Effectiveness of Visual Representations in Undergraduate Human Anatomy and Physiology I & II

University of Mississippi University of Mississippi

eGrove eGrove

Honors Theses

Honors College (Sally McDonnell Barksdale

Honors College)

Spring 5-1-2021

Effectiveness of Visual Representations in Undergraduate Human Effectiveness of Visual Representations in Undergraduate Human

Anatomy and Physiology I & II Anatomy and Physiology I & II

Mary Agnes Mestayer

Follow this and additional works at: https://egrove.olemiss.edu/hon_thesis

Part of the Anatomy Commons, and the Biology Commons

Recommended Citation Recommended Citation

Mestayer, Mary Agnes, "Effectiveness of Visual Representations in Undergraduate Human Anatomy and

Physiology I & II" (2021).

Honors Theses

. 1793.

https://egrove.olemiss.edu/hon_thesis/1793

This Undergraduate Thesis is brought to you for free and open access by the Honors College (Sally McDonnell

Barksdale Honors College) at eGrove. It has been accepted for inclusion in Honors Theses by an authorized

administrator of eGrove. For more information, please contact [email protected].

EFFECTIVENESS OF VISUAL REPRESENTATIONS IN UNDERGRADUATE

HUMAN ANATOMY AND PHYSIOLOGY I & II

Mary Agnes Mestayer

A thesis submitted to the faculty of The University of Mississippi in partial fulfillment of

the requirements of the Sally McDonnell Barksdale Honors College.

Oxford, MS

May 2021

Approved By

______________________________

Advisor: Dr. Carol Britson

______________________________

Reader: Dr. Carla Carr

______________________________

Reader: Dr. Brian Doctor

Mary Agnes Mestayer

iii

ACKNOWLEDGEMENTS

Thank you to my friends and family for their support and grace during this process.

Thank you to my advisor, Dr. Britson, and two readers, Dr. Doctor and Dr. Carr. This

would not have been possible without your help.

ABSTRACT

MARY AGNES MESTAYER: Effectiveness of Visual Representations in Undergraduate

Human Anatomy and Physiology I & II (Under the direction of Dr. Carol Britson)

The objective of this project was to explore intersections between student

preferences and student performance on anatomical visual representations in Human

Anatomy and Physiology I and II. Visual representations are a critical resource for the

formation of relationships between function and structure furthermore; students interpret

these representations uniquely based on specific factors (learning objective, prior

knowledge, the diagram studied, etc.). Phase I of this project gathered undergraduate

responses to ten Likert-style questions on their opinions on diagrams and their use in the

A&P classroom. Phase II of this project presented participants with twelve manipulated

diagrams sourced from three diagram with four manipulations (a control in which no

manipulations were applied, a change in the portion of the leader lines located on the

diagram, both the part of the leader line on the diagram and the label location of the

leader line, and a 25% decrease in the number of leader lines) applied to each of the three

diagrams. Participants were asked to correctly identify anatomical structures on the

twelve diagrams and rate the confidence in the correctness of their answers. Students’

responses in Phase I indicated that when viewing two-dimensional diagrams, students

preferred simplified diagrams with leader lines labeling every anatomical structure

represented. In Phase II, this preference translates into a higher proportion of correctly

identified structures associated with the diagrams contain supporting visual details and the

use of visual cutes to separate structures. Generally, however, students performed better

and felt more confident on the sarcomere diagram, which had the least amount of total

leader lines. The order of diagram presentation sequence had a significant effect on

student performance (F

(2,335)

= 15.61, p= 0.00) with students preforming significantly

better on the sequence featuring diagrams randomly grouped together. The conclusions

made from this project support current research, which suggest that while students prefer

three-dimensional diagrams (Tan et al 2012), students preform best on detailed, two-

dimensional diagrams (Fenesi et al 2017). The practical applications of this project have

the potential to better inform educators as they choose diagrams to integrate into the

classroom.

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

MATERIALS AND METHODS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

LITERATURE CITED. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

TABLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Appendix A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Appendix B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Appendix C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

vii

LIST OF TABLES

TABLE 1

Human Anatomy and Physiology I student responses to Likert-

style statements on phase I survey which recorded undergraduate

opinions on diagrams and their function in the A&P curriculum

viii

LIST OF FIGURES

FIGURE 1

Human Anatomy and Physiology II student performance in Phase II

which, consisted of correct identification of a single anatomical

structure on each of the twelve manipulated diagrams (Amerman 2018;

Appendix C).

