Investigating students’ meaning-making of multiple visual representations of epigenetics at different levels of biological organisation

ABSTRACT

The aim of this study is to investigate students’ meaning–making of multiple visual representations of epigenetics at different levels of biological organisation, and to discern what visual aspects of the multiple visual representations might influence students’ reasoning. Adopting an exploratory approach, we analysed how students made meaning of visually communicated epigenetics phenomena while pointing at and reasoning about the multiple visual representations as part of semi-structured focus groups. We investigated students’ meaning-making of the multiple visual representations by analysing their indications through physical pointing and accompanying verbal utterances. The analysis revealed meaning-making and the nature of linking between levels of organisation in four distinct patterns, namely intra horizontal linking, inter horizontal linking, one level vertical linking and two level vertical linking. In addition, five different visual characteristics of the multiple visual representations emerged as influencing students’ reasoning while linking between different organisation levels: multiple visual representations, salient visual features, analogous visual features, familiar visual elements, and textual adjuncts. The study shows that multiple visual representations at different levels of organisation can support students’ meaning-making of epigenetics, indicating that this way of communicating can be transferable to other biological domains. Potential implications for future research and teaching practice are provided.

KEYWORDS:

• Levels of biological organization; visual representations; yo-yo reasoning

Introduction

Building an understanding of biological phenomena across different levels of biological organisation is essential for developing a coherent conceptual understanding of biology (Knippels & Waarlo, 2018). However, integrating knowledge across different levels of biological organisation is challenging for students. To help facilitate students’ understanding of biological phenomena, Knippels (2002) has proposed the yo-yo strategy, the ability to develop coherent conceptual understanding and to approach the abstract and complex nature of different biological phenomena and processes at different levels of organisation. Nevertheless, there remains a lack of empirical studies investigating how to induce this in biology, and in science education at large. Knippels and Waarlo (2018, p. 16) affirm this position in a recent review by stating, ‘yo-yo thinking as a general idea needs more research-informed elaboration so as to transform adoption of this metacognitive tool into its implementation’.

Using multiple visual representations to communicate biological phenomena that occur at levels imperceptible to the naked eye provide opportunities for students’ meaning-making of complex and abstract structures and processes (Schönborn & Anderson, 2009). Moreover, multiple visual representations are tools that can bridge the gap between our visual perception and extracting intended understanding of complex processes (Moreno & Mayer, 2007). In addition, Ainsworth (2006) argues that multiple visual representations can facilitate learning of complex scientific ideas that span several organisational levels.

Epigenetics is one of the fastest growing areas in the life sciences with important societal implications for teaching at secondary school (Gericke, 2021). Epigenetic mechanisms at the molecular level explain how environmental factors can potentially influence gene activity and thereby human characteristics such as health and the development of diseases at the macro level. Hence, epigenetic phenomena often refer to and intertwine different levels of biological organisation. Therefore, epigenetic phenomena as presented in multiple visual representations serve as a relevant case for exploring challenges around students’ meaning-making and linking of biological processes and concepts at different levels of organisation. Consequently, in this study, we investigate 15–16 year-old secondary students’ meaning-making about epigenetics when presented with multiple representations of epigenetics at different levels of organisation. In doing so, we explore if and how multiple visual representations can induce students' yo-yo reasoning in their meaning-making of multiple visual representations of epigenetic phenomena, and what aspects of the multiple visual representations might trigger this knowledge development.

Background

The importance of navigating levels of biological organisation

Translation across multiple representations of biological knowledge is a common challenge in science education (Tsui & Treagust, 2013). However, in practice, these issues are described and applied differently in physics, chemistry, and biology. In physics, the use of multiple representations is closely related to the nature of physics phenomena themselves, which often involves converting physics modelling into mathematical modelling, and vice-versa (Opfermann et al., 2017). Such translations are crucial for learning and teaching relationships in physics that rely on interpreting mathematical representations in terms of several representational forms. In chemistry education, the different levels are often articulated in terms of symbolic, molecular, and macro aspects, often articulated as the chemistry triplet (Johnstone, 1991). Johnstone (1991) addresses the macro component with visible and tangible concepts, the molecular component with invisible concepts (atoms, molecules, and ions) and the symbolic component for abstract representations (symbols, formulae, equations, and graphs). Biology is also unique since it is described by hierarchically organised levels that contain different biological entities (Tsui & Treagust, 2013). Biological organisation, the focus of this study, refers to a heuristic comprehension of complex systems (Mayr, 2004). Here, biological organisation can be viewed as greater than the sum of its constituent parts – a biological phenomenon can be described by how the various components interact and that all biological systems are ordered systems. Mayr (2004) argues that a complex biological phenomenon has to be understood and studied separately in smaller components such as tissues further reduced into cells. No higher-level biological phenomenon can be fully understood until it has been analysed into the components of the next lower level; this process is to be continued downward to the level of the purely physicochemical components and processes. In contrast, the composite wholes have properties that are not evident in their components, in what Mayr (2004) defines as emergence. Mayr (2004) also asserts that knowledge of the components at the lowest level permit the reconstruction of all higher levels.

