View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/D2RP00095D (Paper) Chem. Educ. Res. Pract., 2022, 23, 898-912

How teacher enthusiasm affects students’ learning of chemistry declarative knowledge in video lectures

Qian Huangfu *a, Hong Li a, Sanshan Tang a, Jianrong Wang b, Qian Liu c and Guojun Chen d
aSouthwest University, Chongqing, 430079, China. E-mail: chemqian16@swu.edu.cn
bGuanDu Experimental School, Yunnan Province 650000, China
cChengdu Shude Xiejin High School, Sichuan Province 610000, China
dChongqing No. 1 Secondary School, Chongqing, 430079, China

Received 6th April 2022 , Accepted 15th June 2022

First published on 20th June 2022

Abstract

Although the chemical literature contains many studies of multimedia-based learning and teacher enthusiasm, there is a paucity of research on whether and to what extent teacher enthusiasm in video lectures affects students’ learning, especially in chemistry. In this context, this mixed-method study used eye tracking and quantitative analysis to investigate how a teacher with different levels of enthusiasm influenced students’ learning in video lectures. Junior middle-school students were selected to engage in this eye-tracking research. We set up 35 such students as a group to view an experimental video with a low level of teacher enthusiasm, and 35 others as another group to view another experimental video with a high level of teacher enthusiasm. The essential tool for capturing the students’ visual attention was an EyeLink 1000 Plus eye tracker. The total dwell time, fixation counts, average fixation duration, and transition counts were recorded and analyzed, and the results showed that the teacher enthusiasm in the video lectures had an indirect positive effect on the students’ self-efficacy and learning performance and was negatively associated with cognitive load. In addition, students paid more attention to the teacher with the higher level of enthusiasm.

Introduction

Educational technology has changed how we teach and learn, and as a preferred learning resource for learners, video lectures (VLs) have gained much attention (Tse et al., 2019). VLs are aimed at offering high-quality resources to enrich students’ experiences (Xie and Ke, 2011). Based on the cognitive theory of multimedia learning and cognitive psychology, many scholars have discussed how the mutual interaction of the presentation elements of teaching videos influences students’ learning, such as (i) whether there are subtitles (Grimmer, 1992; Markham et al., 2001), (ii) whether to show the image of teachers (Ozcelik et al., 2010), (iii) the proportion of teachers’ heads in VLs (Van Gog et al., 2014), and (iv) physical cues in videos (Mautone and Mayer, 2001), but they seldom focus on the teachers themselves.

Teacher enthusiasm (TE) has usually been regarded as an effective way for teachers to deliver teaching information to learners (Shuell, 1996) and a crucial feature of high-quality effective learning (Kunter et al., 2013). TE can be effective for enhancing students’ learning interest (Kunter et al., 2008), motivation (Allen et al., 2006; Frenzel et al., 2009), and outcome (Brophy and Good, 1986). To date, many studies have investigated the definition, dimensions, and role of TE in classroom instruction (Collins, 1978; Brophy and Good, 1986), but few have explored whether TE can play a similar role in video teaching as it does in traditional classroom teaching. In previous studies, some of the learning content presented in video teaching was declarative knowledge (DK) while some was procedural knowledge (PK), and certain empirical studies have shown that the effects of teaching approach on DK and PK in video teaching are different (Hoffler and Leutner, 2007; Fabio and Antonietti, 2012).

The present study used DK of chemistry as the teaching content and collected data through eye movement to explore how TE influenced students’ cognitive load (CL), attention, learning performance (LP), and self-efficacy (SE). The aim was to reveal how different levels of TE affect learners in the process of multimedia learning and provide a theoretical basis for producers of instructional videos.

Theoretical framework

Knowledge types and video teaching

Anderson (1976) divided knowledge into declarative and procedural as different forms of knowledge: DK is knowledge of what is, i.e., knowledge of concepts and facts; PK is how-to knowledge, i.e., knowledge of specific skills and processes in a field. According to knowledge classification theory, chemical knowledge can be divided into chemical DK and chemical PK. Chemical DK refers to the knowledge about chemical facts, concepts, and principles, which reflects the essential properties and internal laws of substances and their changes, such as the properties of elemental compounds, chemical phenomena and facts, and the basic laws of chemical reactions. Chemical PK is the application and calculation of concepts, principles, and rules, such as identifying categories of substances and complexes, naming organic compounds, calculating formula quantities, etc., or chemical experimental skills according to the relevant principles, such as preparing gases, purifying materials, synthesizing organic matter, etc.

Fabio and Antonietti (2012) compared the effects of students learning PK and DK in video teaching and traditional teaching; they found that the effects of video teaching of both types of knowledge were better than those of traditional teaching, but the effects were not completely the same, with the learning effects of DK being better. In addition, some studies have explored whether seeing the teacher's face affects students’ academic performance in the process of learning PK and DK in VLs. Van Gog et al. (2014) used a PK “Frog Leap” demonstration as the learning material, while Van Wermeskerken and Van Gog (2017) used a DK molecular structure model to build knowledge-point video teaching, carried out a comparative study of video teaching with or without teachers’ images, and added teachers’ eye guidance to video demonstrations with teachers; the results of Van Gog et al. (2014) showed that learners with teachers were significantly better than learners without teachers, while those of Van Wermeskerken and Van Gog (2017) showed that there was no significant difference in students’ academic performance with or without a teacher.

Therefore, we speculate that the influence of TE on students’ learning in video teaching may produce different results for different types of knowledge. In the present study, DK was selected so that the teacher was continuously present in the teaching video.

Teacher enthusiasm

Collins (1978) was the first to propose TE, laying the foundation for future study. TE can be characterized by trait-like, habitual, recurring emotion (Kunter et al., 2008). Some researchers have found the enthusiasm that teachers display is stable and idiosyncratic (Keller et al., 2018). The generally accepted definition of TE is the degree of enjoyment, excitement, and pleasure that a teacher typically experiences in their professional activities (Kunter et al., 2008), and TE can be divided into two facets: one is lively and engaging teaching, which we call experienced enthusiasm (Frenzel et al., 2009; Keller et al., 2016; Lazaridesa et al., 2018); the other is the teacher's behavioral expressions encompassing both verbal and nonverbal behaviors, which we call displayed enthusiasm (Feldman, 2007; Kunter et al., 2011). Even though they do not necessarily occur simultaneously in one individual (Keller et al., 2016), they both offer strong support to students and help to motivate them (Kunter et al., 2013).

The purpose of the present study was to determine whether and how TE in chemistry VLs is related to students’ learning, and so the TE in this study was conveyed through nonverbal behaviors such as rich facial expressions, frequent gestures, varied intonation, and even humor. Numerous studies have shown that there is a strong relationship between (i) a sender's verbal and nonverbal cues of instructional communications and (ii) the emotional responses of the receivers, with the receivers being more likely to be affected cognitively and emotionally by the sender's nonverbal messages (Mehrabian, 1972; Sueyoshi and Hardison, 2005; Batty, 2020). Liew et al. (2017) showed that pedagogical object enthusiasm (including smiling, nodding frequently, a high level of animated gestures, and speaking enthusiastically) significantly enhanced students’ positive emotions, and Arguedas and Daradoumis (2021) showed that affective pedagogical tutor feedback could reduce students’ negative emotions such as boredom and anger; these findings are consistent with emotional response theory (Mottet et al., 2006). Meanwhile, past research results show that the facilitating effects of agent enthusiasm on intrinsic motivation and cognitive results are fully mediated by a learner's positive emotions (Liew et al., 2017). A pedagogical agent who performs both gestures and facial expressions leads to better retention (Schneider et al., 2021), and the instructor's facial expressions influence students’ medium-term recall (Wang et al., 2019). The present study is similar to those of Keller et al. (2016) and Murray (2007), i.e., we focus on displayed TE instead of experienced TE.

TE benefits various student outcomes, such as LP (Keller et al., 2013; Burić, 2019), SE (Tuan et al., 2005), CL (Liew et al., 2017), visual attention (VA) (Chen and Wu, 2015), intrinsic motivation (Patrick et al., 2000; Burić, 2019), and interests (Kim and Schallert, 2014), among others. More concretely, TE is always regarded as an indispensable part of effective teaching and has excellent effects on students’ LP (Becker et al., 2014). The findings of the study by Rosenshine (1970) reflect the substantial empirical support for the positive impact of TE on students’ performance. Keller et al. (2016) used concept maps and paper-and-pencil tests to examine the relationship between TE and students’ LP and obtained the same result as that of Rosenshine (1970), i.e., that there is indeed a significant positive relationship between them. Later, Burić (2019) found that TE fully mediated the relationship between the emotional labor strategy of hiding feelings and students’ outcomes, SE, and intrinsic motivation. Those findings matched those of Patrick et al. (2000), who found that students had greater SE to learn the lecture material and had a higher level of vitality after watching enthusiastic VLs.