FIGURE 2

Human Anatomy and Physiology II student performance in Phase II

which, consisted of participant self-assessment of their confidence in the

correctness of their identification of an anatomical structure from one of

the twelve manipulated diagrams (Amerman 2018; Appendix C).

INTRODUCTION

Human Anatomy and Physiology (A&P) is a two-semester course that focuses on

the structure and function of tissues, organs and systems of the human body. The main

learning objective of the course is to establish relationships between the identification of

specific anatomical structures and the biological process in which the specific structure

functions. In students, the association of structure and function is created by a

combination of textual and visual information to create a holistic mental model (Lin et al

2017). The current understanding on the role of visual representations in the A&P

curriculum suggests that diagrams are crucial for successful topic storage and retention as

they allow students to form a mental model in conjunction with knowledge from other

storage. Mental model formation is what allows students to create a deeper and more

critical understanding of anatomical topics.

Instructors test students’ comprehension of identification and formation of a

correct mental model through a series of practical laboratory examinations and lecture

examinations. These assessments focus on different styles of learning; practical

laboratory examinations focus on identification while lecture examinations focus on

information synthesis (Britson 2020). Identification of anatomical structures in laboratory

practical examinations prepares students for further study by ensuring a concrete

understanding of anatomical structure location and the spatial relationships between

structures (Britson 2020). Both types of examinations often utilize anatomic diagrams as

part of their question sequence. To excel on these exams, students must be able to

comprehend information from multiple diagrams, that vary in complexity and focus, to

gain a level of knowledge that allows synthesis of information from multiple sources,

including anatomical visual representations (Fenesi et al 2017). The ability of learners to

comprehend information from multiple sources and compile that information into a

concise mental model indicates a high level of comprehension (Fenesi et al 2017).

Diagrams are an important resource that helps students of all levels visualize

relationships between anatomical structures that they have previously unexplored.

Butcher (2006) and Mayer and Sims (1994) concluded that students learn better from text

that is accompanied by diagrams than just text. Diagrams provide instructors and students

a multipurpose approach to topics, which allows for students from both high-prior

knowledge and low-prior knowledge backgrounds to succeed in the classroom (Khalil et

al 2005). Visual representations are used as learning tools to encourage students to

visualize concepts and understand structures as they relate to each other (West et al

1991).

Instructors aim to facilitate students’ formation of a mental model, which

facilitates a proficiency in a topic. A mental model is the association between recently

acquired information and established information to build a visualization that allows

information from multiple sources to be related (Seel and Strittmatter 1989). This mental

model can be formed from a variety of sources including diagrams presented during

instruction (Khalil et al 2005) and is an essential part of developing long-term memory

for the topic (Fenesi et al 2017). If the mental model is created correctly, students use

their mental model to make interpretations and inferences about information adjacent to

their model (Khalil et al 2005). Mental models have the unique ability to allow students

to create new relationships between previously unrelated structures. This fluid, applicable

nature of mental models created by students has made visual representations a critical

aspect of the A&P curriculum.

According to Kottmeyer et al (2020), diagrams have become valuable in biology

education because “they are holistic, spatial arrangements; show both temporal and size

scales; and depict structure-function relationships”. Diagrams are tasked with

communicating biological concepts while referencing spatial relationships which has led

to a substantial variety of anatomical diagrams available. Butcher (2006) found that

learners synthesize information from anatomical visual representations uniquely in

response to the representation. This unique response means that certain representations

are more effective than other representation in expressing ideas (Butcher 2006). One

student may find that a diagram that represents a comprehensive overview of an

anatomical system the best way to understand a system while another student would

prefer simplistic diagrams that target specific aspects of the same system. Students’

comprehension of a specific diagram, Diagram Comprehension Ability, is influenced by

specific factors such as learning objective and the diagram studied (Kottmeyer et al

2020). Diagrams range from print to digital models and from detailed to broader

representations. The type of diagrams used has a unique influence on students’

understanding of anatomical topics and their construction of an accurate mental model

(Kottmeyer et al 2020).