Within biology education research, Bahar et al. (1999) has applied thinking about multilevel representations to biology by describing the macro level as referring to organisms (population, individual), the micro level as referring to cells, and the biochemical level to DNA. Tsui and Treagust (2013) argue that if students are to make meaning of biological phenomena they need to often incorporate all these levels (macro, micro and sub micro level) given that the nature of biological entities is embedded into multiple hierarchical levels of organisation. Therefore, to acquire a complete understanding of biological phenomena, these three levels of representation need to be interlinked (van Mil et al., 2016). With regard to systems thinking in genetics education, Marbach-Ad and Stavy (2000) also address these three levels, and relate the macroscopic level with structures, i.e. biological traits, that are visible to the naked eye, the microscopic level with structures visible under a light microscope, and the sub microscopic level with molecules, such as DNA and proteins. In relation to scale, Meijer et al. (2009) visually communicated how biological levels of organisation can be approximated to macroscopic (10⁻⁰–10⁻⁴), microscopic (10⁻⁴ –10⁻⁶) and sub microscopic (10⁻⁶ −10⁻⁸) magnitudes (Figure 1). In this study, we draw on all three of these levels.

Figure 1. Visually representing levels of biological organisation in relation to three levels of scale (metres), namely the macro level (3), the micro level (2), and the sub micro level (1).

Yo-yo reasoning for facilitating meaning-making of biological phenomena

Duncan and Reiser (2007), van Mil et al. (2016) as well as Knippels and Waarlo (2018) argue for the need to link organisational levels to promote meaningful and coherent conceptual understanding of biological phenomena, and to structure the learning and teaching process. Knippels (2002) proposed an approach referred to as the yo-yo strategy to make this explicit to students by repetitively linking between different biological organisational levels during teaching. The idea is that learning biological phenomena represented at different biological organisational levels can be supported if students are able to apply the yo-yo strategy themselves when thinking and reflecting on different biological phenomena (Knippels & Waarlo, 2018). Here, we denote the ability to reason between the levels as yo-yo reasoning. Van Mil et al. (2016) elaborate on this issue further by showing that knowledge of molecular properties at the sub micro level that explains cellular systems at the micro level can be further extrapolated as explanations at the macro level. Such an upward explanation is of utmost importance to learn about biological phenomena. Hence, molecular properties, interactions, and behaviours at the sub micro level are used as causal tools to functionally explain aspects of biological phenomena as a whole. Van Mil et al. (2013) introduced the term molecular mechanistic reasoning to refer to the reasoning skills needed by students to construct and understand these functional domain-specific forms of mechanistic explanations. In this study, mechanistic reasoning is included in our definition of reasoning, but we also include reasoning underlying downward linking through different biological organisation levels in which entities at higher organisation levels are used to explain entities at lower levels. Therefore, we use the more general term reasoning instead of mechanistic reasoning. It follows that if students can link upwardly and downwardly through different levels, where they are able to explain and interlink between the different biological organisational levels, would signal their meaning-making of a biological phenomenon.

In this study, we use the concept of meaning-making rather than learning or understanding because we are investigating students’ visual and oral interactions with multiple visual representations. In this sense, meaning-making is described as the process of how people construe, understand, or make sense of life events, involving the development of the person’s distinction between self and other (Kegan, 1980). Since meaning-making occurs through human thought, feelings, and actions (Kegan, 1980) we observe meaning-making as a process of reasoning and pointing when students interact with the multiple visual representations (cf. Pettersson et. al., 2022). In doing so, we explore how students reason while linking across and between different levels of biological organisation. Verhoeff et al. (2008) suggest that it is important to distinguish and match biological concepts to specific levels of biological organisation as well as display horizontal coherence (interrelate concepts at one level) and vertical coherence (interrelate concepts with concepts at higher levels of organisation). Verhoeff et al. (2008) also demonstrated that teaching specific modelling skills can promote students’ acquisition of knowledge through levels of biological organisation.

Adding to the perspective of the role of multiple visual representations in pursuing conceptual understanding in biology, Schönborn and Bögeholz (2009) argue that meaningful learning in biology is fostered by horizontal and vertical transfer (induced by translation across different visual representations) processes when learners integrate knowledge at different levels of biological organisation. Schönborn and Bögeholz (2009) define horizontal transfer as applying knowledge across different visual representations communicated at the same level of biological organisation. Vertical transfer requires connecting and applying knowledge between differently visually communicated levels of biological organisation. In this regard, translation captures the ability to link across, in a multi-directional manner, and connect multiple visual representations that represent an underlying biological concept, principle or process at a particular level of biological organisation (also see Ainsworth, 2006). In this study, students’ meaning-making is explored by observing student gestures through pointing combined with their accompanying uttered reasoning. The process of pointing at the multiple visual representations is explored as linking across and between different levels and thereby connecting the multiple visual representations representing different organisation levels with the underlying epigenetic phenomenon.

Multiple visual representations for supporting meaning-making of biology phenomena at different organisational levels

Interpreting multiple visual representations helps bridge the gap between human visual perception and integrating knowledge about complex processes (Moreno & Mayer, 2007). Through visual communication, students acquire meaning of scientific ideas and biological phenomena that are challenging to express verbally. Therefore, multiple visual representations are useful tools to visualise biological phenomena at different levels of organisation, especially at the micro and sub micro level (Rundgren & Tibell, 2010). Consequently, visual representations play a critical role in influencing how students reason about biological phenomena (Schönborn & Bögeholz, 2009). With the increasing emergence of multiple visual representation in many forms of contemporary science education, discerning which aspects of visual representations facilitate or hinder meaning-making and learning about biology remains an important question in science education research (Mijung & Qingna, 2022).

Schönborn and Anderson (2009) have identified three principal factors affecting students’ ability to gain knowledge from a visual representation. These factors comprise the existing conceptual understanding that students ‘bring’ to the visual representation, students’ cognitive and reasoning skills needed to acquire meaning from the visual representation, and lastly, the visual mode and nature of the graphical markings constituting the external representation itself. In this study, we focus particularly on this last factor in exploring what visual aspects of the multiple visual representations might influence students’ reasoning at different levels of organisation.

