Abstract

The growing variety and availability of live streaming video content has led to many users, especially younger users, turning to it to learn new information. While previous research has documented some motivations of users who seek out live streaming environments, not much is known on how those motivations and other psychological needs differ in comparison to nonlive video content, often available on the same social media platforms. Live versus nonlive social media videos (SMVs) offer different affordances and ultimately different experiences. We employed a survey of 18- to 24-year-olds (N = 369) to further explore these differences. Specifically, we think affordances of live versus nonlive streaming video has the potential to shape what may come before viewing (motivations), what happens while viewing (information behaviors, psychological needs being met), and what someone takes away from viewing, after it has ended (perceived learning). Results show that informational, entertainment, social, and community motivations were all greater for nonlive SMVs than live streaming videos, as was perceived learning and several information behaviors. Two of the three psychological needs explored were better met by live streaming viewing experiences. The influence of motivations and psychological needs on perceived learning and information behaviors was also explored. Theoretical and practical implications are discussed.

Keywords: live streaming, social media videos, motivations, perceived learning, information behaviors

Funding: This research received no funding.

Disclosures: The authors have no conflicts of interest to disclose.

Data Availability: The data from this project have been made public and are available at https://osf.io/auq7k/?view_only=b844ff4d29764372b107d6f792b15f82 (Rubenking & Strawser, 2023). Any additional information can be requested from the corresponding author and will be provided.

Correspondence concerning this article should be addressed to Bridget Rubenking, Nicholson School of Communication and Media, University of Central Florida, 12405 Aquarius Agora Drive Orlando, FL 32816, United States [email protected]

Across content-specific apps and more widely used social media, live streaming is growing in popularity. In 2019, 126.7 million individuals in the United States viewed live streaming content on mobile devices, while 23% of Americans live streamed content themselves (Ceci, 2022). While more individuals are live streaming and viewing live streaming regularly, the variety of content available continues to diversify. Indeed, live streaming is no longer synonymous with eSports and gaming: In the United States, 37% of those who watch live streaming watch live news content (Ceci, 2022).

Live streaming content is now becoming an important information source across platforms and genres. In September 2022, a New York Times headline read, “For Gen Z, TikTok is the new search engine” (K. Huang, 2022). The article cites research that finds 40% of young people turn to search functions on TikTok or Instagram when looking for basic information, such as where to go for lunch. The present study is interested in better understanding why young people are turning to live streaming content and how they are engaging with that information and perhaps learning from it. Specifically, this study explores what may precede the live (and nonlive) streaming video experience (i.e., motivations driving the use), along with what they are doing while engaging with the content (i.e., meeting various psychological needs and engaging in information behaviors), alongside their evaluations of what they have learned from it after viewing has ended (i.e., perceived learning). We propose that common motivations to engage with social media and video content (e.g., informational, social, entertainment) as well as psychological needs (e.g., autonomy, relatedness, and competence) may differ according to the different affordances that live versus nonlive social media video viewing experiences offer. Further, information seeking, sharing, and commenting, as well as perceived learning from the content itself may be differentially influenced by both motivations and psychological needs.

While TikTok and Instagram have live streaming social media video affordances, such that users can “go live,” they offer other more general social media functions including nonlive videos as well. Much of the previous research that examines motivations behind viewing live streaming, including informational or learning-based motives, has been done with topic-specific live streaming sites that focus on gaming or esports, such as Twitch (e.g., Cabeza-Ramírez et al., 2020; Diwanji et al., 2020; Sjöblom & Hamari, 2017; Sjöblom et al., 2019), although some research has investigated general social live streaming site use as well (Long & Tefertiller, 2020).

This article begins with an articulation of what affordances differentiate live streaming social media video (LSSMV) from nonlive social media video (SMV) content. It then introduces motivations to view under a uses and gratifications theoretical approach. Affordances differentially allow for different motivations to view, such as information-seeking motives being more easily obtained by content with information presented across multiple processing streams or social motives being better met by videos that encourage greater communication between viewers and creators. Next, the psychological needs of autonomy, relatedness, and competence, following self-determination theory (SDT; Deci & Ryan, 2002), are discussed also through the lens of how different affordances may allow viewers to better meet these needs.

With greater numbers of individuals, and specifically younger individuals, turning to online video content on social media for information, it is important to understand how their motives and needs may predict both what they do with the information in these environments and their own assessments of what they have learned. Knowledge gained on what differentiates live streaming and prerecorded video content, often found on the same sites, should be relevant to both content creators and viewers.

