The teaching hand in remote accounting education: bringing mirror neurons into the debate

ABSTRACT

This study sheds light on the challenges of communicating knowledge in remote accounting education that has traditionally been demonstrated on a blackboard or whiteboard. In a synchronous online education setting, the study investigated whether students’ learning experience is facilitated by observing the instructor’s hand while examples are demonstrated. Students of two courses in fundamental accounting observed four demonstrations where the conditions instructor’s hand visible/not visible were alternated using a document camera (hand visible) and a writing pad (hand not visible). The study also explored the reasons for students’ preferences for demonstrations with or without visible hands. In terms of theory, the study contributes to previous research on cognitive load theory and social agency theory. Providing helpful directions for instructional design, the findings indicate that observing the instructor’s hand facilitates students’ learning experience to a considerably higher degree than not observing the instructor’s hand. Preferences for demonstrations with visible hands primarily relate to embodied cognition effects that are generated when observing and imitating human movement due to the mirror neuron system. Other explanations relate to social cue advantages – observing the instructor’s hand induces a feeling of connectedness, which in turn enhances cognitive engagement.

KEYWORDS:

• Accounting education
• remote teaching
• mirror neurons
• cognitive load theory
• connectedness
• social agency theory

Introduction

This paper sheds light on one of many challenges that educators and students have faced during the COVID-19 pandemic when faculty members have been forced away from traditional teaching formats and adopted distance learning tools that enable online live or recorded video interaction (Marinoni et al., 2020; Sangster et al., 2020). The paper addresses the challenges of communicating knowledge that has traditionally been demonstrated on a blackboard or whiteboard. In blackboard/whiteboard demonstrations the educator can present a particular line of reasoning stepwise that the students can follow or participate in (cf. Mayer, 2014a). The focus of our study is on observational learning where understanding is assumed to occur when students are able to construct meaningful connections between the verbal and the pictorial representations. Our main research question is whether accounting students’ learning experience is facilitated by observing the instructor’s hand while examples are demonstrated in a synchronous online education setting.

It has been argued that decisions about how to demonstrate the learning material are often made based on fads, convenience, intuition, or personal preference, rather than on rigorous scientific research (Fiorella & Mayer, 2016). Furthermore, Sangster et al. (2020) highlighted the need for accounting educators to have access to legible online whiteboard equivalents in times of distance education in order for students to observe hand-drawn shapes or text, such as T-accounts, in front of their eyes. Fogarty (2020) pointed to the tensions between technology and pedagogy as a consequence of the shift from traditional to online education caused by COVID-19. We find that a lack of scientific guidelines on how to use technology when demonstrating learning material using online whiteboard equivalents motivates an investigation into concrete aspects of remote accounting education that are essential for students’ learning experience.

When knowledge is best demonstrated on a blackboard/whiteboard, that is, when knowledge can be communicated through drawings or writings (as opposed to non-written demonstrations such as chemical experiments or the dissection of animals or plants), different techniques are available. Generally speaking, these techniques can be divided into static visualisations (e.g. still illustrations or slides) or dynamic visualisations (e.g. video animations or videos). In relation to the synchronous online education setting, our focus is on dynamic visualisations where a document camera and a writing pad are used as online equivalents to the whiteboard. In order to answer our research question, we will analyse the learning experiences of students of two synchronous online courses in fundamental accounting when they observe demonstrations of handwritten examples where the instructor’s hand is visible (using a document camera) or where the instructor’s hand is not visible (using a writing pad).

While many previous research studies have focused on comparisons between static and dynamic visualisations (see, for instance, Castro-Alonso et al., 2019 for an overview), only few studies have specifically investigated the benefits of a visible hand in dynamic visualisations (e.g. De Koning et al., 2019; Marcus et al., 2013). These studies have mainly been conducted in laboratory settings. Their findings, moreover, are inconclusive. Given the absence of research on demonstrations with or without a visible hand in synchronous online education in the field of accounting, we employ a rather explorative approach. We ask the following specific research questions:

RQ1: Are demonstrations of fundamental accounting exercises with or without the instructor’s hand easier for students to follow?

RQ2: Do demonstrations of fundamental accounting exercises with or without the instructor’s hand give students a greater understanding of how to perform similar exercises on their own?

Furthermore, we will explore the reasons for students’ preferences by asking the following question:

RQ3: What makes students choose one demonstration technique over the other?

Drawing on insights gained in the field of observational learning, specifically about the implications of mirror neurons for education (cf. Van Gog et al., 2009), we will relate the reasons for students’ preferences for demonstrations with or without visible hands to embodied cognition effects that are generated when observing and imitating human movement. A further analytical dimension is enabled through research on students’ feelings of connectedness in online education, which relate to their engagement in learning (cf. Fiorella et al., 2019; Mayer, 2014c). Both research streams are concerned with questions about what facilitates learning.

Contribution

Our study contributes to accounting education research by bringing the potential impact of mirror neurons on synchronous online education into the debate. More concretely, it provides helpful directions for instructional design in order to enhance students’ learning experience when demonstrating learning material, such as drawings of T-accounts, when using online whiteboard equivalents. Thus, the implications of the paper are not limited to the ongoing pandemic, but can also be helpful in a situation described as the ‘new normal’ when educators have become acquainted with the pros (and cons) of online education.

The paper is organised as follows. The next section reviews the literature on the implications of mirror neurons for education and students’ feeling of connectedness in online education. The research method and results are presented and discussed in the following sections. The conclusion section outlines implications for accounting educators, limitations of the study, and future research options.