The effectiveness of a representation involves both the nature of the visual information provided in the representation and how it is communicated (Ainsworth, 2006). Ainsworth (2006) argues that multiple representations have three key functions, namely: complementary functions (with different processes or different information), construction of deeper understanding (the learner extends knowledge from a known to an unknown representation) and relational understanding (associates two representations without reorganising knowledge). All these processes are important factors in a biology teaching and learning setting.

Tsui and Treagust (2013) define meaning-making of multiple visual representations in biology in terms of three dimensions. Firstly, in terms of the modes of multiple visual representations, secondly, in terms of represented biological organisation levels, and thirdly, in terms of the domain knowledge of biology. Simultaneously, with the level of abstraction of external representations, visual representations can be classified into a variety of modes from abstract to realistic representation of phenomena (Schönborn & Anderson, 2006). Symbolic modes are abstract and constitute one of the most common modes of visual representations in science education (Gilbert, 2010). Symbols in science education function as visual stimuli to represent a mechanism or concept of a phenomenon. The symbolic mode is often used in visual representations that depict entities that are too small to be seen with a light microscope, the submicroscopic level. A realistic mode depicts a phenomenon as perceived by the naked eye, i.e. at the macro level, while semi pictorial visual representations are positioned between the two (Schönborn & Anderson, 2006). Semi pictorial modes are commonly used to represent entities at the micro level since microscopic pictures often need descriptions and symbolic explanations to be understood. In this study we explore visual representations in realistic, semi pictorial- and symbolic modes because they commonly represent the different levels of biological organisation.

Epigenetics as an arena for exploring yo-yo reasoning

Meloni and Testa (2014) claim epigenetics to be a molecular juncture between nature and nurture and thus as a new critical frontier in the social studies of the life sciences. Consequently, epigenetics is becoming an important part of the genetics component of modern biology curricula (Gericke & Mc Ewen, 2023). Epigenetics is a genetic process during which genes are expressed differentially and selectively due to the influence of environmental factors (Nicoglou & Merlin, 2017). Methylation is a common epigenetic mechanism in which methyl groups attach to the DNA-strand, which leads to transcription being hindered, rendering the gene deactivated. In this study, we considered methylation as an underlying mechanism in epigenetics. Epigenetics explains how environmental factors at the macro level can influence gene activity at micro and sub micro levels, thereby explaining biological traits at the macro level such as health and disease development. It follows that students need to reason across and between different organisation levels to understand epigenetics (Gericke, 2021), which also draws on a need for systems thinking. Epigenetics serves as a compelling case to investigate students’ meaning-making about multiple visual representations of a phenomenon.

Aim and research questions

The literature indicates the importance of building conceptual knowledge at different levels of organisation to acquire a coherent conceptual understanding of biology (e.g., Ainsworth, 2006; Bahar et al., 1999; Knippels, 2002; Knippels & Waarlo, 2018; Schönborn & Bögeholz, 2009). Multiple visual representations that represent epigenetic phenomena at different levels of organisation might induce students to link across and between the levels in their meaning-making of epigenetics. The aim of the study is to investigate students’ meaning–making of multiple visual representations of epigenetics at different levels of biological organisation, and to discern what visual aspects of the multiple visual representations might influence students’ reasoning.

Specifically, we pose the following research questions (RQ):

1. How do students link across and between multiple epigenetic visual representations at different levels of organisation as revealed by their pointing and reasoning?

2. What visual characteristics of the representations influence students’ reasoning when linking across and between different levels of biological organisation?

Methods

The overall study design consisted of students first being introduced to genetics and epigenetics concepts. Then, students’ interacted with their fellow peers in focus groups, and with a selection of multiple epigenetic representations in different visual modes and levels of organisation. Students’ pointing and reasoning were filmed, fully transcribed and analysed to unpack the meaning-making process.

Study setting

Five focus groups comprising two to four students (a total of 13 students, 1 boy and 12 girls) in grade 9 (aged 15–16 years old) in a Swedish compulsory lower secondary school participated. An advantage of focus groups this size is that every student has an opportunity to express themselves while also discussing with peers (Flores & Alonso, 1995). Also, focus groups provide an opportunity to observe rich interaction on a topic in a limited period of time (Morgan, 1997).

Study design

First, the students were introduced to the study objectives and sequence of activities. A film clip¹ was then presented to introduce genetics to induce a common base of understanding among students. Examples of communicated content in the film included that human cells contain a nucleus with 46 chromosomes. Each chromosome contains a long strand of DNA, and sections within DNA are termed genes that contain instructions for synthesising proteins. Processes of transcription and protein synthesis were also communicated.

A week after the introduction, students who volunteered to participate in the study were assigned to focus groups. Each group commenced with students viewing another film clip.² This film introduced the students to epigenetic processes conveyed through real life experiments with mice and in an animation.

The film communicated how the methylation process inactivated genes in one of the mice, which caused a normal weight, while the other mouse developed obesity. Environmental epigenetic effects were discussed in the film. Showing the film clips to the students introduced them to a subject that they possibly where not exposed to previously.

After viewing the film clip, the students were seated round a table and exposed to and briefed about eight visual representations of complementary functions that communicated epigenetic phenomena at sub micro, micro and macro levels of organisation through different modes of representation (also see description below). The multiple visual representations were positioned and fixed at the centre of the table. In response to a semi-structured focus group protocol (see Appendix 1) executed by the first author, the students reasoned about the multiple visual representations for approximately 20 min. The students were encouraged to use a provided pointer, and/or their fingers to point at what visual features of the multiple visual representations they were referring to during their reasoning. As part of the semi-structured approach, follow-up questions were formulated to further explore emerging lines of reasoning. The protocol was informed by the approaches of Knippels (2002), Knippels and Waarlo (2018), Schönborn and Anderson (2009) and Venville and Treagust (1998).