Background

Live Streaming Versus Prerecorded Social Video Affordances

The concept of affordances was introduced by Gibson (1977), who noted that environmental stimuli afford, or allow, different behaviors or action possibilities. Objects and environments suggest ways in which to interact with them. Affordances, in other words, are “generally used to describe what material artifacts such as media technologies allow people to do” (Bucher & Helmond, 2017, p. 3). Scholars in various fields have defined affordances in a multitude of ways. For a review of social media affordances, see Bucher and Helmond (2017), and for a model of affordances in digital media, the modality, agency, interactivity and navigability model is articulated by Sundar (2008). Pertinent to the current article is the view that social media afford users new actions and behaviors, which can, in turn, create new user needs or motivations (Sundar & Limperos, 2013). We posit that LSSMVs and SMVs offer somewhat different affordances, which may influence the motivations to seek them out and perceptions about basic psychological needs being met.

While the affordances of viewing these live streaming and nonlive videos share many similarities, they also differ in ways that are critical to the viewing experience. The two content delivery systems differ in the mode of transmission: Live streaming is created and shared simultaneously, whereas nonlive videos are available for viewing and/or downloading after being recorded. Several platforms limit which users can live stream, largely based on the number of followers one has. For instance, users of TikTok must have 1,000 or more followers to go live.

Lin et al. (2019) state that much of the academic research on live (gaming) streaming has been descriptive, as opposed to explanatory or predictive. They propose a tripartite definition of live gaming streaming, which includes the performative nature of play (a), the interaction between streamers and audiences (b), and the synchronicity of those interactions (c) (Lin et al., 2019). This interaction and its synchronous nature are quite relevant to the present study, although their first part, on the “performative nature of play,” is more specific to gaming than the present study’s goals. This asynchronous interaction is the first of two key differences we propose that makes live streaming (of any content) different from prerecorded videos available on the same platforms: (a) synchronous communication and (b) the availability of video visuals and effects.

Synchronous Communication

First, live streaming video on social media platforms has a more conversational back-and-forth between viewers and content creators. This is due to the synchronous interaction between content creators and viewers and among viewers with one another afforded by live streaming (Morain & Swarts, 2012). Streamers and viewers alike can see comments made in real time during a live stream. Content creators can respond through videos or comments, with some of the most popular content creators using moderators to help. This interaction’s synchronicity is explicitly due to its live nature: When commenting on a posted video or reel, commenters do not expect an immediate response. The time-sensitive nature of live streaming videos invites more interaction at the moment, while viewing.

Availability of Video Visuals and Effects

Second, live streaming videos also do not offer the same video visuals and effects that prerecorded videos often include (Gilbert, 2019). Video visuals and effects can increase the perceived production value of a video and can introduce other visuals and sounds that enhance the content being accessed. Some types of information or content benefit from additional video or audio. There is considerable research on the types of content that make for popular or “viral” videos (Guadagno et al., 2013), but less research on what form or medium variables, such as the use of visual effects, influence outcomes, such as popularity, learning, and information behaviors. However, we know from previous research on television news content that redundancy of information in both the audio and visual processing streams leads to increased memory for the content presented (Fox, 2004; Lang, 1995). Since it is considerably more likely for prerecorded videos to include some complementary visuals to the content presented, it may be more likely for information to be retained to a greater degree.

Uses and Gratifications

When new media technologies and platforms emerge, or when new affordances and applications within these platforms become widespread, media and communication scholars have often first explored users’ motivations. These motives can be studied under a uses and gratifications approach (Katz et al., 1974). Uses and gratifications presume that audiences are active, rather than mostly passive, and that they make goal-directed media (and nonmedia) selections to fulfill gratifications (Katz et al., 1974). Uses and gratifications have been applied to television viewing (Rubin, 1983), television and user-generated video viewing (Bondad-Brown et al., 2012), social media platform switching and using (Tandoc, 2019), live streaming (Young & Wiedenfeld, 2022) and viewing live streaming. The approach outlines a continuum of media use, which runs between active, rational audiences and passive, complacent audiences. Indeed, one can see the approach’s lasting effect in that we now refer to internet audiences not as “audiences” at all but as “users” (Sundar & Limperos, 2013).

Under uses and gratifications, there is also a distinction between gratifications sought and gratifications obtained, which can be separated from a measurement perspective to look at both motives and outcomes of media use (Palmgreen et al., 1985). While a person may actively select a live streaming video experience to escape and relax, those goals could be met, or not. For example, although someone tuned in for escapism and relaxation motives, they may instead find social support and community, experiencing different gratifications obtained. During any media experience, gratifications sought may be obtained, as can others which were not sought. Often, multiple gratifications are both sought and obtained.