Collins (1978) constructed the first reliable criteria for measuring TE, her original intention being to develop TE to be evaluated by other professional observers. Subsequently, many scholars developed new instructions for measuring TE that were adapted from the work of Collins (1978). These instructions are used to assess displayed TE via students’ perceptions, and even though the validity of students’ perceptions is still disputable, many types of research have shown that students’ overall perceptions of instruction provide a reliable measure of classroom processes (Trautwein et al., 2006). Table 1 presents several tools for assessing TE, and for the present study, we chose the one by Murray (1983) to evaluate the level of TE in chemistry VLs, the main reason being that compared with the other tools, the items associated with this one are more suitable for evaluating displayed TE (Frenzel et al., 2009).

Table 1 Assessment tools for evaluating displayed teacher enthusiasm
Instrument Scale specifics Item wording
Frenzel et al. (2009) Four items, five-point rating scale from (1) strongly disagree to (5) strongly agree; α = 0.85 • Our teacher teaches with enthusiasm
• Our teacher is humorous during teaching
• Our teacher tries to get students excited about the subject of mathematics
• Our teacher seems to take pleasure in teaching
Kunter et al. (2008) Three items, four-point rating scale from (1) strongly disagree to (4) strongly agree •Our teacher seems to really enjoy teaching
•Our teacher is an enthusiastic teacher
•Our teacher is enthusiastic about his/her subject
Feldman (2007) Not reported • The instructor shows interest and enthusiasm in the subject
• The instructor seems to enjoy teaching
• The teacher communicates a genuine desire to teach students
• The instructor never showed boredom for teaching this class
• The instructor shows energy and excitement
Patrick et al. (2000) Four items, seven-point rating scale from (1) strongly disagree to (7) strongly agree; α = 0.93 • The teacher just lights up the room when he/she teaches
• The teacher is a bit dull
• The teacher has a contagious energy about him/her
• The teacher is full of dynamic energy when he/she teaches
Marsh (IDEA; 1994) Three items, five-point rating scale from (1) hardly ever to (5) almost always. Item description: “Describe the frequency of your instructor's teaching procedures.” • Enthusiastic about the subject
• Spoke with expressiveness
• Dry and dull presentations
Murray (1983) 11 items, five-point rating scale rating frequency of occurrence from (1) almost never to (5) almost always • Uses humor
• Speaks expressively or emphatically
• Shows facial expressions
• Moves about while lecturing
• Reads lecture verbatim from notes
• Shows energy and excitement
• Smiles or laughs
• Gestures with hands and arms
• Shows strong interest in subject
• Avoids eye contact with students
•Speaks softly
Marsh and Ware (1982) Three items; five-point rating scale • Was enthusiastic about the subject
• Has a good sense of humor
• Made learning enjoyable
Marsh (1982) Four items, five-point rating scale from (1) very poor to (5) Very good. Item description: “As a description of this course/instructor, this statement is…” • Instructor was enthusiastic about teaching the course
• Instructor was dynamic and energetic in conducting the course
• Instructor enhanced presentation with the use of humor
• Instructor's style of presentation held your interest during class

Teacher enthusiasm and self-efficacy

SE refers to an individual's judgment of their abilities to organize and execute courses of action required to achieve desired performances (Bandura, 1997). For Bandura (1982), SE is amechanism that activates performance, effort, attention, and persistence concerning situational demands. Previous research has shown that TE benefits students’ interest (Kim and Schallert, 2014) and that there is a directional link from interest to SE (Brown et al., 2008; Sheu et al., 2010). In other words, teachers showing joy and encouragement in class or VLs could have a considerable effect on students’ SE (Dou et al., 2018), and teachers with high levels of TE are more likely to stimulate students’ interest, with highly interested students being more likely to develop high SE. In addition, according to emotional response theory, their emotional state will influence learners’ inclination to approach or escape from the subject or teachers in the process of learning. Teachers promote students’ absorption of academic commitment and achievement values by showing their enthusiasm and pleasure for certain topics or learning tasks (Pekrun et al., 2002; Pekrun, 2006). Mehrabian (1972) found that students were much more easily attracted by persons or things of which they were fond; in contrast, they escaped from things that they did not like. More precisely, teachers can improve students’ evaluation of value through enthusiastic teaching, i.e., students’ judgment of the importance and effectiveness of learning and achievement in a certain field. Therefore, showing enthusiasm is a role model for students, which may internalize their attitudes toward learning in such aspects as likes and values, thus improving their learning (Keller et al., 2016).

Teacher enthusiasm and cognitive load

CL is the total amount of psychological resources generated by information processing when processing specific tasks that need to be completed (Sweller et al., 1998). Based on cognitive load theory (CLT), in the process of multimedia learning, CL is classified into three types: germane, intrinsic, and extraneous. Germane load is linked to the mental effort that students devote to learning when shown learning materials. Intrinsic load is the inherent difficulty, which depends on the type of multimedia instructions; taking the example of chemical reactions, the intrinsic load for oxidation is lighter than that for oxidation–reduction. Intrinsic CL can be low or high, depending on the material being learned and its interaction with prior knowledge (Sweller, 2005, 2010), therefore intrinsic CL may vary from person to person. Extraneous load refers to the format presentation of the information that causes students to use additional mental resources to process it; poorly designed information presentation may bring a higher level of external load to learners. Extraneous CL directly affects students’ learning state, leading to problems such as inattention and reduced learning efficiency.

To acquire new knowledge, learners must process incoming information in their working memory, which is integrated with schemas stored in long-term memory. Working memory has a limited capacity, allowing only a limited number of items to be processed at one time (Cowan, 2001), and excessive information processing will cause CL and affect learning effect and efficiency. The core principle of CLT is that teaching design and learning environment should optimize learners’ limited working memory capacity to promote efficient learning. Based on CLT, different types of multimedia materials should reduce CL and make full use of working memory. Therefore, instructional designers should consider how to optimize learning and reduce unnecessary load as much as possible through effective design.

Compared with less-enthusiastic teachers, those with high levels of enthusiasm exhibit more-energetic tones and more facial expressions and gestures. Previous studies have shown that the nonverbal cues of pedagogical agents may distract the learner from salient learning (Clark and Choi, 2005; Woo, 2009; Frechette and Moreno, 2010). The results of Wilson et al. (2018) suggest that VLs involving teachers cost more to understand than do VLs containing only audio narrations by teachers. On the other hand, according to evidence supplied by various eye-tracking studies, the instructor also attracts a large amount of VA from the student (Kizilcec et al., 2014; Wang and Antonenko, 2017). Homer et al. (2008) examined how two versions of a computer-based multimedia presentation—i.e., a video version (including a video of a lecture synchronized with slides) and no video (including only the slides and an audio recording of a lecture)—affected LP; they found that having video as well as PowerPoint slides had a split-attention effect, subsequently increasing the CL. In summary, learners do divide their attention between the actor and the learning material (Louwerse et al., 2008). According to CLT, this may cause visual and cognitive interference to the learning process of learners, occupy the limited resources of working memory, and further distract learners from the processing of learning materials, resulting in additional CL (Clark and Choi, 2007).

However, some studies have shown that more verbal and nonverbal information in pedagogical agents does not increase students’ CL. Liew et al. (2017) compared the CL of students after watching teaching videos of highly enthusiastic or neutral pedagogical agents, and the results showed that there was no significant difference between the two groups of students. Wang et al. (2019) also showed that adding teachers’ facial expressions did not increase students’ CL.

According to the signaling principle of multimedia instruction design, the cues in multimedia learning resources can effectively guide learners to pay attention to important learning contents, to relieve the CL caused by attention delay or information screening in the learning process and improve their academic performance (Mayer and Moreno, 2003). Cues in video teaching include nonhuman ones (color, arrow, flicker, etc.) and human ones (teacher's expression, gaze, posture, etc.). Previous studies have found that social cues provided by humans guide learners’ attention far more than do these nonhuman cues (Johnson et al., 2015). Pi et al. (2017) found that teachers’ pointing gestures could direct learners’ VA to the corresponding teaching content to allocate more working memory resources to relevant learning activities by avoiding unnecessary visual searches on PowerPoint slides, and the results showed that learners’ LP was higher. Wang et al. (2019) investigated the effects of teachers’ facial expressions on students’ LP and CL; they found that the facial expressions of teachers are important cues for students’ memory, which could help teachers to emphasize key teaching points and eliminate uncertainty to help students understand well the information being conveyed. Van Gog et al. (2014) found that seeing the human model's face in the video fosters learning, a possible reason being that the speaker's gaze direction could guide learners’ attention at relevant moments and effectively transfer their attention from the model's face to the demonstration area. Therefore, a highly enthusiastic teacher with more facial expressions and gestures can reduce competition for working memory resources between relevant and irrelevant information and thus reduce unnecessary CL.