Literature review

In this section, we outline assertions from two research streams that both concern observational learning: (1) observing and imitating visualisations may facilitate learning due to the mirror neuron system and (2) a feeling of connectedness between the instructor and the student during learning, established by social cues embedded in visualisations, may facilitate learning. We then present research that has specifically investigated the effect of making the instructor’s hands visible in demonstrations, and arguments for and against the presence of the instructor’s hands in visualisations.

Mirror neurons and their implications for education

Observational learning implies learning by viewing and interpreting the actions of others and has been identified as one of the most powerful mechanisms by which people learn (Bandura, 1986). The discovery of mirror neurons in the early 1990s relates to observational learning because it is assumed to be responsible for our ability to learn by observing and imitating others. More or less by accident, a research team noticed that specific neurons in a macaque monkey’s premotor cortex reacted when it reached for a nut and when it saw scientists reach for it (Di Pellegrino et al., 1992). This observation led to a series of experiments showing that the same brain areas that are involved in the execution of an action also respond to the observation of another individual executing that action. Furthermore, the activation of mirror neurons seems to prime (i.e. prepare the brain for) the execution of similar actions (Van Gog et al., 2009). In recent years, researchers have provided evidence that mirror neurons also exist in humans (e.g. Mukamel et al., 2010). As opposed to monkeys, humans can encode the details of actions, which enables learning by observing not only what someone else is doing but also why the individual is doing it (Rizzolatti, 2005). This means that the mirror neuron system enables individuals to infer intentions of actions. To understand actions is critical for attaining transfer, that is, the application of what has been learned in new tasks or contexts (Mayer, 2014a).

A relevant question concerning the consequences for instructional design is what role mirror neurons play for learning different types of knowledge. Knowledge can be divided into biologically primary knowledge and biologically secondary knowledge (Geary, 2007). Primary knowledge has evolved in humans to help ensure our survival as a species, often involves movement (e.g. to gesture) and can be acquired easily, quickly and with relatively little effort. Secondary knowledge (e.g. to read and write or to understand graphs) has been culturally important to function in contemporary society and tends to be learned slowly as humans have not evolved the mechanisms to acquire it effortlessly. In line with cognitive load theory, which is concerned with how the design of instruction affects working memory and learning (e.g. Paas & Sweller, 2014; Sweller et al., 2011), primary knowledge is easier to learn because it requires minimal cognitive resources, while secondary knowledge places high demands on learners’ limited cognitive capacity (limitations of the working memory). Research thus suggests that demonstrations may be most effective for learning tasks that involve human movement, that is, tasks related to primary knowledge (Ayres et al., 2009; Van Gog et al., 2009). Such visualisations automatically trigger an effortless process of embodied simulations, which primes the execution of similar actions. De Koning and Tabbers (2011) proposed that the benefits associated with the human movement effect should be extrapolated beyond manipulative (biologically primary tasks), and thus be used for instruction promoting secondary knowledge. Because a great deal of educational research has indicated the effectiveness of observational learning for cognitive tasks (e.g. Berney & Bétrancourt, 2016; Castro-Alonso et al., 2019; Höffler & Leutner, 2007), mirror neurons might not only be involved in the learning of motor tasks, but also of cognitive tasks such as technology, engineering and mathematics. Van Gog et al. (2009) outline possible reasons, including the fact that many complex tasks have both a motor and a cognitive component. It has also been suggested that the mirror neuron system is relevant for social cognition, for example in understanding social intentions that do not include movement.

We assume that the fundamentals of accounting relate primarily to biologically secondary or cognitive knowledge. Thus, in order to make the cognitive ‘action’ observable, it needs to be inferred from physical actions (cf. Van Gog et al., 2009). To visualise the logic of double entry bookkeeping by drawing T-accounts enables this inference. An instructor who comments on the drawing of T-accounts can enhance observational learning of the cognitive task of fundamental accounting (cf. Fiorella et al., 2019; Mayer, 2014b). We argue that demonstrations of accounting examples in the above-described way enable observation, imitation, interpretation and prediction related to the mirror neuron system. In other words, the cognitive advantages of biologically primary knowledge are used as a vehicle to learn biologically secondary skills.

Online education and the feeling of connectedness

Already in the late 1980s, Grint (1989) highlighted that students regard the lack of real time (synchronous) communication in distance education as disadvantageous. In distance education settings, students often identify themselves as peripheral rather than full participants. Lake (1999) argued that both physical and psychological isolation might occur. Physical isolation in distance education refers to geographical isolation from material and human resources. Psychological isolation refers to the feeling of segregation from other students and from the instructor and the content and administrative help that the instructor could supply. Research on online education has shown that students who feel connected to others have reduced feelings of isolation, tend to perform better academically, and are more likely to complete their studies (Levy, 2007). High levels of social connectedness and interaction between the instructor and students attribute to higher levels of retention and overall satisfaction of students (LaBarbera, 2013).

Jamison and Bolliger (2020) investigated, among other things, students’ perceptions of connectedness in two online graduate-level business programmes and found that many students asked for more frequent and timely interaction with instructors and synchronous methods of learning and communication. Similarly, the compilation of personal reflections from accounting educators around the world, conducted during the COVID-19 pandemic, showed that students prefer online-synchronous teaching where they are able to observe the instructor in a video as if he/she is teaching an on-campus course (Sangster et al., 2020). In a similar vein, Fogarty (2020) outlined a possible need for encouragement and handholding of accounting students due to a general condition of isolation during remote delivery.