Motivations to View SMVs

As the content featured in LSSMVs evolves, so do the motivations to engage with it. Much of this research has employed a uses and gratifications theoretical approach. While social and community motivations, along with entertainment motives, have been found to be among the most important motivations to view LSSMVs (Hilvert-Bruce et al., 2018; Sjöblom & Hamari, 2017), so too have information-seeking or cognitive motivations. For instance, Hilvert-Bruce et al. (2018) found three of eight motivations to engage in Twitch live streams were significant predictors of increased time spent viewing: entertainment, information-seeking, and social interaction. Further, Cabeza-Ramírez et al. (2020) found informational motives to be a larger contributor to viewing live streaming than both social and entertainment motives among those viewing gaming live streams. Yet, another study that explored motivations to watch live social streaming videos that included hosted live shows, live game broadcasting, and other types found five motives predicted viewing: real-life communication, escape, fun-seeking, partnership-seeking, and social interaction (Long & Tefertiller, 2020).

To the researchers’ knowledge, a direct comparison of motivations to watch live versus nonlive SMVs does not exist in the academic literature. Research on nonlive online video consumption varies, largely by platform or device, or type of content (e.g., advertising videos, viral videos). Research on YouTube or TikTok videos often does not distinguish between live and nonlive videos. For instance, entertainment and informational motives seem to be common among viewers of YouTube (Lagger et al., 2017), and information seeking, social interactions, and relaxing entertainment motives have been demonstrated to positively influence engagement with online videos, including liking and commenting on video content (Khan, 2017). Further, a study of TikTok consumption and other behaviors examined five motivations to consume videos (Omar & Dequan, 2020). They found that escapism was the largest predictor, followed by social interaction and archiving motivations.

Other research has explored motivations to engage in (online) synchronous and asynchronous communication about live video content. The realm of Social TV research explores viewing a TV show, live, as it airs, while simultaneously chatting with (known or unknown) others via social media (or a second screen) about the content. This shares both similarities (asynchronous communication about content, live, as it airs) and dissimilarities (the typical use of a second device; no interaction with the content creators) with our examination of live versus nonlive SMV viewing. Research in this vein has found that motivations typically seen in social media use, television viewing, and (computer-)mediated communication are also found in Social TV interactions, such as information sharing and seeking, feelings of coviewing, and curiosity about others’ opinions (Han & Lee, 2014), as well as motives to communicate with others, gather information, and be entertained (Krämer et al., 2015).

Of specific interest to the present study, Unkel and Kümpel (2020) explored the motivations of Game of Thrones viewers to post on Reddit pages both before and after broadcasting of the final seasons’ episodes (which afforded asynchronous communication between fans) and posting during the broadcast of each episode (which afforded synchronous communication). They found that information-related motives were better met via asynchronous communication periods, while the need for feeling like they were with others, labeled “company,” was the only motive better met by synchronous interactions (Unkel & Kümpel, 2020). This suggests that in the present study, LSSMVs may better support social/community motives, whereas informational motives may be better met by viewing SMVs.

Previous research suggests specifically that informational and social motives may be better met by different types of SMVs. Nonlive videos potentially offer two advantages over live streaming videos when it comes to eliciting informational motives. Informational motives may be better met by content that can include additional video editing and graphics that can repeat and enhance the information being delivered through the auditory stream, as previous research has suggested that information redundancy can be beneficial for memory and learning outcomes. Further, SMVs may have an advantage for learning motivations in that they can be saved for future viewing and paused, stopped, and returned for later reference. Having this additional control over the content may encourage individuals to seek out this type of content when they want to gain specific information. As for social and community motives, we believe the synchronous communication afforded by live streaming affords viewers a more social and community-driven experience. The ability to talk, in real time, with content creators and other viewers should create a more social community experience than SMVs, which certainly allow for comments but do so in a less conversational way. Therefore, we propose:Hypothesis 1: Informational motives will be greater for SMVs than for LSSMVs. Hypothesis 2: Social interaction and community motives will be greater for LSSMVs than for SMVs. Research Question 1: Will entertainment motives vary by video type?

SDT

This study also employs SDT as a framework to understand the underlying motivations of LSSMV and SMV viewers. At its core, SDT recognizes that we have growth tendencies and psychological needs that serve as the underlying foundation for our self-motivation (Ryan & Deci, 2000). Three distinct psychological needs heavily influence motivation: autonomy, the feeling that we have a choice regarding our behavior or acting by will or volition (Jang et al., 2009); competence, the need to be effective and demonstrate mastery especially in social contexts (Deci & Ryan, 2002); and relatedness, the need to feel a sense of belonging or connectedness to others (Bowlby, 1969). If these conditions or needs are met, our intrinsic motivation will increase. Specifically, if our psychological needs for autonomy, competency, and relatedness are satisfied, then feelings of interest and curiosity, variables especially important for learning environments, will be increased (Reeve, 2009).

In digital and virtual contexts, SDT can be a helpful lens by which researchers can view the behavior of online participants. For instance, Ryan et al. (2006) applied SDT to video game play and found that the theorized psychological needs represented in SDT (autonomy, competence, and relatedness) predict enjoyment and future gameplay. Furthermore, Berezan et al. (2018) found that the virtual happiness of social media users can increase when our self-determination needs are met. Montag et al. (2021) believed that motivated behavior, in our case using SMV platforms, should be high when the platform allows users to feel autonomous, competent, and connected to others.