Based on the above discussion, a higher level of TE may either (i) distract students and thus increase CL or (ii) avoid adding extraneous CL by providing students with visual cues for learning. Therefore, it is very meaningful and necessary to determine the linkage between TE and CL. In the present study, we used a CL scale to measure students’ subjective CL and an eye-movement index to measure students’ objective CL and study the influence of higher TE on students’ CL in video teaching.

Eye-tracking technology in chemistry education

It is known that when our eyes are focused on different objects in our field of vision, we do not process each piece of information individually. Because of the attention mechanism, we are more likely to retrieve a certain amount of information, so we define VA as selectivity in perception (Orquin and Mueller, 2013). VA is an adequate representation of mental processing; in other words, VA is an indicator of mental attention. Numerous studies have shown that VA data can depict mental attention, and eye movements can be regarded as an important indicator of VA (Anderson et al., 2004; Hoang et al., 2008).

Eye-tracking technology is a crucial research tool in the field of education and is used mainly to track the process of people's eye movements that reflect their representation of mental processing in problem solving. Since Havanki and VandenPlas (2014) introduced eye-tracking technology into chemistry education research formally and regarded it as a useful and valuable tool for interpreting results more scientifically, more and more studies have chosen to use this visual survey tool in recent years, but there are still few studies related to the field of video teaching. With the help of eye-tracking technology, chemistry education researchers can explore learners' visual attention and cognitive processes while completing different tasks (Karch et al., 2019; Tang and Pienta, 2022), problem-solving methods and strategies (Tothova et al., 2021; VandenPlas et al., 2021), confirmation of setup demonstration principles (Nehring and Busch, 2020), understanding of chemical phenomena (Hansen et al., 2019), perceptual navigation patterns (Pande and Chandrasekharan, 2022) and some other aspects.

In addition, eye-tracking is an objective measure used in experimental psychology and in CL research through behavioral measures (Paas et al., 2003). The average fixation duration on relevant information, as well as the transitions between different areas of interest (AOIs)—providing information about learners’ cognitive engagement in information processing—should be considered as indicators of cognitive activity. The average fixation duration refers to the average duration of a fixation point in an AOI, with a longer fixation duration indicating a deeper and more effortful cognitive processing of visual information (Just and Carpenter, 1980; Holmqvist et al., 2011). For example, fixation duration increases as the complexity of texts or grammatical constructs increases (Rayner, 1998; Rayner et al., 2012). Fixation duration may thus be considered as an indicator of effort needed for visual information processing (Hodds et al., 2014; Krejtz et al., 2016; Strohmaier et al., 2020). According to the study of Paas and Vanmerrienboer (1994), the intensity of effort being expended by learners can be considered as the essence for obtaining a reliable estimate of CL. A widely used measure of CL is self-reported mental effort (Paas et al., 2003). In summary, average fixation duration can reflect the effort intensity of students in cognitive processing, and it is a very useful indicator in eye-tracking to confirm CL. Korbach et al. (2018) and Kruger and Doherty (2016) came to similar conclusions. Transition counts are considered to represent the learner's cognitive integration of multiple representations, which is the most critical step in multimedia learning for sense making (Mayer, 2014). High transition counts are seen as a strong cognitive investment in information integration (Schmidt-Weigand et al., 2010; Park et al., 2015), i.e., a higher transition count means that learners are exerting more mental effort to complete the information processing. Therefore, transition counts can be regarded as a indicator to confirm CL. Korbach et al. (2018) also proposed that the number of transitions could provide information on the causes of CL. Another eye-based measurement of CL is an increase in pupil size (Van Gog et al., 2009), but it is necessary to acknowledge that pupil size is too sensitive to other factors such as changes in light and brightness, and so researchers should be careful about interpreting this measure (Van Gog and Jarodzka, 2013). Therefore, in the present study, we chose the average fixation duration and transition counts as objective indicators to measure participants’ CL.

In the present study, eye-tracking technology was used to measure students’ VA to specific parts (learning-content materials or the teacher) of a certain screen and their cognitive load in the cognitive process. By analyzing the data, we discuss herein how TE influences students’ VA and CL in VLs.

Research questions

The ultimate goal of the present study was to investigate the connection between TE in VLs and students’ SE, VA, CL, and LP. Eye tracking is a new area of research in chemical education (Tang et al., 2014; Cullipher and Sevian, 2015). Based on the above literature review, we used eye-tracking technology in our study as a research tool for testing hypotheses. With this intent, we proposed the following four hypotheses related to the following two questions: (i) Is TE in chemistry VLs relevant to students’ SE, VA, CL, and LP? (ii) How do students’ SE, CL, VA, and LP compare across different levels of TE in chemistry VLs?

H1. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would show higher SE.

H2. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would pay more attention to the teacher.

H3. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would have a lower CL.

H4. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would show higher LP.

Methods

The research reported herein was quantitative in its nature, using eye tracking and experimental methods.

Participants

In this study, 70 junior middle-school students (34 boys and 36 girls, Mage = 14.45, SD = 0.52) attending seventh grade in a middle school in the southwest of China were engaged voluntarily with teacher and parental agreement. In China, classes are usually assigned according to students' learning ability, and in this study we selected two classes of grade seven in a junior middle school in the southwest of China as the research samples. The two classes were parallel, with students of similar learning ability, and the 70 selected students were ranked between 25% and 75% of their grade level in overall academic performance. Before the experiment, participants were randomly assigned to two groups to ensure the consistency of the two groups in all aspects as much as possible, so as to control other variables of the experiment. This sample had no contact with chemistry and therefore had no prior knowledge. In addition, even though we selected participants without bias toward ethnicity and family background, the gender balance and students’ blank previous chemistry knowledge should still be taken into account. After the experiment, each participant was offered a gift (a ballpoint pen) in appreciation of their support and enrolment.

Variables

In this study, the dependent variables were SE, VA, CL, and LP, and we compared the effects of different levels of TE on these variables. Also, in analyzing the students’ eye-gaze data, we measured their eye dwell time, fixation counts, and the average fixation duration and the transition counts on the AOIs. Table 2 describes the variables.
Table 2 Definitions of eye-movement measures
Variable name Description
Dwell time Total time for which the eye gaze of a learner remained on a certain AOI
Fixation count Total number of fixations on a certain AOI
Average fixation duration Average duration of a fixation point
Transition count Number of direct-gaze switches between two AOIs

Procedure and data collection

The full experiment took between 45 min and 60 min. After signing consent and release forms, all the participants were assigned randomly to two groups. Each participant was asked to sit a test of pre-knowledge (∼10 min), and these pre-test items corresponded to the key content in our instructional videos. According to the results of the pre-knowledge test, there was no significant difference in the scores of each group regarding previous knowledge (p = 0.921), indicating that the two groups had the same level of prior knowledge. Two instructional videos with different degrees of TE were prepared in advance: one was the video with a low level of TE to be watched by group A, and the other was the video with a high level of TE to be watched by group B.

The two experimental groups were taken to separate classrooms and followed the same procedures. Participants were asked to sit in a chair in front of an experimental monitor equipped with an eye-tracking device. Before starting to play the video instruction, the research assistant verbally explained the steps that the participants were expected to complete, and the same research assistant supervised the participants on every occasion. While viewing the uninterrupted video, the participants’ eye movements were recorded. After the video instruction, the participants completed three post questionnaires (SE questionnaire, CL questionnaire, and a test of post-learning comprehension) for 20 min at once. A flowchart of our method is shown in Fig. 1.


image file: d2rp00095d-f1.tif
Fig. 1 Experimental process of present study.