Related to observational learning, social agency theory posits that social cues embedded in the instructors visualisations help establish a sense of social partnership between the instructor and the student (Mayer, 2014c). The sense of partnership or connectedness causes students to feel that the instructor has an interest in their learning, and motivates them to engage in deeper cognitive processing aimed at making sense of the learning material. Engagement and motivation have previously been recognised as the very prerequisites for learning (e.g. Shulman, 2002). Examples of social cues are a personalised language, the visibility of the instructor’s body or face, gestures and eye contact with the camera. Social cues are intended to prime a social response in students that lead to better learning outcomes. Fiorella et al. (2019), for instance, found that eye contact with the camera helps students feel that they are engaged in a partnership with the instructor; they are thus more motivated to make sense of the learning material. However, social agency theory does not make strong predictions about what specific cues should be explicitly present in teaching. Some social cues may be more important than others.

We argue that social agency theory is applicable to remote accounting education. When visualising the logic of double entry bookkeeping by drawing T-accounts, social cues, such as eye contact with the camera, a personalised language or the visibility of the instructor’s face or hands, may be embedded that help students to establish a feeling of connectedness which in turn enables them to engage with the learning material.

Pros and cons of making the instructor’s hand visible

To our knowledge, only few studies specifically investigated the pros and cons of a visible hand in dynamic visualisations. Marcus et al. (2013) and De Koning et al. (2019) examined in laboratory studies whether instructional animations have an advantage over static graphics in hand-manipulative tasks, that is, tasks that require the use of hands. In both studies, the tying of the Truckers’ Hitch knot and the Bowline knot was demonstrated. Both studies concluded that dynamic visualisations (either with or without hands) outperformed the static graphics condition. However, the findings related to the necessity of a visible hand are inconclusive. Marcus et al. (2013) concluded that showing the hands had no effect on the learning of knot tying, but that it had beneficial effects related to efficiency. De Koning et al. (2019) similarly concluded that for animations, viewing the hand or not made no difference for students’ learning outcome. However, the researchers also found that for the Bowline knot, participants that observed demonstrations with hands were significantly less able to reason cognitively about the knot. De Koning et al. (2019) summarised that showing the hands hindered, rather than supported, understanding of the knot-tying task. Both studies concluded that showing hands actually completing a task in an attempt to generate embodied cognition effects might not be required.

There are also studies that relate to non-hand-manipulation tasks, that is, to cognitive knowledge. In a series of four experiments, Fiorella and Mayer (2016) let students view animated videos presenting static and dynamic visualisation of the physical phenomenon of the Doppler Effect. The researchers found that observing the instructor demonstrate with the instructor’s hand visible resulted in better understanding compared to a demonstration where the instructor’s whole body was visible, even though the effect of observing the instructor’s hand draw diagrams was smaller than expected. Nonetheless, the researchers suggested that there may be unique benefits associated with the presence of the instructor’s hand. Schroeder and Traxler (2017) examined the influence of the instructor’s hand in an instructional video compared to a video without the hand. The classroom-based study was conducted in the context of a university physics lesson about friction on inclined planes. The researchers found that students who viewed the video without the instructor’s hand performed significantly better on a learning test and experienced a significantly better training efficiency than the learners who viewed the video with the instructor’s hand. They suggested that the addition of the instructor’s hand might have made the already complex material too difficult to learn and thus inhibited student performance. On the other hand, Schroeder and Traxler (2017) found that the addition of the instructor’s hand was perceived as significantly more engaging than the video where the hand was absent.

One argument outlined in favour of showing the instructor’s hands is that they can offer a processing benefit as individuals have developed an embodied cognitive system that relates to the learning of biologically primary knowledge, appropriate for processing the information that is involved in observing and imitating such movements (cf. Castro-Alonso et al., 2014; Geary, 2007). This argument is in line with the view of cognitive load theorists, who suggest that learning of biologically secondary tasks can be facilitated by directing mental effort towards the content being demonstrated (e.g. Paas & Sweller, 2014). In line with the signalling principle, visualisations with hands could be used to guide attention to specific parts (Fiorella & Mayer, 2016; Mayer & Fiorella, 2014; Van Gog, 2014). Signalling can help students to select, organise and integrate the information presented. Showing hands can also enhance an understanding of how hands are involved in a task, such as their exact position (De Koning et al., 2019). In line with social agency theory (Mayer, 2014c), the presence of the instructor’s hands in visualisations may provide an important social cue that help establish a sense of connectedness between the instructor and the students during learning (Fiorella & Mayer, 2016). In other words, observing human movement may decrease the feeling of isolation in online education. Further, students who imitate a demonstration may experience a process of co-creation; they do something together with the instructor and other students. The feeling of connectedness can motivate students to invest effort towards making sense of the learning material.