Representations that are less detail oriented and focus on a broader representation,

typically diagrams with simplified lines and grey-scaled coloring, can seem more abstract

to learners (Lin et al 2017). More abstract representations can be beneficial to learners

because the increased effort in understanding the diagram initially leads the learner to

encode a more defined and robust representation of the information depicted in the

representation (Fenesi et al 2017). Detailed representations are associated with a more

aesthetic appearance (e.g., use more color and a greater emphasis of realistic details) -

which encourages students to study the representation. The emphasis on realistic detail

also encourages students to connect abstract ideas to more concrete examples (Lin et al

2017).

However, detailed representations are not always associated with a positive

learning effect on the student by requiring a greater cognitive load to process perceptual

details, which can reduce learning (Lin et al 2017; Fenesi et al 2017). Students have a

maximum cognitive load that they can process (Lin et al 2017) and excessive colors and

detail can consume a high cognitive load with low academic return (Fenesi et al 2017).

Detailed representations force students to distinguish between relevant and irrelevant

information which, for students with low prior knowledge of anatomical topics, can be

overwhelming and impede integration of anatomical function and structural

identification. (Lin et al 2017).

As technology has become a valued asset in the anatomy classroom, visual

representations have transitioned from representations found on paper in textbooks to

digital representations that can be manipulated. The progression from print to digital was

initiated by advancement in technology and a generation of students with unique styles of

learning (Khalil et al 2005). Digital representations allow for self-paced learning by

students on a variety of academic levels (Khalil et al 2005) as well as the opportunity for

animated representations (Kriz 2007). The animation of anatomical representation allows

for information to be presented in a manner most reflective of human anatomy (Kriz

2007). Digital representations allow for three-dimensional diagrams to be created and

manipulated by students, which can be a disadvantage for students with low special

rotational abilities (Fenesi et al 2017). The transition to digital representations also

allows for students to study representations from sources outside of their course or

textbook.

Students and educators have access to an immense library of anatomical

representations to utilize in their learning process. The variance in representations

available to students and educators leads to the formation of preferences which are rooted

in a variety of factors including IQ, gender, age, spatial reasoning, and verbal ability

(Ainsworth 1999) and are established in response to the ease of interpretation by students.

Preferences are formed based on ease of representation interpretation, which supports the

preference of representations that are easy to comprehend and work in parallel with

student understanding of the material. A student's ability to interpret a visual

representation is influenced by a student's knowledge of a subject matter and their ability

to interpret visual information (Offerdahl 2017). Student interpretation of visual

representation can be categorized by two processes- top-down interpretation and bottom-

up interpretation. Top-down interpretations are driven by a student’s prior knowledge and

learner expectations while understanding of perceptual details propels bottom-up

interpretations of visual representations (Kottmeyer 2020). These two systems of

interpretation work together allowing students to comprehend complex information from

visual representation. Students begin their interpretation by using a bottom-up

interpretation where students categorize and compartmentalize information found in the

diagram. This information is organized and utilized to construct a mental model of the

information begin received. A top-down interpretation is provided to the newly

constructed mental model (Kottmeyer 2020) by applying prior knowledge and experience

to contextualize their understating of the representation, a bottom-down interpretation.

(Kriz 2007).

Diagrams have always been central in their importance in A&P curriculum and in

response; efforts have been made to understand how students comprehend diagrams.

Students interpret diagrams differently and diagrams are not equal in effectiveness for all

students (Butcher 2006). To incorporate multiple representations into the curriculum and

compensate for the unique interpretations of visual representations by all students,

educators began to implement multimedia learning in the classroom. By providing

students with multiple representation of one topic, multimedia learning encourages

students to gain a deeper understanding of a particular topic and synthesize multiple

representations to create an accurate mental model of the topic from multiple sources

(Ainsworth 1999). However, students often pick a singular visual representation to focus

on which leads to the formation of mental models that are not accurate or do not include

all aspects of the topic (Ainsworth 1999). If multimedia are correctly applied, students

will be provided representations that “complement, constrain, and construct” meaning

these representations aim to reinforce complimentary information, constrain

misinterpretations, and encourage students to construct a thorough understanding

(Ainsworth 1999). Educators face the challenge of providing students with multiple

representations without including irrelevant information to topics (Lin et al 2017).