Instruments

In this study, we used different tools to collect different data to test the research hypotheses, including two experimental videos (a video with a high level of TE and one with a low level of TE), a test of pre-knowledge, an SE questionnaire, a CL questionnaire, a test of post-learning comprehension, and eye-tracking devices.
Experimental videos. As instructed by Murray (1983), we produced a VL with a high level of TE and one with a low level of TE. In the former, the teacher's performance—such as expressions and body gestures—was as instructed by Murray (1983), whereas in the VL with the low level of TE, the teacher did the opposite. Except for this, the two different videos had the same DK content, the same slides, the same lesson plan, the same length (13′25′′), and were taught by the same experienced female teacher. The content of the VLs was “Changes and Properties of Matter,” which stems from the Chinese textbook entitled “Chemistry (Compulsory Edition I)” published by the People's Education Press in China in 2007. Then we started to evaluate the VLs. The Chinese seventh-graders (n = 70) were assigned randomly to two groups (group A = 35, group B = 35) with different VLs, and they were asked to rate them on a scale of 1 (never) to 5 (always), still according to Murray's TE scale. The total scores of TE were calculated by adding the scores of the five items. The results showed that there were significant differences between groups A and B (p = 0.000): the group with the higher scores (MgroupB = 19.50, SDgroupB = 2.470) was the one that had experienced the higher level of TE, and the group with the lower scores (MgroupA = 5.60, SDgroupA = 1.057) was the other one. In addition, the reliability coefficient of the students—as measured by Cronbach's alpha—was α = 0.68 (group A) and 0.72 (group B). Therefore, the results indicated that both VLs followed Murray's instructions and were suitable for this study.
Test of pre-knowledge. We used an original pre-test paper to ensure that the participants in this study had no relevant knowledge. The test paper was created by two experienced teachers of junior middle school. Chemistry education experts and high-school chemistry teachers appraised the suitability of the items in the test paper, ensuring that it had good content validity. It included eight single-choice questions (1 point each) and one multiple-choice questions (2 points) relevant to the teaching content of the experimental VLs; each question had four possible answers, but only one of them was correct. For instance, “The following is a chemical property (….). A. Solubility; B. Volatility; C. Ductility; D. Stability.” and “The following is a chemical change (….). A. Ice melts to become water; B. A pot is made from pure iron; C. Heat rice for cooking; D. Gasoline evaporation.” The pre-test scores ranged from 1 to 10, which reflected the participants’ degree of prior knowledge of this topic. Cronbach's alpha was used in this study as an indicator of scale reliability or internal consistency (Taber, 2017), and the value of Cronbach's alpha for the pre-test paper was 0.93, indicating that the internal reliabilities of the scales were good.
Test of post-learning comprehension. The post-test paper was compiled by the same teachers who created the pre-test one, and all the questions in the post-test paper were on the video content. The post-test paper contained 13 questions of three types (completion, matching, and multiple choice), and full marks corresponded to 20 points. Scores were given by three trained graders, and the final score was the average score of the three. The consistency coefficient of the raters was 0.69. In addition, the post-test paper was entirely different from the pre-test one. All the items were relevant to the content that appeared in the experimental VLs, such as “2. Matter is made up of molecules and….” and “3. Which of the following statements is true? (….). A. Diamonds and pencil lead have the same composition; B. Fireflies glow with chemical changes; C. Flammable alcohol refers to the physical properties of alcohol; D. Diamond is a spatial network structure; graphite is a layered structure.”
Self-efficacy questionnaire. SE testing was another crucial part of the post-test. SE is often defined as an individual's perception of whether he or she can succeed (Bandura, 1986). In the present study, our aim was to use the SE scale developed by Tuan et al. (2005) to determine whether the different levels of TE influenced the students’ belief in their capabilities to obtain the desired effects. The specific scale is divided into three dimensions (understanding, confidence, and giving up) and contains six questions on a five-point Likert scale (see the electronic supplementary information). The scores of this scale ranged from 6 to 30 points, with a higher score meaning a higher SE. The reliability of the scale was α = 0.89.
Cognitive-load questionnaire. We chose an adapted version of the paper-and-pencil CL questionnaire of Moon and Ryu (2020). Participants responded to each of the 20 items on a five-point Likert scale ranging from 1 (never) to 5 (always). The questionnaire had five four-item scales, i.e., (i) task difficulty (ii) mental effort, (iii) perceived task difficulty, (iv) self-evaluation, and (v) usability, and the students could spend 10 min completing it. In this study, we also used eye-tracking methods to evaluate the participants’ CL.
Eye-tracking apparatus. In this study, we carried out an eye-tracking analysis to capture the participants’ VA when experiencing video instruction. In this experiment, we used an EyeLink 1000 Plus eye tracker with matching EyeLink Camera Link, Experiment Builder, and Data Viewer software to record the monocular eye movements. The eye tracker had a high degree of accuracy and a 2000 Hz sampling rate. It was fixed in front of the computer, ∼30 cm from the computer screen. The device contained two cameras that continuously monitored and tracked the eye and collected eye-gaze data. It required the participant's head not to move, so the latter was held in place by a headrest; by fixing the position of the headrest, the participant was guaranteed to be 70 cm away from the screen. The VLs appeared on the computer screen with a resolution of 1366 × 768. Before beginning the eye tracking, all participants completed C correction and V correction, with a fixation threshold of 1000 ms. C correction is the correction of the position deviation of eye movements to ensure that the eye-tracking cameras can detect the eye movements in the whole computer screen. V correction is the correction of angle deviation to ensure that the sampling point is consistent with the point seen by the eye. Only when C correction is well corrected and V correction is corrected to within 2° of visual angle can the eye tracking start to work.

While viewing the VLs, participants’ eye movements were recorded. To address the second research hypothesis, we analyzed dwell time and fixation counts. For this purpose, AOIs were defined to separate the teacher and the learning content of the video in visually (Fig. 2). The AOIs were created manually, and in the video, we identified two areas: (A1) learning-content materials; (A2) the teacher. The dwell time measures how long the learner's eye gazed at an AOI (including fixations and quick scans) and was relativized by the duration of the video. Areas with higher dwell times can be interpreted as having been prioritized more. Such measures have recently been used in teacher-gaze studies (Stahnke and Blömeke, 2021). Similarly, we analyzed the numbers of fixations on the two AOIs as a second indicator of participant’ VA; a high fixation count indicates that the learner repeatedly allocated their attention to the corresponding area (Holmqvist et al., 2011). Concerning the third research hypothesis, we focused on average fixation duration and transition count, which reflect students’ mental activities. The data can provide indirect evidence for judging the size and change of students’ CL.


image file: d2rp00095d-f2.tif
Fig. 2 A screenshot of the high-TE video lecture (VL) (The red box represents the area of learning-content materials and the yellow box represents the area of the teacher).

Data analysis

The experimental data of this study were collected immediately after the end of the experiment. All the data were collected in Excel and processed statistically in SPSS (Statistical Package for the Social Sciences) to compare how TE influenced the students’ SE, VA, CL, and LP. Specifically, the students’ VA was assessed by analyzing eye-movement data (dwell time, fixation counts), the students’ CL was assessed by analyzing eye-movement data (average fixation duration, transition counts) and the scores on the CL scale, and the students’ LP and SE were assessed by analyzing the pre-test and post-test scores and the SE scales.

Before running the analyses described below (t-test), we checked the normality by a Shapiro–Wilk test (p > 0.05) and the homogeneity of variance by Levene's test (p > 0.05). When these conditions were not met, we ran nonparametric statistics on independent samples; the nonparametric test was the Mann–Whitney U test. In the present study, the measures for hypotheses H1 (SE) and H4 (LP) were analyzed using a t-test, those for hypothesis H2 (VA) were analyzed by nonparametric tests, and those for hypothesis H3 were analyzed using a t-test (CL scales) and nonparametric tests (eye-tracking outcomes).

In addition, the statistical hypotheses were tested at an alpha error rate of 5%. Effect sizes estimate the magnitude of effect or association between two or more variables (Snyder and Lawson, 1993); effect sizes are resistant to sample-size influence and thus provide a truer measure of the magnitude of effect between variables (Ferguson, 2009). In present study, to describe whether the effects had a relevant magnitude, the effect-size measures of Cohen's d and η2 were used to describe the strength of the phenomenon. Specifically, Cohen's d was used for the t-test, and Cohen (1992) defined the size of Cohen's d value as follows: 0.2, indicating that the experimental effect was small, 0.5, indicating that the experimental effect was medium, and 0.8, indicating that the experimental effect was large. η2 was used for the Mann–Whitney U test, with baseline definitions of small (0.01), medium (0.06), and large (0.14) effects provided by Cohen (1988).

Results and discussion

H1. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would show higher SE.

Fig. 3 shows the outcome for the SE of all the students in the two groups. The t-test indicated that the SE scores in group B were significantly higher than those in group A (MgroupA = 22.77, MgroupB = 24.69, p = 0.002, Cohen's d = 1.13). Cohen's d was 1.13, indicating a large effect size. This result supports hypothesis H1 and is consistent with emotional response theory and the results of some other scholars (Rosenshine, 1970; Horan et al., 2012). The mere act of watching a video is primarily a passive learning experience, likely leading to a lack of student engagement and hindered learning (Mirriahi et al., 2021). For learning to be successful, students must hold adequate SE beliefs and keep investing sufficient effort in learning. There is evidence that students’ SE increases significantly when they receive positive emotional expressions and when motivating messages are present (Van der Meij, 2013). Consistent with our research results, positive emotions exhibited by enthusiastic teachers may enhance students’ SE, and students with more SE may transform the passive experience of watching videos into a more active experience and promote learning (Mirriahi et al., 2021).


image file: d2rp00095d-f3.tif
Fig. 3 Means and standard deviations of self-efficacy (SE) of students in groups A and B.

H2. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would pay more attention to the teacher.

Hypothesis H2 focuses on how the different levels of TE affect the distribution of the students’ attention on both the learning-content materials (area A1) and the teacher (area A2) during the VL learning. The eye-tracker data showed that the different VLs resulted in a significant difference in the students’ VA. We analyzed the students’ total dwell time, fixation counts, and heat maps, and the specific data for each AOI are given in Table 3.