Autonomy, relatedness, and competence have also been studied in relation to information-seeking and sharing behaviors online. Lee and Lin (2016) found that all three were reported as greater in online health information-seeking sessions than in face-to-face sessions with physicians and that autonomy and relatedness led to greater intentions to seek out health information online. Similarly, Wu et al. (2021) found autonomy to be a critical predictor of knowledge-sharing behavior in a health care setting. In summary, by creating a space for users to feel connected, competent, and free to act according to their own choices, social platforms can positively increase motivation for use.

As a framework for further understanding learning, especially perceived learning and learner behaviors in virtual environments, SDT can also be applied to show how the fulfillment or obstruction of these psychological needs can enhance or undermine learning and behavior (Y. Huang et al., 2019). Learners, then, should have autonomy, competency, and relatedness needs met in order to see a higher level of engagement in a virtual context (Chiu, 2022). Diagnosing how these needs are met in video learning environments, especially social live streaming platforms, can provide insight into the underlying motivation of participants and their perceived level of learning that was achieved throughout the virtual social experience.

Previous empirical research does not provide much direction on how autonomy and competence may be better met by SMVs or LSSMVs. However, we think the same synchronous communication afforded by live streaming videos that should encourage social motivations for viewing, may also better meet relatedness needs. Since relatedness is a psychological need centered on connectedness and togetherness, we believe the video viewing experience that best encourages dialogue among viewers and creators should better meet this need.Hypothesis 3: Relatedness needs will be better met by LSSMVs than SMVs. Research Question 2: Will autonomy and competence vary by video type?

Perceived Learning

While not the sole indicator of whether learning has been achieved, perceived learning can help us understand how learners view their experiences. Alavi et al. (2002) defined perceived learning as “changes in the learner’s perceptions of skill and knowledge levels before and after the learning experience” (p. 406). The key factor in perceived learning is the actual perception of the learner that learning has been accomplished. In traditional educational environments, like colleges and universities, students who have a higher degree of perceived learning, meaning students who believe they have learned content related to their courses, tend to be more active participants in class (Fredericksen et al., 2000), have higher grades (Rockinson-Szapkiw et al., 2016), and enhanced learning experiences (Alavi et al., 2002). It is beneficial for learners to believe they have learned something throughout their participation.

Perceived learning as a key learning indicator continues to be a major component of content effectiveness (Wright et al., 2006). As students or consumers of educational content move throughout the learning process, a cognitive change should ensue. Specifically, individuals should be able to acquire, retain, and then actually apply the knowledge that has been communicated (King & Witt, 2009). These categories were further reiterated by Frisby et al. (2014) who defined cognitive learning as the acquisition, retention, and application of knowledge. For the purpose of this study, the perception of this cognitive progression rather than the change itself is most important. Actual learning, according to Bacon (2016), refers to a change in knowledge that is typically identified through rigorous measurements of learning. Perceived learning, on the other hand, is more concerned with a student’s self-report of any knowledge gained based on in-depth self-reflection. In some ways, perceived learning is also closely related to a positive effect for the learning experience itself (Bacon, 2016). In nontraditional virtual education contexts, like live streams, the perception that viewers have learned may also highlight perceptions of the learning experience, including the community, satisfaction, and other variables. All told, measuring perceived learning can help us identify how participants view their own knowledge after viewing LSSMV (live) or SMV (nonlive) content.

Based on what we know about motivations to view live and nonlive SMVs, it seems reasonable that informational motives should be related to perceived learning. This follows logic from uses and gratifications that if gratifications obtained fail to materialize after being sought, alternative choices will be made. Less is known about the relationship between other motives and perceived learning. SDT is quite clear that when basic psychological needs are met, a host of sought-after outcomes can be identified. Because autonomy, relatedness, and competence inform intrinsic motivations and are associated with a number of positive outcome variables when studied in relation to media use, we believe meeting these needs should be indicative of greater perceived learning. Therefore, we propose:Hypothesis 4: Information-seeking motives, autonomy, relatedness, and competence should lead to greater perceived learning.

We have outlined synchronous communication as an affordance of live videos and video visuals and effects as an affordance of nonlive videos. We believe the professionalism and audio–video redundancy afforded by some nonlive videos may lead to greater perceived learning.Hypothesis 5: Perceived learning will be greater when viewing SMVs than LSSMVs.

Information Behaviors

Information-seeking behavior has often been studied to a greater degree than other information behaviors (Wilson, 2000). However, information behaviors are an umbrella term that encompasses many different actions related to consuming, seeking, foraging, retrieving, organizing, sharing, and producing information (Diwanji et al., 2020). Wilson (2000) broadly defined information behavior as “the totality of human behavior in relation to sources and channels of information, including both active and passive information seeking and use” (p. 49). He further delineates between information seeking, searching, and use behaviors.