Epigenetic visual representations selected for the study

Multiple visual representations are used to support complementary functions (Ainsworth, 2006). Thus, each representation has to be understood independently as well as being understood in relation to the others. Therefore, a main design consideration in this study was to select representations that all connect to an epigenetic phenomenon to be explained, as well as doing so at different levels of organisation. Consequently, the multiple visual representations used in the study were chosen to represent different levels of biological organisation through different modes of representation (Table 1). Realistic images included young and older twins communicated at the macro level (V1, V3), together with semi-pictorial representations of chromosomes (V2, V4) showing methylated and de-methylated chromosomes, at the micro level. Symbolic ‘switches’ that represented the sub micro level were provided to communicate the activation and deactivation of genes on the DNA sequence (V5). At the sub micro level, a semi pictorial and symbolic visual representation were selected, illustrating methylation leading to deactivated DNA molecules (V6) and the termination of protein synthesis. The environmental factors were chosen as a salient communication of nature – nurture processes represented at the macro level (V7). Finally, two genetically identical mice adopted from the viewed film clip, represented realistic images at the macro level (V8). The set of multiple visual representations communicates components of epigenetics across all biological organisation levels and modes of representation.

Table 1. Set of multiple visual representations used in the study that represent aspects of epigenetics at three levels of organization (macro, micro and sub micro) through three modes of representation (realistic, semi-pictorial and symbolic)^a.

Note: ^aDue to copyright, V1 and V3 are slightly different from the versions used in the original data collection. Nevertheless, they are highly similar to the originals that visually communicate epigenetic differences between identical twins at the macro level.

Data collection and analysis

The implemented analytical approach was multi method and exploratory. Robson and McCartan (2015) argue that video recording is recommended in focus group studies due to their complexity, and in our study design we sought an analysis of verbal, visual and behavioural (physical pointing) data. Therefore, analysis of videotaped data of the filmed groups was conducted, in which we analysed students’ pointing, interactions and uttered reasoning in first person voice in relation to respective students’ accompanying pointing at the multiple visual representations. Analyses of the student reasoning focuses on when pointing occurs. The video camera was positioned so that all the visual representations, and each respective individual’s interactions, could be captured and observed (see Tables 4–8).

The analysis of the video data was conducted in two phases. In the first phase, student interaction with the multiple visual representations was analysed in terms of pointing behaviours – events where students used a provided pen, stylus, or their finger(s) to explicitly point to visual features while making meaning of the multiple visual representations. Each pointing event was coded, identified and treated as a measure of students’ indication of a visual feature under meaning-making. In addition, all the pointing events were summed and their number of links quantitatively described relative to the visually communicated organisation level of the respective visual representation. We methodologically identified students’ linking across and between biological organisation levels in four ways. Pointing can occur at the same level of organisation, when students point at, and within the same single visual representation (termed intra horizontal linking), or across different representations at the same level (termed inter horizontal linking). We identify students’ pointing at an upward or downward level of visual organisation as one level vertical linking. Thereafter, if students are observed pointing at a further upward or downward represented level, this is referred to as two level linking. In addition, the chronological sequence of observed pointing events in relation to linking across or between the organisational levels was recorded and described in time-level diagrams (see Figure 4). Also, students’ reasoning in relation to their linking patterns was analysed to identify what potential meaning-making was associated with respective linking patterns and multiple visual representations based on students’ reasoning when linking the visualisations to each other.

In the second phase, we investigated students’ reasoning when they linked across (horizontally within the same, or across different representations at the same organisational level), or between one organisational level (pointed at a visual representation at one organisational level) to another (pointed at another visual representation representing a different organisational level). Students’ reasoning when indicating a visual representation was linked to the respective measuring points. The film was paused at each pointing event at a visual representation and the accompanying uttered student reasoning was analysed to qualitatively identify what features or characteristics of the multiple visual representations triggered any upward or downward level linking (yo-yo reasoning), between the multiple visual representations. These codes were categorised into themes that discerned what visual characteristics of the multiple visual representations influence students’ reasoning when linking across and between different levels of biological organisation.

Themes and patterns emerged from the data naturally and were then defined iteratively (Robson & McCartan, 2015; Mayring, 2000; Figure 2). In step 1, we searched collected video data for meaning-making units that indicated reasoning that induced the students’ linking between levels or between the different visual representations. In step 2, these units were condensed by elucidating patterns of similar characteristics within the multiple visual representations that induced the linking. In step 3, the condensed units were clustered into themes while identifying similarities in expressed characteristics of the multiple visual representations. In step 4, these themes were re-checked in order to affirm students’ reasoning with the multiple visual representations in relation to the emergent themes. Step 1–4 were performed in repeated steps to yield emergence of the eventual themes. In step 5, the themes were labelled and condensed into narratives that provide exemplars of how students linked between the visualisations representing different levels of organisation when they pointed at and reasoned about them.

Figure 2. Analytical process used to identify visual characteristics that influence students' reasoning with the multiple visual representations.

Results

The findings of the study are presented by first addressing RQ1 in the form of four linking patterns and responding to RQ2 in relation to five themes of characteristics of the multiple visual representations that induce linking.

Students' linking across and between multiple epigenetic visual representations at different levels of organisation as revealed by their pointing and reasoning

Results from the focus group sessions showed that students referred to multiple visual representations representing all three organisational levels, and thereby expressed yo-yo reasoning in their meaning-making of the representations. Interestingly, the sub micro level was the most pointed to level in which students exposed most of their reasoning (Figure 3). This is a surprising result when one considers the abstractness of visual representations at the sub micro level (L1).