Table 3 Eye-movement data for visual attention in high- enthusiasm and low-enthusiasm groups
Indicator Group N Mdn IQR Mann–Whitney U test
U p η 2
Dwell time [ms] A 35 573124.18 137812.85 498.00 0.179 0.056
B 35 538932.09 232035.18
Dwell time: A1 [ms] A 35 459754.10 80163.80 422.00 0.025 0.045
B 35 428060.49 153114.89
Dwell time: A2 [ms] A 35 78684.23 44959.21 316.00 0.000 0.131
B 35 99893.79 59642.09
Fixation counts: A1 A 35 1825.61 403.11 485.00 0.134 0.011
B 35 1534.13 488.74
Fixation counts: A2 A 35 300.69 213.02 474.00 0.104 0.046
B 35 328.12 107.38

According to the Shapiro–Wilk tests, the eye-tracking data were not normally distributed. All measures were compared between groups A and B by the Mann–Whitney U test for independent samples. Basic descriptive statistics (median Mdn and interquartile range IQR) of the numerical variables were determined. The Mdn of the dwell time and the fixation counts in the teaching-content display area (area A1) for group A were higher than those for group B, and the situation was just the opposite for the teacher display area (area A2). The comparison in Table 3 shows that the dwell time in area A1 for group A was significantly higher than that for group B with small effect sizes (p = 0.025, η2 = 0.045), but the Mdn of the dwell time in area A2 for group A was significantly lower than that in group B with large effect sizes (p = 0.000, η2 = 0.131). That is to say, TE had a significant effect on students' dwell time in area A2, significantly increasing students' attention on the teacher, but the effect on the dwell time in area A1 is not obvious. This indicates that TE did not have a significant effect on students' attention to the learning-content materials (area A1), i.e., the “distraction effect” was not significant, although the high-TE teacher caused students' attention to shift more toward her. Also, heat maps are often used to show observation areas. This representation includes the so-called “hot spots” (marked in red) that show where students paid most attention. The heat maps in Fig. 4 show that the course content for group A received considerable VA.


image file: d2rp00095d-f4.tif
Fig. 4 Sample distributions of students’ eye fixations on a VL: (a) low level of teacher enthusiasm (TE) (group A); (b) high level of TE (group B). (The red box represents the area of learning-content materials and the yellow box represents the area of the teacher).

The analysis of the eye-movement data showed that the teacher with the higher level of TE added to the students’ VA, which supports hypothesis H2. Analyzing the watching behavior of individual learners combined with detailed learning assessments offers rich insights into how different levels of TE might aid or hinder learning. This study found that characterizing learners’ watching behavior emphasized the great impact of TE on VA. When watching the VL with the higher level of TE, the students paid more attention to the teacher (area A2) and less attention to other areas, this being because the teacher exhibited richer expressions (smiles, surprise, questions), gestures, lively intonation, etc., which led to the students being drawn more to the teacher. As shown in Fig. 4, the teacher was asking the question “Is chemistry a subject that studies explosions?” The teacher with the lower level of TE asked this question in a neutral mood, while the teacher with the higher level of TE asked this question with a brisk tone and hand gestures describing explosions. The heat maps show that students in the low-TE group paid more attention to area A1, while students in the high-TE group paid more attention to area A2. This is consistent with other research that showed that teachers’ gestures (Rueckert et al., 2017) and facial expressions (Gullberg and Holmqvist, 2006) can draw people's attention to the speaker.

H3. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would have a lower CL.

The CL scales were analyzed using a t-test, and the results showed that the CL of group A (M = 65.06) was higher than that of group B with medium effect sizes (M = 62.66, p = 0.015, Cohen's d = 0.598). This shows that the group that experienced the higher level of TE experienced the lower CL. Furthermore, we also used the transition counts and average fixation duration as indicators for measuring the students’ CL, and the CL scores of the two groups are shown in Fig. 5. The nonparametric Mann–Whitney U test indicated that there was a significant difference with medium effect sizes (p = 0.015, η2 = 0.076) in the transition counts between AOIs in the groups, and the average fixation duration for group B was much lower than that for group A with medium effect sizes (p = 0.023, η2 = 0.065).


image file: d2rp00095d-f5.tif
Fig. 5 Box plot of students’ CL scores.

In VLs, attentional cues such as gestures and facial expressions can guide learners’ VA in a timely manner to the information mentioned by the instructor, which can be a very effective way to reduce visual searches in a multimedia explanation. The teacher with the lower level of TE kept her body still, did not move her arms, did not have a directional gaze (i.e., did not shift her eyes from the camera to the learning content being discussed), and kept facing the students, whereas the teacher with the higher level of TE used pointing gestures while turning her upper body to the knowledge area; for example, the latter teacher pointed to the periodic table of elements in the demonstration area and turned her eyes to it when she said that it was discovered by the Russian chemist Mendeleev. Visual search efficiency describes the speed at which observers locate relevant visual information when they hear or read the corresponding word. The gestures and facial expressions of the high-TE teacher helped those participants shift their attention from her to the demonstration area in a timely and effective manner, and she guided students to pay attention to important informative words or concepts, thereby enabling the students to pay attention to relevant information more quickly and increasing their search efficiency (de Koning et al., 2009).

In addition, high fixation counts are associated with low search efficiency (Goldberg and Kotval, 1999), so the lower fixation counts of group B compared to group A (area A1) support the assertion that TE provides attention cues. Previous work has shown that attention cues can guide learners’ VA in a timely manner and reduce the competition for working memory resources between relevant information and irrelevant information in VLs, thereby reducing unnecessary CL (Van Gog et al., 2014). The present results are consistent with those of Pi et al. (2017) in that when the students were watching the VLs, their eyes and attention consciously followed the teacher's guiding behaviors to lead to the teaching content, thereby better allocating attention to the process of memorizing the teaching content and reducing CL. Therefore, TE can be effective in helping students to capture those parts of the learning materials that need to be processed deeply in rapid learning, thereby avoiding increased consumption of ineffective cognitive resources. Our research results support hypothesis H3, and the specific data for each AOI are given in Table 4.

Table 4 Eye-movement data for cognitive load (CL) in high- and low-enthusiasm groups
Indicator Group N Mdn IQR Mann–Whitney U test
U p η 2
Average fixation duration [ms] A 35 281.18 82.79 419.50 0.023 0.065
B 35 230.77 128.38
Transition counts A 35 132.32 63.86 404.50 0.015 0.076
B 35 107.23 65.98

H4. Compared to the students in the group that experienced a low level of TE, those in the group that experienced a high level of TE would show higher LP.

We used a t-test to identify how the different levels of TE affected the students’ post-test scores. As was shown earlier, there was no significant difference in the students’ prior knowledge between the two groups (MgroupA = 2.60, MgroupB = 2.57, p = 0.921). Also, all the students in this study had blank knowledge about this field and needed to learn more about the content of “Changes and Properties of Matter” in the VLs. The LPs of the two groups are shown in Fig. 6, and clearly the LP scores of group B (M = 9.08, SD = 1.221) were higher than those group A (M = 6.46, SD = 0.817). Moreover, there was a significant difference between the two groups with large effect sizes (p = 0.000, Cohen's d = 1.817). The results show that the students in group B outperformed those in group A, which supports hypothesis H4. This is consistent with previous work that shows that TE has a positive impact on teaching effectiveness and that student performance is positively correlated with TE (Streeter, 1986; Kunter, 2013; Mahler et al., 2018). Based on hypothesis H4, we carefully conclude that the enthusiasm of chemistry teachers in VLs may be very important for students’ LP and should therefore be cultivated in pre-service and in-service teacher education. For example, teachers could develop awareness and skills to remain enthusiastic when facing the camera by participating in professional development courses or self-study.


image file: d2rp00095d-f6.tif
Fig. 6 Average scores of students’ learning performance in both groups.

Conclusions

The purpose of this study was to investigate how different levels of TE in chemistry VLs affected students’ SE, VA, CL, and LP. The resulting findings show that a higher level of TE did indeed improve students’ SE, VA, and LP. Moreover, although the enthusiastic teacher exhibited more-abundant nonverbal behavior, her enthusiasm did not add to the students’ CL. Consequently, our data make an empirical contribution to TE design in multimedia learning environments, in terms of which the importance of teachers’ nonverbal design is highlighted.

According to the “split-attention effect” (Ayres and Sweller, 2014), more nonverbal information delivered by teachers with high TE will distract students’ attention and lead to them paying more attention to their teachers. We also showed the existence of this effect in hypothesis H2. However, the high-TE teacher did not reflect the harmful effect of split attention on learning. In the present study, the LP of the students in the high-TE group was significantly higher than that of those in the low-TE group. This result is similar to previous studies. For example, Wang and Antonenko (2017) studied the effects of instructor presence or absence on students’ VA distribution and learning effect in video teaching, and the results showed that the presence of the teacher divided the students’ attention to the teacher (26% of the students’ dwell time), and the students’ learning of simple questions was significantly better. Van Gog et al. (2014) conducted an eye-tracking study in which students learned problem-solving tasks from video modeling examples in which the model's face was either visible or invisible; their results showed that the percentage of fixation duration on the demonstration area decreased by 22.91% when the model's face was visible. This suggests that the model's face distracted learners from the presentation area, but the students’ performance improved significantly after they had seen the video modeling example twice. As discussed in the theoretical framework of the present study, the nonverbal information of the high-TE teacher played an important signaling function, directing the learners’ attention to the most relevant and important aspects of the teaching content, enabling them to participate in cognitive processing of the learning materials, and promoting the integration of visual demonstration and verbal explanation. It is reasonable to speculate that high TE has a positive effect on students’ LP because the signaling advantage brought by teachers’ nonverbal cues is greater than the negative effect of distraction.