Similar to the different phrasing and definitions of motivations to use media, categorizing and defining information behaviors varies across studies as well. In their study of information behaviors on Twitch, Diwanji et al. (2020) looked at four categories of such behaviors: information production, information reception, information reaction, and information reward. Omar and Dequan (2020) explored three TikTok behaviors related to information: consuming, participating, and producing content and how each of these was differentially predicted by personality traits and motivations. They found user motivations, including social interaction, archiving, self-expression, peeking, and escapism to better predict all three information behaviors than the Big Five personality traits measured. Khan (2017) also found that different motives for use predicted different information behaviors in a YouTube context. They found information seeking predicted increased liking, disliking, and commenting on videos, although it did not predict sharing or uploading videos (Khan, 2017).

The present study is most interested in active information behaviors related to viewing videos. As such, video interacting is explored as an outcome variable, as well as information sharing, which encompasses online and offline sharing of information gleaned from the two types of video content studied here. Based on previous research (Omar & Dequan, 2020) that found social interaction to positively influence information interaction, we believe that social motives, community motives, and relatedness should positively predict information interacting and information sharing. Further, informational motives should predict engagement in information behaviors. Since competence is related to demonstrating effectiveness and mastery, we believe it should be implicated in information sharing and information interacting.Hypothesis 6: Information motives, social motives, community motives, relatedness, and competence experienced should lead to greater engagement in information-interacting behaviors (a) and information-sharing behaviors (b). Hypothesis 7: Information interacting should be greater for LSSMVs than SMVs. Research Question 3: Will information sharing vary by video type?

Method

Sample and Procedure

Participants were randomly assigned to one of two conditions or surveys. One survey asked participants to answer all items in reference to LSSMVs (N = 184), and the other asked exclusively about viewing nonlive SMVs (N = 185). This allowed us to differentiate information behaviors and perceived learning across live streaming content and other streaming content.

Participants were undergraduate students from a university in the southern United States, who received extra or course credit for their participation. All procedures were done with the approval of the university’s institutional review board. After completing demographic items, participants answered questions about the frequency of viewing videos both in general and in relation to how frequently one sought informational content; motivations to view; information behaviors; as well as measures of autonomy, relatedness, and competence, followed by measures related to the perceived credibility of the content and perceived learning. The survey took participants approximately 12 min to complete. The data from this project have been made public and are available on the Open Science Framework at https://osf.io/auq7k/?view_only=b844ff4d29764372b107d6f792b15f82.

Due to the focus on younger users of live streaming videos, 11 participants’ data were discarded for being over 24 years of age. The remaining sample had an average age of 19.30 years (SD = 1.34) and was 66.4% female, 31.7% male, and 1.9% nonbinary/preferred not to say. Across all participants, 62.9% identified as White, 11.4% as Black, 25.7% as Hispanic/Latino/a/x, 14.9% as Asian or Asian American, 1.1% as Hawaiian or Pacific Islander, .3% as American Indian or Alaskan Native, and 2.7% as other.

Measures

All scales detailed below, including the four critical study variables, motivations to view, psychological needs, perceived learning, and information behaviors, as well as our two control variables, perceived credibility and accuracy, are measured on 5-point scales of agreement.

Motivations

Information, entertainment, community, and social interaction motives were measured for both LSSMV and nonlive SMV content viewing. The four informational motive items and three entertainment motive items were modified from Hilvert-Bruce et al.’s (2018) study, which also employed Peterson et al.’s (2008) eight-item Brief Sense of Community Scale, also adapted here. The five items used to measure social interaction motives were borrowed from Long and Tefertiller (2020). Reliabilities for each of the four scales, across both viewing conditions, ranged from .819 to .909.

Autonomy, Competence, and Relatedness

The variables of autonomy, competence, and relatedness were measured with a modified version of a scale employed by Berezan et al. (2018). Each scale contained three or more items, and each of the three subdimensions, across the two conditions, reached reliability coefficients of .700 or above, except for autonomy in the nonlive video condition, which reached an α = .630.

Information Behaviors

Measures of information behaviors were based on concepts discussed in several articles, including Diwanji et al. (2020), but ultimately created specifically for this study by the researchers. This study is interested in two categories of active information behaviors: information interacting, which is concerned with commenting and liking video content or otherwise interacting with the content creator or audience while viewing, and information sharing, which is considered as a postviewing behavior wherein individuals share the videos or information from the videos with others.