Figure 3. Frequencies of students’ indicated (through pointing) level in relation to each organisation level.

The time-level diagrams in Figure 4 provide examples of emergent patterns of how two different student groups linked between and/or remained at different levels while reasoning with the multiple visual representations. The revealed linking patterns include intra horizontal linking, inter horizontal linking, one- and two level vertical linking (also see Table 2 and 3). When the students remained at and reasoned about one or several visualisations at the same organisation level, these were categorised as horizontal linking (e.g. time sequence 05:40–07:00 in group 4 in Figure 4). Intra horizontal linking refers to repeated pointing within the same visualisation, while inter horizontal linking refers to pointing across different multiple visual representations that represent the same organisation level. One level vertical linking is the linking upward from a lower organisation level (+1) or downward from a higher organisation level (−1). For example, at 04:27 group 5 (Figure 4) performs a one level upward linking (+1), and then a one level downward linking (−1). Finally, two level vertical linking is the linking upward from the sub micro level directly to the macro level (+2) or downward from macro to sub micro (−2). An example of this linking occurs at 12:21 in group 5 (Figure 4).

Figure 4. Time-level diagrams of linking patterns of two student groups (group 4, top, and group 5, below). The elapsed time during focus group discussions is presented on the x-axis and organisational level (1 Sub micro level, 2 Micro level, and 3 Macro level) on the y-axis.

Table 2. Horizontal linking across multiple visual representations representing the same organisational level.

	Intra horizontal linking (number of links)	Inter horizontal linking (number of links)
Macro level (L3)	23	34
Micro level (L2)	16	11
Sub micro level (L1)	74	32
Total	113	77

The number of and distributions of the horizontal linking (see Table 2) and vertical linking (see Table 3) show that horizontal linking was more common (190 versus 150 total linkages) despite opportunities to perform such linking being less considering the number of exposed multiple visual representations. These results indicate the challenges associated with yo-yo reasoning, i.e. to reason and link between different organisation levels, in biology education. In the following sections we provide further detail about how students reasoned when performing the three different linking patterns.

Table 3. Vertical linking between multiple visual representations representing different organisational levels.

	One level vertical linking (number of links)	Two level vertical linking (number of links)
L1 to L2 (+1)	16	n/a
L2 to L3 (+1)	33	n/a
L2 to L1 (−1)	21	n/a
L3 to L2 (−1)	31	n/a
L1 to L3 (+2)	n/a	25
L3 to L1 (−2)	n/a	24
Total	101	49

Intra horizontal linking

Intra horizontal linking was most common at the sub micro level (L1), with fewer such linking at the other levels, see Table 2. The repeated pointing within the same visual representation at L1 was often accompanied by reasoning revealing that students often had difficulties in their meaning-making of the representations. For instance, difficulties were often related to discussions about the ‘on’ and ‘off’ switches (V5, Table 1) and the gene activation of the gene sequence on V6. In these linking patterns, the students first displayed a sequence of intra horizontal linking during their meaning-making attempts. This reasoning, together with pointing at the same visual representation, seems to support meaning-making over time, which is thereafter often manifested as intra horizontal linking. When doing so, it appears that the students use the meaning gained from one visual representation to make meaning of another visual representation at the same level, an observation also addressed in the next section.

Inter horizontal linking

In comparison with intra horizontal linking, inter horizontal linking was less common, see Table 2. For example, the pointing count at the sub micro level (L1) for inter horizontal linking was half as much compared to intra horizontal linking. Inter horizontal linking at the macro level (L3) was the most commonly revealed pattern, and was often linked to reasoning about environmental factors that affect the phenotype. Hence, a translation between visual representations V7 and V1, V2 and V8 was often involved in these accompanying student discussions.

Interestingly, inter horizontal linking at the sub micro level (L1) was almost as common as the macro level linking showing the usefulness of representations at the sub micro level (L1) for the students' meaning-making of epigenetics. These linkages were often associated with intense reasoning of the function of methyl groups on the DNA sequence (V6). A transition between V5 and V6 was often involved in these discussions. Meaning-making of the symbolic ‘on’ and ‘off’ switches of V5 was connected to the appearance of methyl groups in V6.

Inter horizontal linking at the micro level (L2) occurred less frequently. Reasoning about differences in young and old twin chromosomes occurs at these pointing’s (V2 and V4). The students were shown to reason about the variation in colour or shape between the two different chromosome pairs of the twins.

One level vertical linking

One level vertical linking patterns were observed in four different ways (Table 3). However, no specific differences could be identified between upward (+1), and downward (−1) linking. Linking from micro (L2) to macro (L3) upward (+1), was the most common observed linking. Students’ reasoning in these linkages often focused on features in the chromosomes (V2, V4) in relation to macro level visual features of the mice or humans (V1, V3 and V8). This reasoning was often accompanied by such linking, which was regularly observed at the commencement of meaning-making during the focus group sessions, which might indicate its importance. Likewise, the downward (−1) linking from macro (L3) to micro (L2) was the second most common one level vertical linking and was accompanied by the same reasoning. It seems like linking between macro and micro levels served as the easiest starting points for students’ meaning-making of epigenetic representations.

The linking between sub micro and micro levels, i.e. L1 to L2 (+1), and L2 to L1 (−1), occurred less frequently. It appeared challenging for students to connect these two abstract organisational levels, which confirmed the importance of linking these two organisational levels to the macro level. We also noted that students’ meaning-making of the sub micro level was usually related with horizontal linking.

