The COVID-19 pandemic caused the great virtual pivot, moving learning to a digital platform (Kelly, 2023). Despite the world's return to many pre-COVID conditions, education is not anticipated to revert to a primarily face-to-face experience. In fact, the National Center for Education Statistics (2023a) indicates that more than half of students enrolled in undergraduate higher education are having virtual learning experiences regardless of whether they have enrolled in a face-to-face, hybrid, or online education program. Zoom and other video conferencing software have become a normal mode for synchronous and asynchronous (i.e., pre-recorded) virtual instruction (Yang et al., 2023).

When instructors use video conferencing software for teaching, most guidance recommends use of a virtual background to obscure the environment around them for a number of reasons (Barchas, 2020; Ohnigian et al., 2021; Chen et al., 2023). First, the utilization of virtual backgrounds is an equalizer, placing everyone in the same virtual world rather than homes of varying levels of socio-economic status (SES; Chen et al., 2023). Second, virtual backgrounds help instructors maintain credibility by hiding messiness and external intrusion in their physical backgrounds (Barchas, 2020). Finally, using a virtual background simply keeps one's home life private, creating a barrier between school and home even when school happens at home (Ohnigian et al., 2021). Thus, instructors are encouraged to use a virtual background for face-saving and security reasons.

Despite the general consensus among scholars that virtual backgrounds are useful for education, very little research exists on how to optimize instructors' use of these backgrounds (c.f., Seitz et al., 2023). Though a review of the literature revealed many thought pieces on the use of virtual backgrounds in education, only three research studies were found. Kelly et al. (2023) conducted a virtual learning experiment where an instructor compared use of a professional background (i.e., university branded background) and personal background options; the study found that female students rated instructors as credible and had high perceived cognitive learning regardless of the background choice, but male students most consistently perceived higher cognitive learning and professor credibility with the professional, university branded background. In a pandemic education investigation, Goethe et al. (2022) found that some students become distracted by instructors' virtual backgrounds and in these cases prefer to see the instructors' real environment instead of a virtual background. Finally, Kelly et al. (2024) found that male students experienced less affective learning than female students when online instructors used a blurry virtual background filter compared to an uploaded virtual background or allowed students to see their actual environment.

In a holistic review of research on virtual background use, Seitz et al. (2023) identified only two additional studies on the impact of virtual backgrounds on communicators outside of the education setting. In a workplace meeting study, Palanica and Fossat (2022) found that using virtual backgrounds with nature scenes stimulated more creativity than backgrounds using urban scenes. The second study was a pilot test of the FocalSpace video conferencing software, which found that blurry backgrounds can be less distracting for users than the authentic background (Yao et al., 2013). This research was somewhat contradicted by Kelly et al. (2024) in the educational context, who found male learners responded negatively to instructors' use of blurry background filters.

Together, these studies indicate that communicators pay attention to virtual backgrounds when using video conferencing software and have perceptual responses to them. Yet, the existing literature is not enough to generate evidence-based best practices for online teaching. Despite the arguments for why using a virtual background is a best practice (Barchas, 2020; Ohnigian et al., 2021; Chen et al., 2023), there is still uncertainty about which backgrounds are the best choices for learning (c.f., Goethe et al., 2022; Kelly et al., 2023, 2024). To address this gap, this study aims to advance the extant literature by examining the how perceived appropriateness and enjoyment of an instructor's virtual background affects learning.

Social presence theory suggests that individuals have an innate need for social interaction and connection, even in mediated environments (Short et al., 1976). According to social presence theory, individuals seek cues in mediated communication, such as video conferencing, that enhance the sense of immediacy and warmth, making the interaction feel more personal and engaging. Further, the theory suggests that emotional engagement plays a crucial role in fostering social presence and connection in mediated communication (Short et al., 1976).

People experience strong social presence when they perceive that the technology chosen is appropriate for the conversation (Rice, 1993; Biocca et al., 2003). Also, when perceiving strong social presence, mediated communicative experiences become more enjoyable (Lee et al., 2013; Wang and Lee, 2020). As such, enjoyment and appropriateness of technology can contribute to that desired sense of warmth and immediacy individuals seek when communicating through technology (Short et al., 1976). Therefore, social presence theory would anticipate that students will have more positive social responses to instructor's backgrounds in online learning environments if they find the background to be appropriate and enjoyable.

Yet, what is unclear is whether that positive affective response will translate to learning. While Kelly et al. (2023) found that male students perceived that they learned more when a university branded, professional background was used by their instructors, it is unclear whether they actually did learn more. To address this issue, the present study investigates how instructor choices of virtual backgrounds may affect student learning, making a distinction between perceived learning and actual learning, by raising the following research question.

RQ1: To what extent does enjoyment and perceived appropriateness of the instructor's virtual background affect learning (i.e., perceived cognitive learning and actual cognitive learning)?

The extant literature notes that female and male students tend to respond differently to the way instructors utilize technology in classroom learning. Schrodt and Turman (2005) found that female students perceive instructors who infused a moderate amount of technology into their teaching practices to be more competent than instructors who used no technology, while male students found instructors to be most competent when they used no technology. More germane to the present study's context, Kelly et al. (2023, 2024) noted that male and female students respond differently to instructors' virtual background choices. As such, to further address the potential effects of student sex in learning experiences, the following research question is raised:

RQ2: To what extent does learner sex interact with virtual background choice to predict learning (i.e., perceived cognitive learning and actual cognitive learning)?