Despite support for diagram use in the classroom, research on which variations of

diagrams are the most effective in the classroom is lacking. I have the following

questions: (1) What are students' perceptions on the importance of diagrams in A&P?; (2)

If students understand the importance of visual representations, are they willing to find

representations outside of the classroom?; and (3) When students are preparing for an

anatomical structure identification assessment, do they prioritize learning the mechanics

of the diagram or the anatomical structures? To test these questions, I developed the

following hypothesis; students' responses indicate a preference for a wide variety of

diagrams based on personal preference but most students will prefer simplified diagrams.

I also hypothesize that students presented with diagrams containing manipulated leader

lines will have greater difficulty correctly identifying anatomical structures.

METHODS

Participants for this study were recruited from students enrolled in Human

Anatomy and Physiology I (BISC 206) and II (BISC 207) at the University of Mississippi

in the Fall 2020 and Spring 2021 semesters. Human Anatomy and Physiology is a two-

course sequence in which students must pass Anatomy and Physiology I (with a C or

higher) to enroll in Anatomy and Physiology II. All participants were undergraduate

college students with most ranging between the ages of 18 and 24 and varying in race and

gender. The protocol was approved as Exempt under 45 CFR 46.101(b)(#2) by the

University of Mississippi Institution Review Board (Protocol #21x-082).

This study was divided into two parts. In both parts students were recruited to

participate via an announcement from their Human Anatomy and Physiology I and II

course instructor. Students were told that successful participation in the project would be

compensated with extra credit towards their Anatomy and Physiology I and II grade. In

the initial announcement, students were made aware of an alternative activity of equal

extra credit value in lieu of participation in this project.

Phase I

This phase of the project was offered to students in Anatomy and Physiology I in

the fall semester of 2020. Students interested in participation emailed the PI to gain more

information about the project. After understanding what the project entailed, students

made informed decisions on participation. Once students agreed to participate, they

received an email with specific instructions that included a link to the survey (Appendix

A), project details from the IRB information sheet, and a summarized list of participation

requirements. The survey consisted of ten Likert-style questions with responses ranging

from strongly agree to strongly disagree in response to each statement. Questions focused

on students’ perception and understating of diagrams and their functions in Human

Anatomy and Physiology I. One hundred twenty-four students successfully completed the

survey.

Phase II

This phase of the project was offered to students in Anatomy and Physiology II in

the spring semester of 2021. Students received an email from the PI explaining the

second phase of the project and the extra credit associated with successful phase two

participation. Students interested in participation replied to the PI. Then, the PI emailed

the potential participant the IRB information sheet, a link to a consent form that certifies

participant age, video conference call availabilities, and a summarized list of participation

requirements. Participants and the PI worked to schedule an online video conference call

that both parties attended. An hour before the scheduled call, participants received a link

to the video conference call and a link to three diagrams (Appendix B) with instructions

to familiarize themselves with the anatomical structures labeled in the diagrams in

preparation to being asked to identify structures from an unmarked diagram. The three

diagrams were sourced from the A&P textbook (Amerman 2018) which means that

students had already studied from the three selected diagrams. On the scheduled video

conference call, the PI would begin by certifying that participants had taken the survey

certifying their age and had become familiar with the three diagrams. The PI would then

give verbal instructions to participants regarding participants: “You will be shown 12

diagrams that are similar to the three that you looked at earlier. You will be asked to

identify the anatomical structure indicated by the leader line labeled with a question

mark. After verbal identification, you will be asked to rate your confidence in your

answer on a scale from 1 to 10; 10 being the most confident that you are correct and 1

being the least confidence that you are correct.” After instructions were given,

participants were given the opportunity to ask any questions before the project began.

The PI would utilize the screen-sharing capacity of the video conference call system and

project a PowerPoint slideshow containing the 12 diagrams (Appendix C). The three

original diagrams [an inferior view of the skull (bone diagram), a mid-sagittal view of the

brain (brain diagram), and a sarcomere (sarcomere diagram)] had four types of

manipulations: a control in which no manipulations were used on the original diagrams,

the portion of the leader lines located on the diagram changing, both the part of the leader

line on the diagram and the label location of the leader line, and a 25% decrease in the

number of leader lines. The 12 diagrams were presented to participants in three different

sequences: version A, version B, and version C. Version A of the diagrams featured the

diagrams grouped by their original diagram meaning that all of the sarcomere diagrams

were grouped together, all of the brain diagrams were grouped together, and all of the

bone diagrams were grouped together. Version B randomly grouped the diagrams

together. Version C grouped diagrams by treatment which functionally translated to all

four of control diagrams grouped together, all four of the starting point of the leader line

altered together, all four of the starting and ending point of the leader line together, and

the diagrams with 25% fewer leader lines grouped together. Each participant was

randomly assigned a diagram sequence. The participant would have 90-seconds on each

diagram and would be asked to identify the anatomical structure indicated by the question

mark. If the participant was unsure or answered that they did not know the structure, they

were encouraged to guess. After the participant had completed all completed all 12

diagrams, participants were thanked for their participation and the video conference call

was terminated.