In addition, facial expressions (eye, eyebrow, and mouth movements) are particularly important for online learning (Johnson et al., 2000; Atkinson, 2002). The face is the primary source for expressing emotions nonverbally (Mehrabian, 1971). In the present VLs, the high-TE teacher used different types of facial expressions: happy, surprised, confused, etc. The emotion expressed by the instructor's facial expressions in VLs is critical to shortening this distance between students and online teachers, i.e., it could overcome students’ feelings of isolation and improve their arousal level and participation in learning activities (Borup et al., 2014). Baylor and Kim (2009) noted that persuasion can be facilitated though the physical social cues of an agent's eyes, face, hands, and body; because the agent's facial expressions contain emotions, this enhances the persuasiveness of the lesson to the students, thus benefiting their learning. Mayer and DaPra (2012) reported a similar result that virtual instructors’ appropriate use of facial expressions could promote students’ deep cognitive processing and encourage their learning. Therefore, the nonverbal information conveyed by the high-TE teacher prompted the processing of verbal information.

However, nonverbal cues should be designed carefully to ensure that they do not increase CL. From a practical viewpoint, this study showed that TE can be formulated via the operation of teachers’ expressions and body gestures. Note that cues via immediate controls of teachers’ voice tones, facial expressions, and animation gestures open a myriad of exciting potential benefits regarding the development of learner interaction in VLs. In multimedia learning environments, TE helps students to maintain a higher level of SE. Moreover, TE maintains a higher attention level from student to teacher and improves students’ scores. This study also showed that CL was positively influenced by TE. Therefore, whether in a conventional classroom or a multimedia learning environment, teachers should pay attention to the proactive expression of their enthusiasm.

When interpreting our findings, some limitations should be kept in mind. First, the limited sample size did not involve a nationally representative sample of students, which may restrain us from generalizing the results of our study to all students in China; thus, future studies should expand the sample size by including students from different subjects, grade levels, and regions. Second, because some students were curious about the eye tracker, they might have paid more attention to it than to the VL, and this restricted the scientific nature of the research results to some extent. Third, this study involved only chemistry DK, and PK could be studied in the future. Finally, an important direction of future study would be to understand well the effects and mechanism of teachers’ other nonverbal cues regarding students’ learning.

Conflicts of interest

No potential conflict of interest was reported by the authors.

Acknowledgements

This study was financially supported by the Funds of the Ministry of Education of Humanities and Social Science Project (21YJA880018). We express our sincere thanks to the teachers and students who participated in this study.

