the group.

Conclusion

This review finds mixed views and evidence on the effectiveness of utilising digital cues in reducing perceived aggression and ambiguity during online bantering. On the one hand, digital cues are helpful and disambiguate banter-type humorous comments, but on the other hand, there is room for improvement, especially concerning the variations in interpretations and uses. The shortcoming of digital cues as banter tokens is not limited to plain misinterpretations but in some cases, they can be intentionally utilised with potentially harmful intentions. With the help of more research and awareness raising, the ambiguity and harmful utilisation of digital cues during online banter may decrease. The need for research that is grounded in adolescents’ and emerging adults’ views and interpretations is prevalent for a more comprehensive understanding of digital cues in the context of banter/humour.