Previous research has also outlined possible explanations for detrimental effects of showing hands in visualisations. These relate to so-called extraneous cognitive load, that is, an overload of the learner’s working memory caused by inappropriate instructional design (cf. Paas & Sweller, 2014). Consequently, less cognitive resources can be allocated to the process of making sense of the learning material. One example of a disadvantage of with-hands demonstrations is information redundancy (e.g. Wong et al., 2009). Castro-Alonso et al. (2015), in a study on visualisations of Lego bricks assembling, suggested that there might be detrimental effects of showing the instructor’s hand because the objects used in the demonstration activated the mirror neuron system themselves. For instance, how to pick up and place Lego bricks is unnecessary information. Consequently, a sort of visual overload might occur because the hand’s movements have to be processed in addition to what is of importance – the right position of the Lego bricks (or other to-be-perceived objects). Similarly, De Koning et al. (2019) asserted that there would not be any beneficial effect of visible hands if the hands do not add relevant information to the visualisation in terms of activating brain regions and providing additional cognitive resources. Another explanation for possible detrimental effects of showing the instructor’s hands in visualisations relates to reduced visual clarity (De Koning et al., 2019). This situation occurs, for instance, when the instructor’s hands cover the objects in focus, or parts of them. Consequently, the visible hands result in increased working memory load. Making the instructor’s hands visible may also direct students’ attention to the instructor’s hands and away from the learning material. This type of distraction is referred to as an extraneous social cue that increases cognitive load and may offset or reduce the advantages of social cues (Fiorella & Mayer, 2016; Mayer, 2014c). Schroeder and Traxler (2017) argued that unlikable ‘agents’, which we suggest might take the form of trembling or dirty hands, also might distract students’ learning.

In sum, previous research has suggested both advantages and disadvantages of making the instructor’s hands visible in visualisations. Reasons relate either to the embodied processing of information due to the mirror neuron system (cognitive load theory) or to feelings of connectedness and engagement due to social cues (social agency theory).

Method

In order to investigate whether accounting students’ learning experience is facilitated by observing the instructor’s hand while examples are demonstrated in a synchronous online education setting, students in two online courses in fundamental accounting at a Swedish university were asked a series of questions after the solving of four exercises demonstrated with and without the instructor’s hand. Due to different textbooks, the examples demonstrated in the two courses were not the same, but on the same fundamental level. Acknowledging potential bias, we assume that the exercises in the two courses are comparable. In both courses, the exercises were similar in terms of scope and had a steadily increasing level of difficulty. The courses lasted for more than one week during the spring semester of 2021 and were given successively via Zoom by one and the same educator who has been teaching fundamental accounting for more than ten years. Teaching material such as PowerPoint presentations, chart of accounts, and self-study exercises was accessible via the learning platform Canvas. In order to visualise the logic of double entry bookkeeping, the solving of exercises was demonstrated by drawing T-accounts in a four-field matrix. T-accounts that follow the same pattern (increase or decrease in debit or credit) were drawn in one of the four fields, where the first field is intended for assets, the second for equity and liabilities, the third for revenues and the fourth field for costs (see Figure 1).

Figure 1. Four-field matrix

Further, the solving of exercises was demonstrated using two different technical tools: (1) a writing pad – the Bamboo Slate in combination with the live mode in the Wacom InkSpace App and (2) a document camera – the HoverCam Solo8 together with the HoverCam Flex11 App. Both ways of demonstrating were made observable for the students via the Share function in Zoom. While the first technique (writing pad) only made visible what the instructor’s hand was writing, the second technique (document camera) also showed the hand that was writing. Apart from the visibility of the hand, the intention was to make the demonstrations with and without hands as similar as possible in terms of speed and commitment from the instructor, the tone and volume of the instructor’s speech, the degree of difficulty, the writing style and size, the perspective (top angle for the camera) and the use of a single colour. A small alteration of the way of writing the with-hands demonstrations in the second course, namely a sharper angle of the pen in relation to the paper, was carried out due to students’ answers in the first course. We discuss this alteration in more detail in the discussion section below.

Acknowledging the fact that researchers are impacted, not only by ontological and epistemological convictions (e.g. Chua, 1986), but also, for instance, by cultural, social, ideological and political assumptions, researching own practice requires additional reflexivity (Haynes, 2017). We assume that our research questions can generally be considered quite neutral and ‘unproblematic’ in terms of the above mentioned aspects. The fact that previous research generated mixed findings enabled us to keep an open mind. Also, from an instructor’s point of view, the two alternatives writing pad and document camera did not imply technical differences – in both cases the exercises were written with a pen on a sheet of paper (when using the writing pad, the paper was placed onto the pad). Thus, potential manipulations due to the researchers’ beliefs and experiences are considered quite small.

Informed consent was obtained from students before the start of the study. The enquiry was introduced by three background questions (asked via the Polls function in Zoom) concerning self-assessed prior accounting knowledge, preferred learning method (instruction versus self-study) and preferred educational setting (on-campus versus online education). Subsequently, the instructor demonstrated four accounting exercises. In line with neurological research on the benefits of handwriting (e.g. Ose Askvik et al., 2020), students were encouraged to draw T-accounts on a paper themselves during the demonstrations. Students were also encouraged to perform similar exercises on their own in between the demonstrations. For the first exercise, the writing pad was used. The second and third exercises were demonstrated using the document camera. For the fourth exercise, the writing pad was used again. After these exercise demonstrations, students were asked two questions (via the Polls function in Zoom) about which technique they thought best facilitated their learning. In addition, a question concerning the reasons for their choices was asked (via the Chat function in Zoom). After all three questions were answered, students were informed about the results (Polls results were shared in Zoom) and the technique that the majority of students preferred was then used for the remaining two exercises of the course. All questions are presented in Table 1 below.

Table 1. Background questions and questions asked after the demonstrations

Background questions

How do you assess your knowledge of financial accounting? To be answered on a five-point Likert scale with endpoints anchored by (1) ‘I am a beginner.’ and (5) ‘I have experience of working as an accounting consultant.’

What facilitates your learning – instruction or no instruction provided by the educator? To be answered on a five-point Likert scale with endpoints anchored by (1) ‘I prefer when the educator shows me.’ and (5) ‘I prefer to learn on my own.’