References

  1. Allen M., Witt P. L. and Wheeless L. R., (2006), The role of teacher immediacy as a motivational factor in student learning using meta-analysis to test a causal model, Commun. Educ., 55(1), 21–31.
  2. Anderson J. R., (1976), Language, memory, and thoughts, Hillsdale, NJ: Erlbaum.
  3. Anderson J. R., Bothell D. and Douglass S., (2004), Eye movements do not reflect retrieval processes: Limits of the eye-mind hypothesis, Psychol. Sci., 15, 225.
  4. Arguedas M. and Daradoumis T., (2021), Analyzing the role of a pedagogical agent in psychological and cognitive preparatory activities, J. Comput. Assist. Learn., 37(4), 1167–1180.
  5. Atkinson R. K., (2002), Optimizing learning from examples using animated pedagogical agents, J. Educ. Psychol., 94, 416–427.
  6. Ayres P. and Sweller J., (2014), The split-attention principle in multimedia learning, in Mayer R. E. (ed.), The Cambridge handbook of multimedia learning, New York, Cambridge University Press, pp. 206–226.
  7. Bandura A., (1982), Self-efficacy mechanism in human agency, Am. Psychol., 37, 122–147.
  8. Bandura A., (1986), The explanatory and predictive scope of self-efficacy theory, J. Soc. Clin. Psychol., 4, 359–373.
  9. Bandura A., (1997), Self-efficacy: Toward a unifying theory of behavioral change, Psychol. Rev., 84, 191–215.
  10. Batty A. O., (2020), An eye-tracking study of attention to visual cues in L2 listening tests, Language Test., 38, 511–535.
  11. Baylor A. L. and Kim S., (2009), Designing nonverbal communication for pedagogical agents: When less is more, Comput. Hum. Behav., 25(2), 450–457.
  12. Becker E. S., Goetz T., Morger V. and Ranellucci J., (2014), The importance of teachers' emotions and instructional behaviour for their students' emotions – An experience sampling analysis, Teach. Teach. Educ., 43, 15–26.
  13. Borup J., West R. E., Thomas R. and Graham C. R., (2014), Examining the impact of video feedback on instructor social presence in blended courses, Int. Rev. Res. Open Distributed Learn., 15(3), 232–256.
  14. Brophy J. E. and Good T. L., (1986), Teacher behavior and student achievement, in Wittrock M. (ed.), Handbook of research on teaching, pp. 328–375.
  15. Brown S. D., Tramayne S., Hoxha D., Telander K., Fan X. and Lent R. W., (2008), Social cognitive predictors of college students' academic performance and persistence: A meta-analytic path analysis, J. Vocat. Behav., 72, 298–308.
  16. Burić I., (2019), The role of emotional labor in explaining teachers' enthusiasm and students' outcomes: A multilevel mediational analysis, Learn. Individ. Differ., 70, 12–20.
  17. Chen C. M. and Wu C. H., (2015), Effects of different video lecture types on visual attention, emotion, cognitive load, and learning performance, Comput. Educ., 20, 143–156.
  18. Clark R. E. and Choi S., (2005), Five design principles for experiments on the effects of animated pedagogical agents, J. Educ. Comput. Res., 32, 209–225.
  19. Clark R. E. and Choi S., (2007), The questionable benefits of pedagogical agents: Response to Veletsianos, J. Educ. Comput. Res., 36(4), 379–381.
  20. Cohen J., (1988), Statistical Power Analysis for the Behavioral Sciences, 2nd edn, Hillsdale, NJ: Lawrence Erlbaum Associates.
  21. Cohen J., (1992), A power primer, Psychol. Bull., 112(1), 155–159.
  22. Collins M. L., (1978), Effects of enthusiasm training on preservice junior middle school teachers, J. Teach. Educ., 29, 53–57.
  23. Cowan N., (2001), The magical number 4 in short-term memory: A reconsideration of mental storage capacity, Behav. Brain Sci., 24(1), 87–114.
  24. Cullipher S. and Sevian H., (2015), Atoms versus bonds: How students look at spectra, J. Chem. Educ., 92, 1996–2005.
  25. de Koning B. B., Tabbers H. K., Rikers R. M. J. P. and Paas F., (2009), Towards a framework for attention cueing in instructional animations: Guidelines for research and design, Educ. Psychol. Rev., 21(2), 113–140.
  26. Dou R., Brewe E., Potvin G., Zwolak J. P. and Hazari Z., (2018), Understanding the development of interest and self-efficacy in active-learning undergraduate physics courses, Int. J. Sci. Educ., 10, 1464–5289.
  27. Fabio R. A. and Antonietti A., (2012), Effects of hypermedia instruction on declarative, conditional and procedural knowledge in ADHD students, Res. Dev. Disabil., 33, 2028–2039.
  28. Feldman K. A., (2007), Identifying exemplary teachers and teaching: Evidence from student ratings, in Perry R. P. and Smart J. C. (ed.), The scholarship of teaching and learning in higher education: An evidence-based perspective, Dordrecht: Springer Netherlands, pp. 93–143.
  29. Ferguson C. J., (2009), An effect size primer: A guide for clinicians and researchers, Prof. Psychol. Res. Pr., 40(5), 532–538.
  30. Frechette C. and Moreno R., (2010), The roles of animated pedagogical agents’ presence and nonverbal communication in multimedia learning environments, J. Media Psychol., 22, 61–72.
  31. Frenzel A. C., Goetz T., Lüdtke O., Pekrun R. and Sutton R. E., (2009), Emotional transmission in the classroom: Exploring the relationship between teacher and student enjoyment, J. Educ. Psychol., 101, 705–716.
  32. Goldberg J. H. and Kotval X. P., (1999), Computer interface evaluation using eye movements: Methods and constructs, Int. J. Ind. Ergon., 24(6), 631–645.
  33. Grimmer C., (1992), Supertext English language subtitles: A boon for English language learners, EA Journal., 10(1), 66–75.
  34. Gullberg M. and Holmqvist K., (2006), What speakers do and what addressees look at: Visual attention to gestures in human interaction live and on video, Pragmat. Cogn., 14, 53–82.
  35. Hansen S. J. R., Hu B., Riedlova D., Kelly R. M., Akaygun S. and Villalta-Cerdas A., (2019), Critical consumption of chemistry visuals: Eye tracking structured variation and visual feedback of redox and precipitation reactions, Chem. Educ. Res. Pract., 20(4), 837–850.
  36. Havanki K. L. and VandenPlas J. R., (2014), Eye tracking methodology for chemistry education research, in Bunce D. M. and Cole R. S. (ed.), Tools of Chemistry Education Research.
  37. Hoang D. A., Bays P. and Husain M., (2008), Eye movements as a probe of attention, Prog. Brain Res., 171, 403–411.
  38. Hodds M., Alcock L. and Inglis M., (2014), Self-explanation training improves proof comprehension, J. Res. Math. Educ., 45(1), 62–101.
  39. Hoffler T. N. and Leutner D., (2007), Instructional animation versus static pictures: A meta-analysis, Learn. Instr., 17, 722–738.
  40. Holmqvist K., Nystrom M., Andersson R., Dewhurst R., Jarodzka H. and Van de Weijer J., (2011), Eye tracking: A comprehensive guide to methods and measures, Oxford: Oxford University Press.
  41. Homer B. D., Plass J. L. and Blake L., (2008), The effects of video on cognitive load and social presence in multimedia-learning, Comput. Human. Behav., 24(3), 786–797.
  42. Horan S., Martin M. and Weber K., (2012), Understanding emotional response theory: The role of instructor power and justice messages, Commun. Q., 60, 210–233.
  43. Johnson W. L., Rickel J. W. and Lester J. C., (2000), Animated pedagogical agents: Face-to-face interaction in interactive learning environments, Int. J. Artif. Intell. Educ., 11, 47–78.
  44. Johnson A. M., Ozogul G. and Reisslein M., (2015), Supporting multimedia learning with visual signaling and animated pedagogical agent: Moderating effects of prior knowledge, J. Comput. Assist. Learn., 31(2), 97–115.
  45. Just M. A. and Carpenter P. A., (1980), A theory of reading: From eye fixations to comprehension, Psychol. Rev., 87(4), 329–354.
  46. Karch J. M., Valles J. C. G. and Sevian H., (2019), Looking into the black box: Using gaze and pupillometric data to probe how cognitive load changes with mental tasks, J. Chem. Educ., 96(5), 830–840.
  47. Keller M., Neumann K. and Fischer H. E., (2013), Teacher enthusiasm and student learning, New York: Routledge Press, pp. 247–249.
  48. Keller M. M., Goetz T., Hoy A. W. and Frenzel A. C., (2016), Teachers' enthusiasm: Reviewing and redefining a complex construct, Educ. Psychol. Rev., 28, 743–769.
  49. Keller M. M., Becker E. S., Frenzel A. C. and Taxer J. L., (2018), When teacher enthusiasm is authentic or inauthentic: Lesson profiles of teacher enthusiasm and relations to students' emotions, AERA Open, 4, 1–16.
  50. Kim T. and Schallert D. L., (2014), Mediating effects of teacher enthusiasm and peer enthusiasm on students' interest in the college classroom, Contemp. Educ. Psychol., 39, 134–144.
  51. Kizilcec R. F., Papadopoulo K. and Sritanyaratana L., (2014), Showing face in video instruction: Effects on information retention, visual attention, and affect, Paper presented at proceedings of the 32nd Annual ACM Conference on Human factors in computing systems, ACM, pp. 2095–2102.
  52. Korbach A., Brünken R. and Park B., (2018), Differentiating different types of cognitive load: A comparison of different measures, Educ. Psychol. Rev., 30(2), 503–529.
  53. Krejtz K., Duchowski A. T., Krejtz I., Kopacz A., and Chrząstowski-Wachtel P., (2016), Gaze transitions when learning with multimedia, J. Eye. Mov. Res., 9(1), 5.
  54. Kruger J. L. and Doherty S., (2016), Measuring cognitive load in the presence of educational video: Towards a multimodal methodology, Australasian J. Educ. Technol., 32(6), 19–31.
  55. Kunter M., (2013), Motivation as an aspect of professional competence: Research findings on teacher enthusiasm, in Kunter M., Baumert J., Blum W., Klusmann U., Krauss S. and Neubrand M. (ed.), Cognitive activation in the mathematics classroom and professional competence of teachers, New York: Springer, pp. 273–289.
  56. Kunter M., Tsai Y. M., Klusmann U., Brunner M., Krauss S. and Baumert J., (2008), Students' and mathematics teachers' perceptions of teachers' enthusiasm and instruction, Learn. Instr., 18, 468–482.
  57. Kunter M., Frenzel A., Nagy G., Baumert J. and Pekrun R., (2011), Teachers' enthusiasm: Dimensionality and context specificity, Contemp. Educ. Psychol., 36, 289–301.
  58. Kunter M., Klusmann U., Baumert J., Richter D., Voss T. and Hachfeld A., (2013), Professional competence of teachers: Effects on instructional quality and student development, J. Educ. Psychol., 105, 805–820.
  59. Lazaridesa R., Buchholzb J. and Rubacha, (2018), Teachers' enthusiasm and self-efficacy, student perceived mastery goal orientation, and student motivation in mathematics classrooms, Teach. Teach. Educ., 69, 1–10.
  60. Liew T. W., Zin A. M. and Sahari N., (2017), Exploring the affective, motivational and cognitive effects of pedagogical agent enthusiasm in a multimedia learning environments, Human-Centric Comput. Inf. Sci., 7(9), 1–21.
  61. Louwerse M. M., Graesser A. C., McNamara D. S. and Lu S. L., (2008), Embodied conversational agents as conversational partners, Appl. Cogn. Psychol., 23(9), 1244–1255.
  62. Mahler D., Grossschedl J. and Harms U., (2018), Does motivation matter? – The relationship between teachers' self-efficacy and enthusiasm and students' performance, PLoS One, 13(11), e0207252.