Reliability analyses demonstrated high coefficients for the four-item index of information sharing. Sample items include “I will share (live social video streams/nonlive social media videos) with others outside of the app that I’m viewing them in” and “I often will talk about the content I learned from (social media live streams/nonlive social media videos) with others in face-to-face contexts.” An initial reliability check found values lower for information interacting. An evaluation of items revealed that the reliability would increase if the sole item on “liking” videos was omitted from the two items about commenting, which left two items on comments, including, “I often comment on live social media video streams that I watch” and “I like asking questions and talking with others during live streaming viewing.” Moving forward, information interacting is a measure of information commenting (α = .902 for live streaming video; α = .749 for nonlive video).

Perceived Learning

Individuals’ perception of learning was measured with the Cognitive Learning Measure (Frisby et al., 2014). The measure is from the instructional communication literature and has three dimensions: acquisition, retention, and learning. Since the scale was originally developed to measure perceived learning inside of the classroom, items were modified to refer to the video viewing experience (either live or nonlive). Participants answered the items in reference to the last time they sought out live streaming content for informational purposes. The scale reached acceptable reliability coefficients: α = .777 for live streaming video and α = .767 for nonlive video.

Perceived Credibility

Measures of perceived credibility were measured by adapting two previously used scales and used as covariates in hypothesis testing. The first was a modified News Credibility Scale from Yale et al. (2015). Participants were asked to think of the last time they watched either live streaming video or nonlive SMVs with the goal of learning new information and then rate their agreement with each of 10 descriptors about the “content creator” (e.g., balanced, accurate). This scale showed high reliability (α = .967 for live streaming video; α = .946 for nonlive video). A second measure of credibility is the Accuracy of Depictions subscale of the Trust in News Media Scale (Prochazka & Schweiger, 2019). These four items also achieved high reliability: α = .952 for live streaming video and α = .943 for nonlive video.

Results

In an effort to capture basic descriptive data, time spent viewing LSSMVs and SMVs was measured with two items each. The first question asked how frequently individuals viewed each, on an 8-point scale ranging from never to 7 days per week, and a second question asked about the duration of viewing on an average day that one viewed on a 7-point scale, ranging from less than 30 min to 5 hr or more per day. SMVs were watched more frequently (M = 4.65, SD = 3.34) than LSSMVs (M = 3.49, SD = 3.05) and for longer each day that they were viewed (M = 2.10, SD = 1.59) than LSSMVs (M = 1.81, SD = 1.17).

The majority of the hypotheses and research questions explored differences in the variables of interest (i.e., motivations, psychological needs, perceived learning, and information behaviors) across live versus nonlive videos (SMVs vs. LSSMVs) and were tested via one-way analyses of variance. These results, which cover Hypotheses 1, 2, 3, 5, and 7, and Research Questions 1, 2, and 3, are discussed below in the text, and descriptive data and effect sizes are presented in Table 1.

Hypotheses 4 and 6, which predicted multiple influences on lone dependent variables, were tested via multiple regressions. All analyses were conducted via SPSS v.28.

Differences in Live Versus Prerecorded SMVs

The first set of hypotheses predicted differences in motivations to view by video type, based on different affordances of each type of video. As predicted in Hypothesis 1, informational motives were greater for SMVs (M = 3.91, SD = .85) than LSSMVs (M = 2.88, SD = 1.12); F(1, 366) = 99.733, p < .001, η² = .214. This is a large effect size per Cohen’s suggested interpretation. In fact, all motivations were significantly higher for SMVS as compared to LSSMVs. This is in direct contrast with Hypothesis 2 and answers Research Question 1. The difference was greatest for informational motives, followed by community and entertainment motives, and the smallest (yet significant) difference was for social interaction motives. Data are presented in Table 1.

Hypothesis 3 and Research Question 2 addressed differences in psychological needs by video type, positing that relatedness needs were greater for LSSMVs than SMVs, while a research question was posed about autonomy and competence. Hypothesis 3 was supported: Relatedness was greater for LSSMVs (M = 3.64, SD = .91) than for SMVs (M = 3.36, SD = .84); F(1, 366) = 8.907, p = .003, η² = .024. This can be interpreted as a small effect size, according to Cohen. Similarly, autonomy needs were met to a greater degree in LSSMVs (M = 3.94, SD = .92) than in SMVs (M = 3.52, SD = .77); F(1, 366) = 21.772, p < .001, η² = .056 (medium effect size). No differences in competence by video type were found. See Table 1, for full data.

Differences in the dependent variables explored here—perceived learning, information-commenting, and information sharing—by video type were also explored and results are presented in Table 1. Hypothesis 5 predicted that perceived learning would be greater while viewing nonlive (SMV) content as compared to live videos (LSSMVs). This was supported: Perceived learning was greater among viewers of nonlive content (M = 3.69, SD = .56) than of live content (M = 3.47, SD = .62); F(1, 366) = 12.852, p < .001, η² = .034. Based on Cohen’s suggested interpretation, this is a small-to-medium size effect.