Phase I data were compiled to obtain descriptive statistics about survey responses.

Phase II data was analyzed with a 3-factor analysis of variance (ANOVA) with the level

of significant set at α=0.05 where diagram, treatment, and version were the experimental

factors. Cohen’s d statistic was used to calculate effect size for any significant

differences.

RESULTS

Phase 1 Results

Fifty-seven percent of the 126 students in Anatomy and Physiology I (spring

2020) who participated in the survey had previously completed a biology course at the

University of Mississippi. The number of biology credits ranged from just under 4 credit

hours at the 100-level for 53 participants to 4 credit hours at the 400-level for 8

participants. Results from the remaining survey questions were divided into three general

categories: (1) Spatial Awareness, (2) The Use of Digital Resources, and (3) The Effect

of Diagrams Structure on Learning.

Survey questions focused on spatial awareness provide insight into student

opinions on the importance of spatial awareness and its function in the anatomy

curriculum. Ninety percent of students agreed or strongly agreed that an important part of

preparing for an A&P I laboratory practical was to understand the spatial awareness

between structures in an organism (Table 1). A little over 40% of respondents strongly

agreed that they prefer to lean on three- dimensional structures rather than two-

dimensional structures. As seen in Table 1, three-fourths of students agree and strongly

agree that when they had access to three-dimensional anatomical software, they would

prefer to use the three- dimensional structures to prepare for the laboratory practical.

When studying two-dimensional structures, about fifty percent of students agreed that

they can apply their knowledge to a more complex diagram and more specifically,

knowledge acquired on two-dimensional figures to three-dimensional figures easily

(Table 1).

As seen in Table 1, fifty percent of students agreed or strongly agreed that when

preparing for a laboratory examination, students seek computer-based models

independently of provided diagrams to better contextualize their knowledge. Half of

students also agreed that they feel comfortable applying their knowledge from a

computer-based diagram to an image or diagram of a physical dissected specimen.

Students varied in opinion on feeling comfortable learning to identify anatomical

structures using computer models rather than diagrams or pictures of physical dissected

specimen (Table 1).

The results recorded in Table 1 indicates that 80% of students agreed or strongly

agreed that they preferred simplified diagrams that highlight one structure rather than a

detail-oriented diagram when initially learning about a structure. Students preferred

simplified structures however, 60% preferred structures that had leader lines that labeled

every structure instead of only relevant structures being labeled (Table 1).

Phase II Results

There were no significant differences in the interaction effect between the

treatment applied to the diagram and student performance (F

(3, 335)

= 1.31, p = 0.273).

There was also no significant effect on diagrams times version (F

(4, 335)

= 0.928, p =

0.448) or treatment times version (F

(6, 335)

= 0.115, p = 0.995) on student performance.

There was a significant effect of version on student performance (F

(2, 335)

= 15.61, p =

0.00). The magnitude of the effect size that sequence version has on student performance

can be quantified. There is a medium magnitude effect size between the B and A versions

(0.39) and the A and C versions (0.26). There is a large magnitude effect size between the

B and C versions (0.65) of the diagram sequence. There was a significant effect of

diagram times treatment on student performance (F

(6, 335)

= 9.53, p = 0.00).