Two level vertical linking

The two level linking patterns emerged in two different ways (+2) or (−2), see Table 3. These linkages were almost as common as one level vertical linking. This is perhaps surprising since the cognitive demand might be assumed higher for two level linking. However, the observation might be explained by connection to the macro level in both these instances. Also, the sub micro level provides explanatory potential since this level embeds the scientific mechanistic explanation, i.e. methylation, of epigenetics. Linking from L1 to L3 (+2) was aligned with reasoning about how the methylation mechanism (V6) at the sub micro level had an effect on freckles or bad health of the twins at the macro level (V3). Added to this, linking from L3 to L1 (−2) often revealed reasoning of how environmental factors (V7) might influence the activation or deactivation of protein synthesis (V6).

Summary of the observed nature of linking patters in relation to students’ reasoning

Overall, the analysis revealed four linking patterns that are associated with students’ reasoning, and are indicative of yo-yo reasoning. From the analysis, we can see that longer periods of intra horizontal linking seems to serve as a starting point for many meaning-making processes. These sequences seem to be a threshold for meaning-making, where the students first need to reason for a while during their meaning-making of the multiple visual representation These sequences are then often followed by regular inter level horizontal linking, one level and two level vertical linking as their reasoning progresses further. The results suggest that connecting macro and micro levels are the least demanding starting points for students’ meaning-making processes, but linking sub micro and micro levels appear most difficult. Linking between the levels was performed both upwards and downwards and there were no differences in this respect. Hence, students revealed little difference in commencing their reasoning by first linking upwards or downwards, respectively.

Characteristics of the multiple visual representations that influence students’ reasoning when linking across and between different levels of biological organisation

In the thematic analysis connected to the second research question, five themes were identified, which we denote as characteristics of the multiple visual representations that influence linking between the different levels of organisation. The students' pointing patterns and reasoning about the multiple visual representations show this linking. For each characteristic, we have constructed narratives demonstrating how the multiple visual representations influence students’ linking. The narratives are presented in Tables 4–8 where the image presentation of the multiple visual representations (first column) is related to the students’ interactions with the multiple visual representations, visually presented with student pointing. The students’ reasoning is further shown in the narrative turns (second column) in which we present verbatim excerpts of what the students uttered in the first person while pointing. Our analysis of the students’ meaning-making processes in relation to their linking patterns and reasoning is presented as an indication of linking and reasoning patterns (third column). Each line in Tables 4–8 shows the nature of the linking and reasoning accompanying a narrative turn. Each narrative is chronologically numbered and discussed in the accompanying text together with how that visual characteristic may relate to yo-yo reasoning.

Table 4. Narrative and comparative image presentation for the characteristic multiple visual representations.

Table 5. Narrative and comparative image presentation for the characteristic salient graphical visual features.

Table 6. Narrative and comparative image presentation for the characteristic analogous visual features.

Table 7. Narrative and comparative image presentation for the characteristic familiar visual elements.

Table 8. Narrative and comparative image presentation for the characteristic textual adjuncts.

Multiple representations influence horizontal and vertical comparisons

Students’ comparisons of multiple visual representations induced linking between different organisation levels. In the presented narrative (Table 4), the students of group 4 discuss the realistic visual representations presented at the macro level with the younger and older twins (V1 and V3). Phenotypic traits are compared with different epigenetic patterns in the symbolic visual representations of the chromosomes presented at the micro level (V2 and V4) (narrative turns 1 and 2 in Table 4). One example of the functionality of a clear difference in the visual representations is the expressed features such as spots (freckles) on the twins’ skin (V3) (narrative turn 1 in Table 4). This comparison with phenotype and different epigenetic patterns on the chromosomes leads to the search for a generalisation of this reasoning even in the young twins’ chromosomes (V2) (narrative turn 7 in Table 4). Discussion about similarities and differences between chromosomes and their epigenetic patterns compared with features of young and old twins as well as changes within the chromosomes over time then occurs (V2, V3 and V4) (narrative turns 2–6 in Table 4). Both horizontal and vertical linking are performed.

Salient graphical visual markings support meaning-making of epigenetic concepts between organisational levels

The analysis showed that students’ reasoning with salient features such as the green markings (V5) in the visual representations induced linking between different organisational levels. In the current example students of group 2 revealed the conception that the green colouring of a DNA strand signals something that is acting (V5) (narrative turn 3 in Table 5). However, the students struggle with associating the switches to activation or inactivation of the gene caused by the methyl groups (V5 and V6) (narrative turn 1–5 in Table 5). The switching off is connected to the methylated DNA (V6) (narrative turn 2 in Table 5). The question the students struggle with is whether the turning off causes terminated expression of the whole gene as expressed as the twins’ freckles (narrative turn in Table 5). The students reason at the sub micro level when relating the green colour when linking to the macro level and connect the colouring of the DNA to the characteristic of the mice at the macro level (V8) (narrative turn 6 in Table 5). In other student exchanges we also found that the yellow, red, brown, and green coloured markings of the chromosomes also induced similar linking between multiple visual representations.

Analogous visual features help students connect epigenetic concepts at different levels of biological organisation

The analysis showed that students’ analogical reasoning about the on/off switches triggered yo-yo reasoning. In this regard, the students decode the symbolic language of switches in the visual representations while relating these markings to other visual representations represented at the sub micro, micro and macro level, as exemplified by group 4 (V2, V5 and V8) (narrative turns 1–4 in Table 6). Such student discussions focused on the functions of the ‘on’ and ‘off’ switches in relation to gene activity at the sub micro level (V5) (narrative turn 5 in Table 6), in comparison to the visible differences in the chromosomes at the micro level (V2) (narrative turn 6 and 7 in Table 6).