Hypothesis 7 predicted that information commenting would be greater for viewers of LSSMVs than for SMVs. The opposite pattern emerged, such that information commenting was greater in response to nonlive video content (M = 2.43, SD = 1.12) than for live video content (M = 2.08, SD = 1.18); F(1, 366) = 8.268, p = .004, η² = .022, with a small effect size. Hypothesis 7 is not supported. Last, Research Question 3 asked if information sharing would differ between video types. It did; information sharing was undertaken by viewers of nonlive content (M = 3.65, SD = .91) to a greater degree than by viewers of live content (M = 2.21, SD = 1.17); F(1, 366) = 174.066, p < .001, η² = .322. This effect size is large, per Cohen. All three of these variables, perceived learning, information commenting, and information sharing, were greater for viewers of nonlive SMVs than for LSSMVs.

Influence of Motivations and Psychological Needs

The hypotheses about the influence of motivations to view (information, entertainment, social interaction, and community) and psychological needs (autonomy, relatedness, and competence) on perceived learning, information commenting, and information sharing were tested via three multiple regressions, presented in Tables 2–4. The first block of all three regressions included age, sex, and the two measures of perceived credibility of content as control variables. No multicollinearity issues were apparent, as the highest variance inflation factor was for perceived credibility (3.026), which is well below the cutoffs of 10 and 4 suggested in the literature.

With controls, motives, and needs to be regressed on perceived learning, 48.2% of the variance was explained. Hypothesis 4 predicted that informational motives, autonomy, relatedness, and competence would all positively influence perceived learning. Perceived credibility and perceived accuracy (the second measure of credibility) were both significant predictors of perceived learning in the final block. So too were information motives, relatedness, and competence. However, autonomy, the final hypothesized predictor here, was not significant. Thus, Hypothesis 4 was mostly supported. Results are presented in Table 2.

A total of 36.1% of the variance was explained in the regression on information commenting. Information motives, social interaction motives, community motives, relatedness, and competence were predicted to influence information commenting in Hypothesis 6a. Although informational motives, relatedness, and competence were not found to be significant predictors, social interaction motives (β = .385, p < .001) and community motives (β = .218, p < .001) were. Hypothesis 6a is partially supported, with two of the five predicted relationships found. In addition, perceived accuracy was also a (near) significant predictor (β = −.138, p = .051), such that less perceived accuracy was related to more information interaction.

Hypotheses 6b posited the same set of predictors for information sharing. The regression explained 41.6% of the variance. Information motives (β = .342, p < .001) and community motives (β = .188, p < .05) were, indeed, significant predictors; however, social interaction motives, relatedness, and competence were not. Hypothesis 6b is also partially supported, albeit with different partial support among the relationships posed. Interestingly, autonomy was a significant negative predictor (β = −.169, p < .05), such that less autonomy experienced was related to greater information sharing.

Discussion

The results of the present study produced interesting differences across live and on live videos and explained significant amounts of variance in perceived learning and information behaviors. After discussing several key takeaways from the results, this discussion will focus on theoretical implications, practical implications, and study limitations.

First, motivations to view and psychological needs met by viewing behaved very differently in this data set. All four motivations explored here were greater for nonlive videos than for live videos, with the largest difference observed for informational motives. On the other hand, two of the three psychological needs (relatedness and autonomy) were evaluated as being better met by viewing live streaming videos than nonlive videos. Both motivations and psychological needs predicted greater perceived learning, although only motivations positively predicted information-interacting and information-sharing behaviors. Perceived learning, like motivations, was reported as greater for nonlive videos. The data also demonstrated a robust model predicting variance in perceived learning, where informational motives, relatedness, competence, and the control variables of perceived credibility and accuracy were all significant predictors. Last, both information behaviors were reportedly engaged in to a greater extent during viewing nonlive videos and compared to live streaming, although information interacting and sharing were both influenced by largely different factors.

The impetus for this study was news and industry reports about young individuals turning to SMV content for information at a rate seemingly outpacing other digital sources. We were particularly interested in informational motivations and perceived learning and how the affordances of live streaming videos may factor into these critical concepts. Informational motives were indeed greater for nonlive viewing as predicted as was perceived learning. These results may be due to our rationales behind these hypotheses: The possible addition of video visuals and effects offered a redundant, secondary stream of information previously found to increase learning and memory outcomes (Fox, 2004; Lang, 1995) or they may have been viewed as more professional due to the editing and effects and therefore deemed more credible. One additional affordance that we did not fully expound on previously may also be a contributing factor to these outcomes, as well as relevant to the psychological need of autonomy. That factor is the relative permanence of nonlive videos and the agency a viewer has to stop, start over, rewatch. One could imagine that it is easier to learn something from content that can be readily available to review a second time. This relative permanence and viewer agency in the viewing experience may also explain why information sharing was greater in response to nonlive videos. Physically copying a link to a video that will be there tomorrow, that is short enough to garner interest, is likely more common in nonlive videos. Sharing a live stream may be more difficult due to its timing and mode of delivery.