What facilitates your learning – when the lecture is given on campus or via e.g. Zoom (digitally)? To be answered on a five-point Likert scale with endpoints anchored by (1) ‘I prefer when lectures are given on campus.’ and (5) ‘I prefer when lectures are given via Zoom.’

Preferred demonstration technique questions

Which demonstration technique – with or without the instructor’s hand – is easier for you to follow? To be answered on a five-point Likert scale with endpoints anchored by (1) ‘It is easier to follow the demonstration when I see the instructor’s hand.’ and (5) ‘It is easier to follow the demonstration when I do not see the instructor’s hand.’

Which demonstration technique – with or without the instructor’s hand – gives you the greatest understanding of how you can perform similar exercises on your own? To be answered on a five-point Likert scale with endpoints anchored by (1) ‘I have a greater understanding how to perform similar exercises on my own when I see the instructor’s hand.’ and (5) ‘I have a greater understanding how to perform similar exercises on my own when I do not see the instructor’s hand.’

Explanation of choice question

What made you choose your preferred demonstration technique? Open-ended question.

A total of 138 students (115 in course 1; 23 in course 2) were enrolled in the two courses, and 74 students (60 in course 1; 14 in course 2) participated in the study. Gender ratios and age ranges were derived from the enrolment information for the courses and are presented in Table 2.

Table 2. Participants in the study

	Course 1	Course 2	Total
Participants	60	14	74
Male/Female	34/26	3/11	37/37
Age range	20–49 years	21–34 years	20–49 years
Age mean (SD)	24.08 (5.55)	25.86 (3.16)	24.42 (5.21)

Initially, we inspected all data for outliers, non-attempts and irregular distributions. No issues were identified, and all data were included. In addition to the compiled descriptive statistics, we analysed the qualitative data concerning the reasons for a preferred demonstration technique deductively. We grouped students’ answers to the open-ended question into categories related to the two research streams outlined above: (1) mirror neurons and their implications for education and (2) online education and the feeling of connectedness. In order to enhance inter-coder reliability, a colleague from the Department of Pedagogy verified the coding of the answers. In order to investigate reasons for preferred demonstration technique not captured by the open-ended questions, we performed further analyses by means of correlation tests. Using the statistical analysis software SPSS (version 27) and a p-value .05 as an indicator of significance in all analyses, we ran Spearman-Rho tests to measure correlations between self-assessed prior accounting knowledge, preferred learning method and preferred educational setting (background questions) and preferred demonstration technique. In line with the assertion of Castro-Alonso et al. (2019) that gender is a key participant characteristic to consider when investigating instructional design, we also measured correlations between students’ gender and the preferred demonstration technique by using chi-square tests.

Results

Which demonstration technique do students prefer?

The results for the learning experience questions (RQ1 and RQ2) indicate a clear preference for the with-hands demonstrations of fundamental accounting exercises over the without-hands demonstrations.¹ Most students thought that the with-hands demonstration best supported their learning as it was easier to follow and gave them an understanding of how to perform similar exercises on their own. Descriptive statistics are provided in Table 3.

Table 3. Means and Standard Deviations (in brackets) of the preferred demonstration technique for the demonstration of fundamental accounting exercises

	Technique that facilitates students to follow the demonstration	Technique that facilitates students to perform similar exercises on their own
Course 1, n = 60	2.15 (1.537)	2.15 (1.352)
Course 2, n = 14	1.93 (0.961)	1.71 (0.881)
Total, N = 74	2.11 (1.448)	2.01 (1.289)

Figures 2 and 3 show the distribution of the preferred demonstration technique divided into with-hands, neutral and without-hands preferences for the two learning experience questions. A total of 65 percent of all students (67% in course 1; 57% in course 2) perceived that it was easier to follow the with-hands demonstrations (RQ1), while 17.5 percent of all students (21.5% in course 1; 0% in course 2) found that it was easier to follow the without-hands demonstrations. The remaining 17.5 percent of all students (11.5% in course 1; 43% in course 2) were neutral to demonstrations with or without a visible hand.

Figure 2. Distribution of preference for demonstration technique that facilitated students to follow the exercise for course 1, course 2, and both courses together

Figure 3. Distribution of preference for demonstration technique that facilitated students to perform similar exercises on their own for course 1, course 2, and both courses together

Similarly, 63 percent of all students (62% in course 1; 71% in course 2) perceived that the with-hands demonstration facilitated them to perform similar exercises on their own (RQ2), while 15 percent of all students (18% in course 1; 0% in course 2) found that the without-hands demonstration facilitated them to perform similar exercises on their own. 22 percent of all students (20% in course 1; 29% in course 2) were neutral to demonstrations with or without visible hands.

How do students explain which demonstration technique they prefer?

We grouped students’ answers to the open-ended question why they chose the preferred demonstration technique (RQ3) into five categories: hand as embodied processing advantage/disadvantage, hand as social cue advantage/disadvantage, and a neutral category. The category ‘hand as embodied processing advantage’ (60% of all answers) includes answers indicating that students found it easier to make sense of the information thanks to the instructor’s hand. The answers in the ‘hand as embodied processing disadvantage’ category (18% of all answers) indicate the contrary – students perceived the instructor’s hand as disturbing or distracting. The category ‘hand as social cue advantage’ (8% of all answers) contains answers indicating that the hand induced a feeling of connectedness to the instructor during learning, which in some cases also sparked students’ engagement. We found no answers that we could assign to the opposite category ‘hand as a social cue disadvantage’. The ‘neutral’ category (14% of all answers) was dedicated to answers indicating that the hand did not play any role in relation to students’ learning experience. The distribution of students’ answers for both courses, separately and in total, are shown in Table 4.

