The embedding of the idea of “Smart” in the term Smart Classroom is itself intriguing and most likely arises from joint usage of smart technology. The “Smart Classroom,” as a technology capable of affecting classroom interaction, usually refers to a physical classroom that integrates advanced forms of educational technology to increase the instructors’ ability to facilitate students’ learning and their ability to participate in learning experiences beyond the possibilities of traditional classrooms [17,23,24]. The educational technology includes flexible hardware devices (such as computers, mobile terminals, electronic whiteboards, presentation equipment, and activity desks and chairs) and more interactive software with artificial intelligence (such as online interactive platforms, learning management systems, face recognition, and emotional recognition) [25].

From a theoretical perspective, the Smart Classroom architecture and philosophies are based on constructivist learning theories [26], which emphasize students’ self-development via social interactions [27] (pp. 1–14). As a result, the term “Smart” may thus refer to the ability to maintain a high degree of classroom involvement throughout the learning process, which emphasizes the exploration of interactions as an important point in studies of this phenomenon. According to Yau et al. (2003) [28], Yu et al. (2022) [29], and Yuan (2022) [30], Smart Classrooms are intended to improve teacher–student interactions to thus improve the teaching and learning experience. Based on constructivism, Garrison (2007) [31] proposed a Community of Inquiry (CoI) model to understand and guide online learning experiences. He believes that community cooperative learning and interaction, including social, cognitive, and teaching, are of great significance for students’ advanced learning. Therefore, technology can promote teaching and student learning in these aspects based on the CoI model. Huang et al. (2012) [32] also proposed a “SMART” concept model with the five dimensions of “Showing”, “Managing”, “Accessing”, “Real-time feedback”, and “Testing”, as a framework for the impact of technology on teaching and learning. Researchers further think that Smart Classrooms may improve the presentation of educational information, make it easier for students to acquire learning resources, encourage engagement in classes [17], help teachers assess student learning [32] and take charge of in-classroom teaching [33], and provide better interactions and better physical environments [34].

The term classroom interaction is generally understood to mean the communication between teacher and students and between students themselves [35] (pp. 3–12). It is believed that quality interaction can enhance the classroom atmosphere, as well as promote students’ learning behaviours and engagement, and thus improve the quality of classroom teaching [29,36]. Researches have identified the influence of both a single technology and Smart Classroom on classroom interaction quality.

Some previous research has focused on the influence of interactive whiteboards (IWBs), with evidence from studies by Smith et al. (2006) [37], Manny-Ikan et al. (2011) [38], and Hall and Higgins (2005) [39] indicating that the versatility and multimedia functions of IWBs and the “theatrical tension” that they bring to the classroom help to attract students and increase their interest and engagement. Technology can be used to improve the interactions between the instructor and the students, or in-group collaboration among the students [28]. Some studies, however, also indicated that such student participation was short-lived and that the advantages of IWBs were lost, where the class lacked higher-order thinking skills [38]. Knowledge and understanding of technology were also seen to affect teachers’ use of IWBs and, consequently, their confidence in teaching [40], suggesting that the quality of the resulting interactions may also be affected. Raman et al. (2014) [41] pointed out that IWB acceptance among teachers or students further affects the quality of teaching, while Glover, Miller, and Averis (2007) [42] noted that the interactive function of IWBs was maximized to differing extents based on teachers’ personal technical and pedagogic fluency.

Other research has focused on personal response systems (PRS) and group response systems (GRS), such as clickers, with the findings suggesting that these help to break up traditional lecture models by promoting learner-centred active learning and to increase student participation and student–teacher interaction by removing students’ fear of public mistakes or embarrassment [43,44,45,46]. And, it was found that, with the use of tablet PCs, response systems can improve individual students’ participation and interaction in various group sizes [47]. As a result, it is becoming accepted that technological interventions can have an influence on classroom engagement, though the effect is often associated with the degree of involvement.

It is critical to understand that it is inappropriate to examine the impact on interaction in new Smart Classrooms by separately examining the impact of some single technology, as He and Li (2009) [48] (p. 103) argue that the development of educational technology is cumulative, with previous generations of technology coexisting with new generations. A Smart Classroom is a comprehensive technical system where the function of the whole can exceed that of the sum of its parts. In this sense, Smart Classrooms must be seen as multilevel technology systems containing co-existing generations of technology: these commonly include the simple multimedia environment represented by projection, the interactive multimedia environment represented by the IWB, and the interactive teaching system represented by intelligent terminal technology [49].

Some researchers have suggested that the use of full Smart Classrooms can promote interaction in primary and secondary schools. Wang et al. (2016) [50] analysed 54 English classes in Beijing, Shanghai, and Shenzhen using an interactive analysis scale, finding that Smart Classrooms supported classroom interactions, improving its frequency and enriching its content. The use of Smart Classrooms also enhances the dynamic, effective, and harmonious interaction between teachers and students, thus enabling technology to reinforce the impact of students’ involvement in learning and intelligence with regard to academic performance [16,18,51]. Some scholars also consider that smart technologies may help to improve interactions, with which teachers will be able to choose more suitable ways of teaching (i.e., online, face-to-face, or blended) to meet different types of needs [17], and to extend the limitations of time and space for learning [52]. In a Smart Classroom, students have more opportunities to explore, create, display, and evaluate with the support of smart technology. Teachers can also use technology to present content, detect students’ learning statuses, diagnose the teaching process, and adjust their teaching method in a timely manner [53].