Familiar visual elements help students link everyday life experiences at the macro level to visualisations representing epigenetic concepts at other organisation levels

The analysis showed that students’ reflections about everyday life experiences, such as exposure to environmental phenomena like smoking and diet, were triggered by reasoning about familiar visual elements such as the twins’ skin appearance. Hence, familiar visual elements depicted at the macro level helped students’ meaning-making of visual representations at other levels of organisation, and thereby facilitated vertical linking during students' reasoning. As shown by group 4 in Table 7, differences in genetically identical twins’ appearances is a familiar visual element (V7) (narrative turn 1 in Table 7), which is interpreted and explained by everyday life experiences such as smoking and diet, which then helps the students to perform vertical linking in their reasoning about the other visual representations (V1, V4 and V7) (narrative turns 1–5 in Table 7). Near the close of this exchange, the students were able to connect these results back to the sub micro level and the activation of genes (V5), and then to other environmental factors such as diet (V7) (narrative turns 5–6 in Table 7).

Textual representational adjuncts support students’ ability to reason with the multiple visual representations between organisational levels

The analysis showed that students used the textual adjuncts of the visual representations to support reasoning between different organisational levels. In the narrative provided in Table 8 for group 5 the students are persistent in trying to understand the visual representation with the activated and deactivated DNA sequence (V6) linked to the visual representation with an ‘on’ and ‘off’ switch (V5). When turning to the visual representation describing the methylation process, they encounter text adjuncts describing the function of methyl groups in gene expression (V6) (narrative turn 1 in Table 8). In meaning-making of these, the students acquire an understanding of the function of methyl groups at the sub micro level. In the following sequence we then see how the students make meaning about the gene expression process (narrative turns 1–6 in Table 8), which they subsequently relate to another visual representation representing the chromosome (V4) at the micro level (narrative turn 5 in Table 8). Finally, they return to the sub micro level and in their meaning-making of the methylation process by connecting to the ‘on’ and ‘off’’ switch (V5) (narrative turn 6 in Table 8). Hence, meaning-making of the textual adjuncts of the multiple visual representations seemed to induce vertical linking in the meaning-making process.

Discussion

In this section we focus on how the findings of this study might contribute to further insights about yo-yo related thinking, and the possibilities of using multiple visual representations to support students’ meaning-making with multiple visual representations in biology education.

The importance of different patterns of yo-yo thinking for students’ meaning-making

The findings indicate that the revealed yo-yo like reasoning seems to support students’ meaning-making of epigenetics; reasoning that might be transferable to other biological domains. Students discuss and make meaning of epigenetic phenomena by reasoning about the multiple visual representations representing different levels of organisation. This is an important contribution considering Knippels and Waarlo’s (2018) assertion that there is a lack of empirical studies that interrogate students’ yo-yo thinking. Our study empirically supports the importance of encouraging students to reason across and between different organisational levels when learning biology. With respect to evolution education, Jördens et al. (2016) found that laboratory activities purposely designed for students to explore the interplay between different levels of biological organisation provided students with better vertical explanations of evolutionary concepts. Our study had a similar approach using multiple representations of epigenetics to encourage reasoning across and between different organisational levels indicating the importance of students’ engagement and active reasoning in learning about biology phenomena connecting several organisational levels.

Students’ indication of multiple visual representations during their reasoning revealed four different linking patterns in our study: intra horizontal linking, inter horizontal linking, one level vertical linking, and two level vertical linking. When analysing the linking patterns in combination with the students’ reasoning, we can confirm the central importance of the macro level for students’ meaning-making of biological phenomena as often emphasised in the literature (see Duncan & Reiser, 2007; Knippels & Waarlo, 2018). Connections to the macro level were of most importance for students’ reasoning in this study, both salient in the one level and the two level linking patterns. Also, inter horizontal linking was important at the macro level for students’ meaning-making of epigenetic phenomena.

The sub micro level was reasoned about to a larger degree in students’ linking patterns than the macro level. Also, intra and inter horizontal linking at the sub micro level was very central in students’ meaning-making of the epigenetic content. The usefulness of, and dependence on, the sub micro level in students’ reasoning might be explained by the fact that much of the mechanistic explanations in genetics are present and explained at the molecular level (Haskel-Ittah & Yarden, 2017; van Mil et al., 2016; Verhoeff et al., 2008). Hence, a large component of the biological meaning is embedded at this level, making it central for students’ meaning acquisition, while at the same time also serving as the most challenging obstacle.

Interestingly, our findings in this regard are in contrast with some previous studies, in which students show various difficulties in meaning-making of multiple visual representations at the sub micro level (Knippels & Waarlo, 2018; Marbach-Ad & Stavy, 2000). However, our results indicate that by using multiple representations that communicate biological phenomena at different levels of organisation might help scaffold students’ reasoning and facilitate their overcoming of difficulties with biological phenomena at the sub micro level. In addition, the inter horizontal linking also supported students’ meaning-making of the multiple visual representations. This finding indicates that it is not only necessary to understand phenomena at different organisational levels (see Duncan & Reiser, 2007; Venville et al., 2005), it is also crucial to connect different perspectives and abstractions of visual representations in general. The multimodal approach adopted in this study seems to support both horizontal and vertical linking between the organisational levels.

A further salient finding of this study is that we could not differentiate any difference in students’ meaning-making in whether they linked upwards (+) or downwards (-) during their reasoning. Both linking patterns occurred at similar intensities and supported students’ reasoning and meaning-making of epigenetics in a similar fashion. The yo-yo teaching strategy often suggests that the approach should commence at the macro level, and then transcend stepwise downwardly towards the lower levels, and then revert upwards again for optimal learning (Knippels, 2002). In addition, downward linking has been described as a reason for being able to explain a phenomenon at a higher level with an object at a lower level, while upward linking is accompanied with functional reasoning where the function at a lower level explains the emergence of a phenomenon at a higher level (Boerwinkel et al., 2009). These two respective upward and downward reasoning patterns have been described as potentially problematic for students to navigate since causal explanations do not always translate across the levels of biological organisation (Gericke & Hagberg, 2007; Knippels & Waarlo, 2018). In this study, no such differences in students’ meaning-making that were dependent on linking patterns upward or downward could be identified. Rather, the vertical linking in itself between macro and sub micro level (transcending the micro level) was the pivotal factor regardless of whether performed upwardly or downwardly.