Table 4. Answers concerning the explanation of preferred demonstration technique per category

	Hand as embodied processing advantage	Hand as social cue advantage	Neutral	Hand as embodied processing disadvantage	Hand as social cue disadvantage
Course 1	60%	10%	8%	22%	0%
Course 2	58%	0%	42%	0%	0%
Total	60%	8%	14%	18%	0%

In order to give a nuanced picture of the students’ explanations for why they preferred the demonstrations with or without a visible hand or neither of them, we will summarise the answers and provide some quotations. Students who preferred the with-hands demonstrations due to reasons that we categorised as ‘embodied processing advantage’ argued that this technique ‘looks more natural’ and ‘feels more real, like an on-campus lecture’. ‘In some way, it feels easier to absorb the information when it is ‘real’ (by hand) rather than technical and ‘unreal’ without the hand’. Students also explained that they found it ‘easier to learn by observing (through sight)’ and due to the procedural, step by step guidance. Some students also pointed to the fact that although the hands sometimes covered the T-accounts, during the with-hands demonstrations it was easier to follow the instructor’s thoughts: ‘it shows you exactly how to do it’. ‘You can see where on the paper the hand is writing. When not seeing the hand, I had to search for where on the paper you were writing.’ ‘When you do it the other way [without-hands demonstration], I think there is a greater risk of missing something you wrote without me noticing it!’ Students also pointed to the fact that the hand helped them to focus on what was important and that it was easier to follow the exercises when the instructor used her hand and pen to point to specific areas because ‘it is easier to see the finger than the pointer’. ‘I recognised that I was more focused on what you were saying during the exercises when the hand was visible.’ Some students also found the instructor’s handwriting cleaner and nicer in the with-hands demonstrations and therefore it was easier to read what the instructor wrote. Others argued that the speed was better and the text larger in demonstrations where the hands were visible. According to some students, it was also easier to separate the with-hands demonstration from the task. Some of the students’ answers explicitly indicate that the mirror neuron system was activated during the with-hands demonstrations: ‘I prefer the with-hands demonstration because I prefer writing by hand myself, but also because I can kind of imagine that it is almost me myself who is writing.’ ‘It is easier to relate to the demonstration because I myself write by hand and with a pen, so it looks the same.’

Students who preferred the with-hands demonstrations due to reasons that we categorised as ‘social cue advantage’, indicate a greater feeling of connectedness during with-hands demonstrations: ‘It becomes more personal as if we are doing the exercises together.’ ‘The hand was more personal and it gives me a greater feeling of that you helped me.’ ‘I get a clearer ‘contact’ with you as an educator when I see something physically moving in reality.’ ‘It feels good that not everything happens digitally; that some things still are personal.’ ‘When I saw the hand, it felt as if I was in the ‘right’ place.’ Students also referred to the fact that the hand made them feel more engaged: ‘It becomes more personal and therefore I think it’s more fun and easier to learn.’ ‘It feels more motivating, especially now in times of distance learning.’

Students’ reasons for a ‘neutral’ choice were that ‘both options worked equally well’, that they were able to follow ‘no matter what’.

Students who preferred without-hands demonstrations due to reasons that we categorised as ‘embodied processing disadvantage’ found it ‘easier to understand the numbers and the way of thinking when the hand was not visible’. Some students stated that the hand was distracting or disturbing: ‘I was more focused on what was written when the hand was not visible. When the hand was visible it became a bit disturbing.’ ‘It’s annoying to wait for the hand to move away in order to see.’ Students also argued that the instructor’s handwriting was nicer and clearer, no shadows were visible and that the text was larger during the without-hands demonstrations.

No students who preferred without-hands demonstrations specified reasons that we could categorise as ‘social cue disadvantage’.

What else explains which demonstration technique students prefer?

The answers to the background questions showed that students assessed their previous knowledge of financial accounting as relatively low and that they thought that their learning was facilitated by instruction provided by the educator (as opposed to self-study) as well as by on-campus lectures (as opposed to online lectures). Descriptive statistics are provided in Table 5.

Table 5. Means and Standard Deviations (in brackets) for the background questions

	Self-assessed prior accounting knowledge	Preferred learning method (instruction vs self-study)	Preferred educational setting (on-campus vs online education)
Course 1, n = 60	1.85 (0.963)	2.03 (1.032)	2.36 (1.032)
Course 2, n = 14	1.07 (0.258)	1.79 (0.860)	1.50 (0.824)
Total, N = 74	1.71 (0.926)	1.99 (1.007)	2.21 (1.052)

In order to assess whether there are explanations for a preferred demonstration technique not captured by the open-ended question (RQ3) and that can be statistically derived, we measured correlations between the background questions including gender and the preferred demonstration technique. Correlation tests show no significant correlations for the background questions including gender and preferred demonstration technique. Correlations are provided in Table 6.