The purpose of this study was to better understand the roles that enjoyment and perceived appropriateness of an instructor's virtual background play in student learning. The experimental design required students to watch a 2-min lecture on one-way interviews with one of three virtual background options varying in professionalism and authenticity to assess students' learning experiences when all other conditions are held constant. Overall, the results indicate that students' enjoyment and perceived appropriateness of the background do not influence whether they learn, but they do influence whether students think that they learn.

First, the study finds that both enjoyment and perceived appropriateness of the background were statistically significant predictors of whether students thought that they learned. However, neither variable predicted how students actually performed on the quiz. This finding is potentially explained by the halo effect (Thorndike, 1920). The halo effect explains that individuals associate positive attributes to people and things that they like more easily than those they dislike. Thus, students who found the background enjoyable and appropriate are more likely to think they had a good learning experience regardless of whether they actually learned.

Another possible explanation for the finding lies in social presence theory. Part of the fundamental tenet of social presence theory implies that more enjoyable and appropriate uses of technology will elicit higher social presence (Short et al., 1976), and this experience of perceiving another person to be socially present elicits changes in perception (Biocca et al., 2003). As such, perceived cognitive learning increase may be an indicator of positive perceptual change prompted by the experience of social presence. Perception and reality are not always unified, so it is possible that higher social presence may have caused learners to feel that they were learning more though the same amount of learning took place across learners. Additional research is needed to confirm this supposition.

A supplemental analysis examined students' learning on the easier version of the quiz, using only the three most simplistic items to assess learning. While enjoyment of the background still did not predict students' performance of the easier quiz, perceived appropriateness did. It may be that with backgrounds that students found to be more appropriate, they were less distracted, and therefore more able to capture some of the bigger, repeated ideas during the lecture, which can be easily demonstrated in simple questions. This would be in alignment with the findings of Yao et al. (2013) that learners prefer virtual backgrounds that are not distracting. If this is the case then it is possible that when a higher level of thinking or problem solution skills is needed through a difficult quiz question, it may require more cognitive ability from students beyond what the instructor repeatedly conveyed in the lecture. More studies are required to test this conjecture.

Yet, the finding that appropriateness of the background did predict students' performance must be considered in terms of the magnitude of the effect. Perceived appropriateness only explained 1% of the total variance in how students performed on the easier version of the quiz. This means that students perceiving the background to be appropriate only made a difference in 1% of the variability in whether they were able to recall the most simplistic, high-level concepts overviewed in the lecture within minutes of viewing it. While this study had ample statistical power to make this finding statistically significant, a variance control of 1% is not a substantive impact.

Further, recall, the type of learning assessed in this study is the lowest possible level of learning (Chandio et al., 2021). Students' ability to recall information does not indicate that they can create, evaluate, analyze, or apply the information or that they even understand the information. So, at best, students finding a virtual background to be appropriate can affect 1% of the variance in a student's ability to recall the most basic information from a lecture, which is a negligible impact on the grand scale of learning.

An important question to address is whether considering students' perceived appropriateness and enjoyment matters when instructors choose their virtual background. Certainly, instructors must make pedagogical decisions based on what actually impacts learning rather than whether students feel like they learn (Rudick, 2024). Yet, socioemotional aspects of the classroom are still important for overall satisfaction in the class (Yang et al., 2023). As such, additional research is needed to see whether appropriateness and enjoyment of virtual backgrounds can affect students' satisfaction in a course overall to know whether being concerned about such influence is truly impactful to the learning environment.

In addition to exploring whether enjoyment and appropriateness of instructors' virtual backgrounds affect learning, this study also sought to identify whether learner sex interacted with choices of virtual backgrounds to predict learning as observed in a study of perceived cognitive learning by Kelly et al. (2023). Again, the results of this present study indicate that actual learning did not differ according to learner sex or background choice. Background choice also did not affect how much students thought that they learned. However, female students had higher perceived cognitive learning than male students. Notably, the Levene test was very close to statistical significance for that analysis, raising the probability of Type I Error. That this analysis replicated the findings of Kelly et al. (2023) lends more credibility to the finding, but the results should be accepted with some caution.

As Rudick (2024) points out, perceived and actual cognitive learning are not often equivalent. It would be interesting to know if the Kelly et al. (2023) study design were replicated with the addition of an actual cognitive learning measure whether they would have found that the difference in gender was limited to perception. Female students tend to have better performance and persistence than male students at the college level (National Center for Education Statistics, 2023b). As such, female students in these studies may perceive that they learned more because they are accustomed to learning and performing well, even though there was no difference in actual learning between male and female students in the present study. Future research is needed to better explain this finding.

In the end, the results of this study indicate that whether students enjoy an instructor's virtual background or think it is appropriate does not actually affect whether they learn, but it does affect whether they feel like they have learned. Future research may find other socioemotional classroom variables that justify instructors being concerned with the perceived appropriateness and enjoyment of their virtual backgrounds. The results of this study indicate that instructors cannot actually negatively affect students' cognitive learning based on their choice of virtual background though.

The datasets presented in this article are not readily available due to confidentiality but data can be shared by the authors upon reasonable request. Requests to access the datasets should be directed to sekelly@ncat.edu.

The studies involving humans were approved by University of Central Florida Internal Review Board. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

SK: Formal analysis, Methodology, Writing – original draft, Writing – review & editing. JK: Conceptualization, Data curation, Methodology, Writing – review & editing. JC: Conceptualization, Methodology, Writing – review & editing.