The impact of Smart Classrooms on interaction may, however, differ at different stages of education. The use of Smart Classrooms has led to a significant increase in teacher–student interactions at the K12 level, helping to improve the quality and efficiency of teacher–student interactions [54]. At the primary school level, Jo and Lim (2015) [55] compared the interaction within two lessons in South Korean and found that lessons in Smart Classrooms involved more indirect teaching, a higher question ratio, and less lecture-style teaching. However, it is questionable whether interactions at the university level are positively influenced by Smart Classroom use, as in comparison with research carried out at primary and secondary schools, there are much fewer studies on the influence of Smart Classrooms on interaction at the university level. Chen, Chang, and Chien (2015) [56] used the “Speech-Driven PowerPoint” (SDPPT) system to enhance interactions at a Taiwanese university, determining that student enjoyment and motivation increased with such use, and Jiang et al. (2018) [57] showed that the amount of classroom interaction in Smart Classrooms in mainland China generally increased, although the levels of technology used by teachers were quite different.

However, there is some controversy. Li, Liang, and Xue (2018) [58] suggested that Smart Classrooms did not significantly improve class interaction and that such technology was mainly used to support teacher-centred teaching. Furthermore, although the existing literature provides some overview of the interactions facilitated through the use of technology, the specific relationship between Smart Classrooms and interaction quality in university classrooms remains unknown, and any discussion of the impact of layered technology systems on interaction is thus lacking.

It may be difficult to quantify the classroom interaction quality, and the only way is through some kind of standardized observation method [59]. Hence, the development of classroom observation instruments, such as an observation framework and scale, may be required to measure the quality of interactions.

A great deal of research has gone into developing instruments to analyse classroom interactions based on different criteria for “quality interactions”. There are mainly two kinds of interactions in educational relationships, defined as student–teacher (ST) and student–student (SS). An analysis of ST interactions tends to focus on evaluating the quality of teachers’ influence on and support of students. Flanders (1963) [60] assumed that teachers typically exerted their influence on students by means of verbal statements and thus proposed the “Flanders System of Interaction Analysis” (FSIA) to analyse both teacher talk and student talk. Pianta, Hamre, and Allen (2012) [61] instead divided teacher support into the domains of emotional support, classroom organization, and instrument support, thus developing the “Classroom Assessment Scoring System” (CLASS), which includes 11 dimensions based on the various domains to evaluate the effectiveness of interactions between teachers and children. As Johnson (1981) [62] noted, however, in addition to ST interactions, SS interactions are also necessary for students’ achievement, socialization, and healthy development. Kumpulainen and Wray (2001) [63] thus proposed the “Analytical Framework of Peer-group Interaction” (AFPI) to analyse SS interactions from sociocultural and sociocognitive perspectives. Hillman, Willis, and Gunawardena’s (1994) [64] research incorporated the learner–interface interaction into the instrument used to analyse interaction, while Gu and Wang (2004) [65] adapted FSIA by adding items to increase the focus on students’ behaviour and student–technology interactions, thus proposing the “Information Technology-based Interaction Analysis System” (ITIAS) to analyse interactions in the classroom in a manner integrated with information and communications technology (ICT) use. Mu and Zuo (2015) [66] similarly took ICT into consideration when proposing the “Teaching Behavior Analysis System” (TBAS), which was used to observe teacher and student behaviours, ST interactions, and the use of media in class. Wang et al. (2016) [50] similarly developed the “Smart Classroom-based Interaction Analysis System” (SCIAS) for primary and secondary schools, which includes an analysis of basic information, classroom facilities, and interactive processes.

Other research has focused on constructing an analysis instrument to consider interactions from the perspective of the learning process. Henri (1992) [67] proposed an analytical model that emphasized five dimensions of the learning process in a computer-mediated communications (CMC) environment: these were participation, interaction, social, cognitive, and metacognitive. After examining this model and other studies, Sing and Khine (2006) [68] then concluded that the most commonly used interaction dimensions were participation, cognitive processing, and social interaction.

Due to the variety of different contexts in which these instruments were developed, they cannot be used to analyse the quality of interactions in university Smart Classrooms without alteration, however. FSIA and CLASS do not include sufficient indicators of SS interaction, as they are mainly aimed at evaluating ST interactions, while CLASS is most appropriate for early childhood education. Similarly, the instruments proposed by Henri, and Sing and Khine are predominantly used to evaluate interactions in distance education, rather than prioritizing face-to-face interactions, while ITIAS, TBAS, and SCIAS are largely used in primary and secondary schools and also do not focus on the quality of interactions. As a result, based on the research above, a novel framework adapted to university classroom interaction assessment needs to be developed.