Table 6. Correlations between background questions including gender and preferred demonstration technique (*p < .05, **p < .01, 2-tailed)

		Background questions
		Self-assessed prior accounting knowledge	Preferred learning method (instruction vs self-study)	Preferred educational setting (on-campus vs online education)	Gender
Preferred demonstration technique
Course 1, n = 60	Technique that facilitates students to follow the demonstration	-.022	.034	.019	χ²(4)  =  4.54 ϕ = 0.28
	Technique that facilitates students to perform similar exercises on their own	.023	.159	.010	χ²(4) =  9.18 ϕ = 0.391
Course 2, n = 14	Technique that facilitates students to follow the demonstration	.308	.072	-.145	χ²(2) = 3.96 ϕ = 0.532
	Technique that facilitates students to perform similar exercises on their own	-.232	.210	-.120	χ²(2)  = 1.38 ϕ = 0.314
Total, N = 74	Technique that facilitates students to follow the demonstration	.005	.043	-.013	χ²(4)  = 5.66 ϕ = 0.276
	Technique that facilitates students to perform similar exercises on their own	.049	.176	.039	χ²(4) = 8.09 ϕ = 0.331

Discussion

In line with cognitive load theory, previous research has suggested that learning biologically secondary knowledge can be facilitated by observation and imitation due to the mirror neuron system (e.g. Paas & Sweller, 2014; Van Gog et al., 2009). In demonstrations, the cognitive advantages of biologically primary knowledge are used as a vehicle to learn biologically secondary skills (De Koning & Tabbers, 2011). Furthermore, according to social agency theory, social cues embedded in demonstrations help establish a feeling of connectedness between the instructor and the student, which motivates the student to engage in deeper cognitive processing (e.g. Mayer, 2014c). The findings of research studies that directly investigated possible advantages of with-hands demonstrations over without-hands demonstrations relate to these theories, but are inconclusive. While some researchers argued that showing hands actually completing a task in an attempt to generate embodied cognition effects might not be required or might even inhibit students’ performance (De Koning et al., 2019; Marcus et al., 2013; Schroeder & Traxler, 2017), others suggested that there might be unique benefits associated with the presence of the instructor’s hand (Fiorella & Mayer, 2016). With-hands demonstrations have also been perceived as significantly more engaging than without-hands demonstrations (Schroeder & Traxler, 2017).

Our study showed that with-hands demonstrations of fundamental accounting exercises had a positive effect on most students’ learning experience. We measured learning experience by asking whether demonstrations with or without the instructor’s hand were easier for students to follow (RQ1) and whether demonstrations with or without the instructor’s hand gave students a greater understanding of how to perform similar exercises on their own (RQ2). An analysis of the answers to the third research question, regarding what makes students choose one demonstration technique over the other, indicates that the instructor’s hands can enhance the positive processing effects related to visualisations of biologically secondary knowledge such as fundamental accounting. Thus, there are strong indications for embodied cognition effects of showing the hands. An important explanation seems to be the signalling principle (cf. Fiorella & Mayer, 2016; Mayer & Fiorella, 2014; Van Gog, 2014). The instructor’s hand seems to be more suitable to guide students’ attention to specific parts of the exercise compared to the pointer. But also the instructor’s hand itself (without pointing at any specific parts) seems to facilitate students’ learning of the logics of double entry bookkeeping, because students can observe, imitate and make sense of in which field of the four-field matrix the hand is writing. These observations support cognitive load theory that suggests that easily acquired biologically primary knowledge can be used to leverage the acquisition of biologically secondary knowledge, here enhanced through the instructor’s hand. Intuitively, one might assume that students who prefer demonstration by the instructor in general (as opposed to self-study) also prefer with-hands demonstrations. A correlation test between preferred learning method and preferred demonstration technique did not show any significant correlations, however.

Students that perceived the hand as disturbing or distracting (only course 1) mainly referred to what has previously been described as reduced visual clarity (De Koning et al., 2019), as these students explained that the hand covered part of the exercises. The instructor tried to avoid that when demonstrating exercises for students in course 2, which might explain why no students participating in course 2 preferred the without-hands demonstrations. We found no instances of preference for without-hands demonstrations due to information redundancy among students’ answers to the question why they chose the preferred demonstration technique (c.f. De Koning et al., 2019). We argue that the absence of this detrimental effect strengthens the benefits of the with-hands demonstrations because what has been remarked as a negative effect of making the instructor’s hands visible (reduced visual clarity) can to a large extant be avoided.

Our study also indicates that the instructor’s hand in demonstrations of fundamental accounting exercises can imply a social cue that sparks not only a feeling of connectedness between the instructor and the students during learning, but also a higher level of motivation. Some students’ answers to the question why they chose the preferred demonstration technique refer explicitly to these feelings. This is in line with findings in previous research conducted by Schroeder and Traxler (2017). No students who preferred the without-hands demonstrations referred explicitly to a preferred distance to the instructor. We thus suggest that the hand can have positive effects in line with social agency theory (cf. Mayer, 2014c), and that negative effects are rare. As Mayer (2014c) suggested that cognitive load might offset or reduce the advantages of social cues, we argue that students who found the instructor’s hand disturbing or distracting might have refrained from thoughts about feelings of connectedness or isolation. Previous research has shown that a feeling of connectedness is of particular importance in online education settings (e.g. Jamison & Bolliger, 2020; Sangster et al., 2020). Thus, intuitively, one might expect that students who prefer campus education over online education also prefer with-hands demonstrations due to a greater feeling of connectedness. However, a correlation test between preferred educational setting and preferred demonstration technique did not show significant correlations.

Some students found that the demonstrations with and without a visible hand facilitated their learning equally well. Previous research has suggested that prior knowledge might affect students’ learning in visualisations (Fiorella & Mayer, 2016; Mayer & Fiorella, 2014; Van Gog et al., 2009), and intuitively one might expect that students with prior knowledge of fundamental accounting are indifferent to the demonstration technique. However, a correlation test between self-assessed previous accounting knowledge and preferred demonstration technique did not show significant correlations. We therefore have no indications for why these students chose a neutral answer. A possible explanation is that, on average, these students needed less mental effort to make sense of fundamental accounting compared to their fellow students. Another possible explanation is that some students may have increased their knowledge by studying the course literature during the course, while others relied on lectures only.