Theoretical Contributions

The differences seen between motivations being greater for nonlive video viewing and psychological needs largely being better met for live video viewing speaks to their theoretical distinctiveness, as well as the value in utilizing two different theoretical approaches to better understand the individual-level experience of viewing LSSMV and nonlive SMVs. This difference also highlights the differences in what individuals may seek before going into a viewing experience and what they report actually experiencing after the fact. Future research using a uses and gratifications perspective could explore both gratifications sought and obtained to further explore these differences in expectations and experiences.

That we saw any differences at all as a result of LSSMVs versus nonlive SMVs speaks to the appropriateness of exploring the differences in affordances each offer. We were not right about how these affordances would affect all variables: Of note, we predicted that social and community motives, along with information interacting, would be greater during live streaming due to the synchronous nature of communication. We had posited that the immediacy of responses and the opportunity to interact with other viewers and content creators would elicit more interaction. Live streaming videos did, however, better meet relatedness needs, as it was posited (with the same rationale). It may be that individuals who are consciously seeking outlets for social and community reasons choose a different mediated (or nonmediated) experience all together, such as texting a friend, scrolling social media, and so forth. But, the interactions afforded by live streaming chats did meet a relatedness need to a greater extent than nonlive viewing. This line of thinking would suggest that the degree of interaction offered in different mediated communication environments exists along a continuum, and individuals negotiate these affordances with their own motives and needs.

This work also contributes to an expanded view of outcome variables associated with psychological needs being met within the SDT literature. Many applications of SDT in media studies focus on psychological needs predicting enjoyment, future engagement with the media being studied, or future behaviors (Berezan et al., 2018; Montag et al., 2021; Ryan et al., 2006), while this study convincingly found relatedness and competency needs to influence perceived learning. It is logical that once basic personal needs are met, we are better equipped to learn and integrate new information or at least perceive ourselves as more able.

One last theoretical insight into the findings suggests that Diwanji et al. (2020) and information sciences, generally, are correct in clearly distinguishing information behaviors rather than grouping them all together. While information sharing and commenting were both found to be greater when viewing nonlive videos, both information behaviors were subjected to mostly different sets of influences as observed in the regression analyses. Future research should further explore the relationships between motivations and information behaviors.

Limitations

This study is not without its limitations. While the study wanted to broadly explore live streaming versus nonlive streaming video, it did not attempt to measure or control other message characteristics, like genre, platform, or type of content, that we know also influence the variables studied here. Second, as a survey study, this data cannot speak to causal relationships. Future experimental work could do so, while also correcting an additional limitation of this study. The present study assumes nonlive videos contain more video visuals and editing effects, although we did not ask participants to refer specifically to a video that did. Accounting for these effects and visuals should be the next step in an experiment in this line of research. Future research should also look at populations other than 18- to 24-year-olds. Although this is where the bulk of the research currently exists, exploring how different demographic groups interact with information and learn from SMVs is an important next step.

Practical Implications

While future experimental research could isolate more content-specific takeaways for content creators, the present study offers several relevant bits of information for content creators and viewers. First, viewers report greater informational, entertainment, social, and community motivations, along with greater perceived learning, information sharing, and information commenting for nonlive SMVs as compared to live streaming videos. This should indicate, at least for now, that nonlive videos should not be ignored or abandoned, despite how much interaction live streaming videos may engender. In fact, if content creators want viewers to learn information from their content, perhaps posting nonlive video versions of information they create during live streaming sessions—even if they are just highlight reels—may encourage further interaction with that content and greater knowledge gain. Live videos are not being viewed as often or for as long as nonlive videos among this younger age group at the moment, although we know these numbers are increasing as time goes on (Ceci, 2022). We know that once one gets viewers watching a live stream, that experience better meets relatedness needs (or feelings of togetherness or connectedness) and autonomy needs (feelings of agency or control)—so live streaming is certainly not something to be ignored. This may suggest that live streaming is especially worth it if you can connect with viewers and make them feel empowered.

As viewers of live streaming content, this research would suggest that one keeps an open mind about what can be gained from the experience. While individuals are more likely to seek out nonlive videos for social and community motivations, their relatedness needs are better met by live video viewing experiences. The data would further suggest that both one’s motivations (i.e., learning motives) and experiences with the content (i.e., perceived credibility, accuracy) contribute to perceived learning. So, maximizing one’s motivations by being conscious of them when making a media choice, paired with reliable, credible content would be an ideal situation for learning new information.

Copyright © 2023 The Author(s)

Received February 2, 2023
Revision received June 8, 2023
Accepted June 9, 2023