Visual characteristics that influence students’ yo-yo thinking

Five characteristics of the multiple visual representations that supported students’ linking between the different organisational levels while reasoning about epigenetics were identified in this study, namely multiple visual representations, salient visual features, analogous visual features, familiar visual elements and textual adjuncts. These characteristics might be applicable in informing the design of multiple visual representations for supporting students’ yo-yo reasoning and meaning-making. In this way, this study builds on the findings by Schönborn and Anderson (2009) by providing new insights about what aspects of the mode factor have the potential to trigger students’ reasoning across and between multiple biological organisational levels in biology and science education. Hence, multiple visual representations can be important tools for learning biology phenomena at different levels of representation. On this note, further studies are sought to test this premise in other biological and scientific domains.

The analogical switches depicted in one of the visual representations (V5) of this study as a representation of regulating gene activity could be compared to Johnstone's (1991) symbolic aspect where understanding of chemical concepts is understood by combining macro and symbolic (formulae, equations, and graphs) and sub micro aspects. In our study, we could see that such symbolic representation was important for students’ meaning-making both in horizontal and vertical linking. Hence, in future development of biology and science education endeavours that expand yo-yo reasoning around transcending organisational levels, the influence of symbolic representations would be important to consider.

Conclusions and implications

In conclusion, we revisit Knippels and Waarlo’s (2018) quest for the community to perform further empirical studies investigating how yo-yo reasoning might be related to teaching and learning.

This study aimed to investigate if and how multiple visual representations can induce students’ yo-yo reasoning in their meaning-making of epigenetics, and what aspects of the multiple visual representations might trigger this development. What insights have our findings contributed? Firstly, we have shown that the use of multiple visual representations representing different levels of biological organisation can be a powerful tool in supporting yo-yo teaching interventions.

Secondly, we have shown that in adopting teaching and learning settings that support students’ yo-yo reasoning, multiple visual representations should be deployed in a comparative way to encourage comparisons across and between levels. Such an approach is further supported if the visual representations at the macro level can be associated with students’ everyday experiences, and visual representations at the micro and sub micro level should be designed to include the need of meaning-making of salient visual features at other organisational levels, and emphasised by textual adjuncts that explain abstract features. Our results also suggest that it might be favourable to employ symbolic analogical features of the representations at micro and sub micro levels to explain scientific mechanistic processes with attributes from students’ everyday life.

While the current study potentially makes the above descriptive contributions to the area, it is limited in that it was performed in a single setting, and with a relatively small number of student participants. Therefore, it is uncertain to what extent these findings around students’ meaning-making of multiple epigenetic visual representations would be generalisable to other domains of science education. Our future work will adopt the findings here to set the scene for further fine-grained analyses of students’ meaning-making of different representation modes conveying epigenetics at different levels of biological organisation, which could include more advanced analysis around students’ mechanistic reasoning as proposed by van Mil et al. (2016).

Acknowledgments

Images in Figure 1 are obtained from: DNA, https://se.clipart.me/; Cell, https://pixabay.com/sv/illustrations/celler-m%c3%a4nsklig-medicinsk-biologi-1872666/; Mice, https://commons.wikimedia.org/wiki/File:Agouti_Mice.jpg; Images in Table 1 are obtained from: V1, https://commons.wikimedia.org/wiki/File:Twin_boys.JPG; V2 and V4, Reprinted with permission from: Fraga MF, Ballestar E et al. (2005), “Epigenetic differences arise during the lifetime of monozygotic twins”. Proc Natl Acad Sci U S A 102(30):10604-10609. https://www.pnas.org/doi/pdf/10.1073/pnas.0500398102, V3, https://commons.wikimedia.org/wiki/File:Twins_-_Gay_and_Gwyn,_Sun_exposure.jpg; V5, https://geneticliteracyproject.org/2019/10/08/easilymisunderstood4-things-to-know-about-epigenetics-including-the-fact-that-most-changes-are-not-passed-on-to-offspring/; V6, Reprinted with permission from Gericke, N.; V7, https://www.researchgate.net/profile/Riya_Kanherkar/publication/266390065/figure/fig5/AS:271795949404174@1441812614075/A-compilation-of-epigenetic-influences-on-humans-Thefigure-representsacompilation-of_W840.jpg; V8, https://commons.wikimedia.org/wiki/File:Agouti_Mice.jpg.

Ethical approval

The research conducted in this study followed the ethical guidelines as recommended by The Swedish Research Council (2017). The study design was reviewed and approved for implementation by the ethical advisor of the Faculty of Health, Science and Technology at Karlstad University, Sweden (Dnr. HNT 2020/383).

Disclosure statement

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

Additional information

Funding

This research was conducted within the Swedish National Graduate School in Science and Technology Education Research (FontD) (funded by The Swedish Research Council [grant 2017-06038]), The Centre of Science, Mathematics, Engineering, Education Research at Karlstad University and Karlstads kommun, Sweden.

Notes

1 https://www.youtube.com/watch?v=gG7uCskUOrA

2 https://www.pbslearningmedia.org/resource/biot09.sci.life.gen.epigenetics/epigenetics/