Conclusion

In the present study, we investigated whether accounting students’ learning experience is facilitated by observing the instructor’s hand while examples are demonstrated in a synchronous online education setting. The findings, that are based on the answers of 74 students, show that with-hands demonstrations to a considerably higher degree than without-hands demonstrations create a positive learning experience when the logic of fundamental accounting is visualised by drawing T-accounts in a four-field matrix. The analysis of students’ reasons for their preferred demonstration technique reveals that the advantages of making the instructor’s hands visible can be related to two research streams within observational learning. First, in line with cognitive load theory, learning not only biologically primary knowledge but also biologically secondary knowledge – such as fundamental accounting – can be facilitated by observation and imitation due to the mirror neuron system. Our study indicates that including the instructor’s hands in demonstrations enhances the embodied processing advantages of observing and imitating human movement for a majority of students. Second, in line with social agency theory, social cues embedded in demonstrations can establish a feeling of connectedness between the instructor and students during learning, which in turn motivates students to engage in deeper cognitive processing and thus facilitates learning. Our study indicates that the instructor’s hands can serve as a social cue and therefore enable a greater learning experience. Furthermore, an important finding is that there are no indications of information redundancy in with-hands demonstration and no direct indications for an offset or reduction of the advantages of social cues. These instances further strengthen the arguments for with-hands demonstrations.

Practical implications

A number of practical implications can be derived from our study. These implications may provide helpful direction for the instructional design of synchronous online education in fundamental accounting. Firstly, showing the instructor’s hands completing accounting examples where T-accounts are drawn in a four-field matrix is recommended. Benefits of such with-hands demonstrations relate to both the cognitive architecture of the human brain where mirror neurons play an important part for learning and to a feeling of connectedness that should be furthered in online education settings. For such visualisations, a document camera seems to be an appropriate device. Secondly, in order to enhance the benefits of with-hands demonstrations, we suggest that accounting educators advise their students to imitate the instructor while demonstrating. Although we did not explicitly study the effect of immediate imitation, our study indicates that students’ learning experience seems to be enhanced when they draw T-accounts by themselves while observing the instructor. Thirdly, reduced visible clarity, that is, the instructor’s hands covering parts of the exercise, should be avoided. Students in one of the courses in our study who preferred without-hands demonstrations mainly referred to reduced visual clarity. During the second course, the instructor tried to avoid covering essential parts of the exercises. Students in this course did not refer to reduced visual clarity, which indicates that a positive learning experience can be achieved for even more students when the instructor’s hand is visible.

Limitations and future research

As with all studies, the results of this research should be viewed in light of its limitations. Our investigation included two groups of students at one Swedish university. Expanding the study to other universities in different parts of the world could help examine the efficacy of the findings over a broader demographic. Future research might also further explore variables not highlighted in previous research on demonstrations with or without hands that affect students’ preferences, such as grading marks from prior studies or whether students drew T-accounts on a paper themselves during the demonstrations. Furthermore, while there are relatively strong indications of embodied processing advantages in with-hands demonstrations that can be related to the mirror neuron system, actual brain activity has not been measured and students’ direct imitation of the instructor’s drawings has not been controlled for. Neither have learning outcomes, such as retention or transfer (cf. Mayer, 2014a), been measured. Although we assume that students’ positive learning experience can lead to positive learning outcomes, to measure learning outcome, as has been done in previous research (e.g. De Koning et al., 2019; Marcus et al., 2013; Schroeder & Traxler, 2017), can provide further insights into the pros and cons of demonstrations with and without a visible hand. In short, the with- or without-hands variable might not have the same effect on learning outcome as it has on learning experience.

A further limitation is related to what we recognise as rather beneficial – the real online-class scenario. While this setting enables helpful directions for accounting educators, it also comprises possible bias. We tried to make the demonstrations with and without a visible hand as comparable as possible, but admit that it is hard to avoid ‘noise’ such as performing the exercises with exactly the same speed as well as tone and volume of the instructor’s voice. This bias is related to the fact that the educator who demonstrated the exercises might unconsciously have acted in a way that made one technique appear superior to the other. At what point in time the exercises are demonstrated during a lecture, for instance just before or after a break, might lead to further bias. Also, we learned from the demonstrations in the first course and adjusted the way of drawing T-accounts to avoid reduced visual clarity, that is, to avoid covering essential parts of the exercises demonstrated in the second course. Highly controlled between-subjects experimental designs, such as those performed by, for instance, Marcus et al. (2013) and De Koning et al. (2019), enable high internal validity, but sacrifice external validity.

Future studies might consider what happens when hands that are not likeable in appearance, such as trembling or dirty hands, are included in visualisations (cf. Schroeder & Traxler, 2017), or under what circumstances advantages of social cues are offset or reduced (cf. Fiorella & Mayer, 2016). Further suggestions for future research include investigations of the relative effectiveness of with-hands demonstrations for students’ learning experience and outcome to without-hands demonstrations in more advanced accounting areas.

Acknowledgements

Thanks to the two anonymous reviewers and the editors for their insightful comments and suggestions. Thanks also to Ronja Bohnenkamp at the University of Rostock, who provided helpful advice. We would also like to thank the Centre for Teaching and Learning at Karlstad University for technical support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Correlation tests between the two learning experience questions show significant correlations for both courses, separately (course 1: ρ = .857**; course 2: ρ = .832**) and in total (ρ = .85**), where **p < .01, 2-tailed.