  63. Markham P. L., Peter L. A. and McCarthy T. J., (2001), The effects of native language vs. target language captions on foreign language students' DVD video comprehension, Foreign Lang. Ann., 34(5), 439–445.
  64. Marsh H. W., (1982), SEEQ: A reliable, valid, and useful instrument for collecting students' evaluations of university teaching, Br. J. Educ. Psychol., 52, 177–195.
  65. Marsh H. W., (1994), Weighting for the right criteria in the instructional development and effectiveness assessment (IDEA) system: Global and specific ratings of teaching effectiveness and their relation to course objectives, J. Educ. Psychol., 86, 631–648.
  66. Marsh H. W. and Ware J. E., (1982), Effects of expressiveness, content coverage, and incentive on multidimensional student rating scales: new interpretations of the Dr Fox effect, J. Educ. Psychol., 74, 126–134.
  67. Mautone P. D. and Mayer R. E., (2001), Signaling as a cognitive guide in multimedia learning, J. Educ. Psychol., 93, 377–389.
  68. Mayer R. E., (2014), Cognitive theory of multimedia learning, in Mayer R. E. (ed.), The Cambridge handbook of multimedia learning, New York, Cambridge University Press, pp. 43–71.
  69. Mayer R. E. and DaPra C. S., (2012), An embodiment effect in computer-based learning with animated pedagogical agents, J. Exp. Psychol. Appl., 18(3), 239–252.
  70. Mayer R. E. and Moreno R., (2003), Nine ways to reduce cognitive load in multimedia learning, Educ. Psychol., 38(1), 43–52.
  71. Mehrabian A., (1971), Silent messages, Wadsworth.
  72. Mehrabian A., (1972), Nonverbal communication, Chicago, IL: Aldine-Atherton.
  73. Mirriahi N., Jovanovic J., Lim L. A. and Lodge J. M., (2021), Two sides of the same coin: Video annotations and in-video questions for active learning, Educ. Technol. Res. Dev., 69(5), 2571–2588.
  74. Moon J. and Ryu J., (2020), The effects of social and cognitive cues on learning comprehension, eye-gaze pattern, and cognitive load in video instruction, J. Comput. High. Educ., 33(1), 39–63.
  75. Mottet T. P., Frymier A. B. and Beebe S. A., (2006), Theorizing about instructional communication, in Mottet T. P., Richmond V. P. and McCroskey J. C. (ed.), Handbook of instructional communication: Rhetorical and relational perspectives, Boston, MA: Allyn and Bacon, pp. 3–32.
  76. Murray G., (1983), Low-inference classroom teaching behaviors and student rating of college teaching effectiveness, J. Educ. Psychol., 75, 138–149.
  77. Murray H. G., (2007), Low-inference teaching behaviors and college teaching effectiveness: Recent developments and controversies, in Perry R. P. and Smart J. C. (ed.), The scholarship of teaching and learning in higher education: An evidence-based perspective, Dordrecht: Springer Netherlands, pp. 145–200.
  78. Nehring A. and Busch S., (2020), Setting up chemistry demonstrations according to the left-to-right principle: An eye-movement-pattern-based analysis, J. Chem. Educ., 97(1), 275–282.
  79. Orquin J. L. and Mueller L. S., (2013), Attention and choice: A review on eye movements in decision making, Acta Psychol., 144, 190–206.
  80. Ozcelik E., Arslan-Ari I. and Cagiltay K., (2010), Why does signaling enhance multimedia learning? Evidence from eye movements, Comput. Human. Behav., 26(1), 110–117.
  81. Paas F. G. W. C. and Vanmerrienboer J. J. G., (1994), Instructional control of cognitive load in the training of complex cognitive tasks, Educ. Psychol. Rev., 6(4), 351–371.
  82. Paas F., Tuovinen J. E., Tabbers H. and Van Gerven P. W. M., (2003), Cognitive load measurement as a means to advance cognitive load theory, Educ. Psychol., 38(1), 63–71.
  83. Pande P. and Chandrasekharan S., (2022), Expertise as sensorimotor tuning: Perceptual navigation patterns mark representational competence in science, Res. Sci. Educ., 52(2), 725–747.
  84. Park B., Korbach A. and Brünken R., (2015), Do learner characteristics moderate the seductive-details-effect? A cognitive-load-study using eye-tracking, J. Educ. Techno. Soc., 18, 24–36.
  85. Patrick B. C., Hisley J. and Kempler T., (2000), “What' s everybody so excited about?”: The effects of teachers' enthusiasm on student intrinsic motivation and vitality, J. Exp. Educ., 68, 217–236.
  86. Pekrun R., (2006), The control-value theory of achievement emotions: Assumptions, corollaries, and implications for educational research and practice, Educ. Psychol. Rev., 18, 315–341.
  87. Pekrun R., Goetz T., Titz W. and Perr R. P., (2002), Academic emotions in students' self-regulated learning and achievement: A program of qualitative and quantitative research, Educ. Psychol., 37, 91–105.
  88. Pi Z. L., Hong J. H. and Yang J. M., (2017), Effects of the instructor's pointing gestures on learning performance in video lectures, Br. J. Educ. Technol., 48(4), 1020–1029.
  89. Rayner K., (1998), Eye movements in reading and information processing: 20 years of research, Psychol. Bull., 124(3), 372–422.
  90. Rayner K., Pollatsek A., Ashby J. and Clifto C. J., (2012), Psychology of Reading, 2nd edn, New York: Psychology Press.
  91. Rosenshine B., (1970), Enthusiastic teaching: A research review, Sch. Rev., 78, 499–514.
  92. Rueckert L., Church R. B., Avila A. and Trejo T., (2017), Gesture enhances learning of a complex statistical concept, Cogn. Res. Princ. Impl., 2, 2.
  93. Tang H., Kirk J. and Pienta N. J., (2014), Investigating the effect of complexity factors in stoichiometry problems using logistic regression and eye tracking, J. Chem. Educ., 91, 969–975.
  94. Schmidt-Weigand F., Kohnert A. and Glowalla U., (2010), A closer look at split visual attention in system-and self-paced instruction in multimedia learning, Learn. Instr., 20(2), 100–110.
  95. Schneider S., Krieglstei F., Beege M. and Rey G. D., (2021), The impact of video lecturers' nonverbal communication on learning – An experiment on gestures and facial expressions of pedagogical agents, Comput. Educ., 176, 104350.
  96. Snyder P. and Lawson S., (1993), Evaluating results using corrected and uncorrected effect size estimates, J. Exp. Educ., 61(4), 334–349.
  97. Sheu H. B., Lent R. W., Brown S. D., Miller M. J., Hennessy K. D. and Duffy R. D., (2010), Testing the choice model of social cognitive career theory across Holland themes: A meta-analytic path analysis, J. Vocat. Behav., 76, 252–264.
  98. Shuell T. J., (1996), Teaching and learning in a classroom context, in Berliner D. C. and Calfee R. C. (ed.), Handbook of educational psychology, New York: Simon and Schuster Macmillan, pp. 726–764.
  99. Sueyoshi A. and Hardison D. M., (2005), The role of gestures and facial cues in second language listening comprehension, Lang. Learn., 55(4), 661–699.
  100. Stahnke R. and Blömeke S., (2021), Novice and expert teachers' noticing of classroom management in whole-group and partner work activities: Evidence from teachers' gaze and identification of events, Learn. Instr., 74, 101464.
  101. Streeter B. B., (1986), The effects of training experienced teachers in enthusiasm on students' attitudes toward reading, Read. Psychol., 7(4), 249–259.
  102. Strohmaier A. R., MacKay K. J., Obersteiner A. and Reiss K. M., (2020), Eye-tracking methodology in mathematics education research: A systematic literature review, Educ. Stud. Math., 104(2), 147–200.
  103. Sweller J., van Merrienboer J. J. G. and Paas F. G. W. C., (1998), Cognitive architecture and instructional design, Educ. Psychol. Rev., 10, 251–296.
  104. Sweller J., (2005), Implications of cognitive load theory for multimedia learning, in Mayer R. (ed.), The Cambridge handbook of multimedia learning, New York: Cambridge University Press, pp. 19–30.
  105. Sweller J., (2010), Element interactivity and intrinsic, extraneous, and germane cognitive load, Educ. Psychol. Rev., 22(2), 123–138.
  106. Taber K. S., (2017), The use of Cronbach's alpha when developing and reporting research in instruments in science education, Res. Sci. Educ., 48(6), 1273–1296.
  107. Tang H. and Pienta N., (2022), Eye-tracking study of complexity in gas law problems, J. Chem. Educ., 89(8), 988–994.
  108. Tothova M., Rusek M. and Chytry V., (2021), Students' procedure when solving problem tasks based on the periodic table: An eye-tracking study, J. Chem. Educ., 98(6), 1831–1840.
  109. Trautwein U., Lüdtke O., Marsh H. W., Koller O. and Baumert J., (2006), Tracking, grading, and student motivation: Using group composition and status to predict self-concept and interest in ninth-grade mathematics, J. Educ. Psychol., 98, 788–806.
  110. Tse W. S., Choi L. Y. A. and Tang W. S., (2019), Effects of video-based flipped class instruction on subject reading motivation, Br. J. Educ. Technol., 50(1),385–398.
  111. Tuan H. L., Chin C. C. and Shieh S. H., (2005), The development of a questionnaire to measure students' motivation towards science learning, Int. J. Sci. Educ., 27, 639–654.
  112. VandenPlas J. R., Herrington D. G., Shrode A. D. and Sweeder R. D., (2021), Use of simulations and screencasts to increase student understanding of energy concepts in bonding, J. Chem. Educ., 98(3), 730–744.
  113. Van der Meij H., (2013), Motivating agents in software tutorials, Comput. Hum. Behav., 29(3), 845–857.
  114. Van Gog T. and Jarodzka H., (2013), Eye tracking as a tool to study and enhance cognitive and metacognitive processes in computer-based learning environments, in Azevedo R. and Aleven V. (ed.), International handbook of metacognition and learning technologies, New York: Springer, pp. 143–156.
  115. Van Gog T., Kester L., Nievelstein F., Giesbers B. and Paas F., (2009), Uncovering cognitive processes: Different techniques that can contribute to cognitive load research and instruction, Comput. Hum. Behav., 25(2), 325–331.
  116. Van Gog T., Verveer I. and Verveer L., (2014), Learning from video modeling examples: Effects of seeing the human model's face, Comput. Educ., 72, 323–327.
  117. Van Wermeskerken M. and van Gog T., (2017), Seeing the instructor's face and gaze in demonstration video examples affects attention allocation but not learning, Comput. Educ., 113, 98–107.
  118. Wang J. and Antonenko P., (2017), Instructor presence in instructional video: Effects on visual attention, recall, and perceived learning, Comput. Hum. Behav., 71, 79–89.
  119. Wang Y., Liu Q. T., Chen W. L., Wang Q. Y. and David S., (2019), Effects of instructor's facial expressions on students' learning with video lectures, Br. J. Educ. Psychol., 50, 1381–1395.
  120. Wilson K. E., Martinez M., Mills C., D’Mello S., Smilek D. and Risko E. F., (2018), Instructor presence effect: Liking does not always lead to learning, Comput. Educ., 122, 205–220.
  121. Woo H. L., (2009), Designing multimedia learning environments using animated pedagogical agents: Factors and issues, J. Comput. Assist. Learn., 25, 203–218.
  122. Xie K. and Ke F., (2011), The role of students' motivation in peer-moderated asynchronous online discussions. Br. J. Educ. Psychol., 42, 916–930.
This journal is © The Royal Society of Chemistry 2022