The three diagrams without any manipulations to the original, the control, had no

significant differences in student performance between the bone and brain diagrams

(Figure 1). However, a two-fold increase in the proportion of correctly identified

structures was seen in the sarcomere diagram. The three diagrams with the location of the

leader line located on the diagram changed, denoted by “starting point” in Figure 1,

corresponds to a ten percent decrease in correctly identified structures in both the

sarcomere and brain diagrams. In the bone diagram, the starting point manipulation has a

positive effect on student performance and is associated with a twenty percent increase in

correctly identified structures (Figure 1). The three diagrams with both the leader line

located on the diagram and the labeled portion of the leader moved, noted as “both

points” in Figure 1, had no effect on student performance on the sarcomere diagram. On

the bone and brain diagrams, both leader line manipulations lead to an increase in the

proportion of correctly identified structures. On the three diagrams with a 25% decrease

in the number of leader lines, denoted as “fewer lines” in Figure 1, a substantial change in

the proportion of correctly identified structures for all three diagrams is noted. The brain

diagram with fewer leader lines had a forty percent increase on student performance. The

bone and sarcomere diagram have a decrease in student performance by about fifty

percent.

There were no significant differences in the interaction effect between the

treatment applied to the diagram and participant confidence (F

(3, 335)

= 1.5, p = 0.215).

There was also no significant effect on diagrams times version (F

(4, 335)

= 0.75, p = 0.558)

or treatment times version (F

(6, 335)

= 1.19, p = 0.313) on participant confidence. There

was a significant effect of version on student confidence (F

(2, 335)

= 9.82, p = 0.00). The

magnitude of the effect size that sequence version has on participant confidence can be

quantified. There is a medium magnitude effect size between the B and A versions (0.39)

and a small effect size between the A and C versions (0.093). There is a large magnitude

effect size between the B and C versions (0.52) of the diagram sequence. There was a

significant effect of diagram times treatment on participant confidence (F

(6, 335)

= 6.25, p

= 0.00).

The three diagrams with no changes to their leader lines had no significant effect

on participant confidence. Participants were most confident in their identification of

structures on the sarcomere diagram, rating their confidence at a 7.4, and least confident

in their identification of structures on the bone diagram, rating their confidence at a 5.6

(Figure 2). The three diagrams with the location of the leader line on the diagram altered,

the “starting point” column on Figure 2, had significant effects on participant confidence

for the bone and sarcomere diagrams. The change in leader line staring point had no

significant effect on the brain diagram, however an increase in participant confidence on

the sarcomere diagram (Figure 2) is noted. The change in starting point correlates to a

16% decrease in student confidence on the bone diagram. The three diagrams with both

the leader line located on the diagram and the labeled portion of the leader moved, noted

as “both points” in Figure 2, had a positive effect on all three diagrams. The magnitude of

this effect varies, the confidence in the sarcomere diagram increases by three percent

while the confidence in the bone and brain diagrams increases by fourteen and eight

percent, respectively. The three diagrams with a 25% decrease in the number of leader

lines, indicated by “fewer lines” in Figure 2, have no effect on participant confidence on

the bone diagram. However, a significant change in the confidence for both the brain and

sarcomere diagram is noted (Figure 2). Participant confidence on the brain diagram is

increased by 25% compared to the confidence on the control. Confidence on the

sarcomere diagram is decreased by a fourth (Figure 2).

DISCUSSION

Phase I

The survey associated with phase one of this project was used to compare

University of Mississippi undergraduate human anatomy students’ opinions and

performance to published research on undergraduate learning.

Students overwhelming agreed that understanding the spatial relationship between

structures is important for mental model formation, which supports Khalil et al (2005)

who states that the goal of diagram incorporation is to facilitate mental model formation,

which allows students to make informed inferences about related structures. Students that

integrate visualization and analysis of three-dimensional models develop a linkage

between previously acquired knowledge relevant to the structure and knowledge

presented in the provided diagram (Wu et al 2000). Another intersection between student

survey response and published data is the preference to learn using three-dimensional

anatomical representation rather than a two-dimensional representation. In agreement

with my data, Tan et al (2012) conclude that students prefer to learn with three-

dimensional diagrams when available, but these diagrams do not yield a statistically

significant advantage on examination performance. Despite the lack of correlation

between the use of three-dimensional diagrams and improved exam scores, Tan et al

(2012) suggest that three-dimensional diagrams can improve learning by increasing

student interaction with the visual representation, which is associated with more

consistent and holistic mental model formation.

Use of three-dimensional visual representations in the classroom requires a virtual

component for integration into the curriculum. Technology has become more integrated

into the classroom and student responses indicate that use of digital resources alongside

classroom provided resources has become standard. This finding supports published data

that encourage integration of digital resources as it allows for structures and concepts to

be presented in a more explicit way (Kriz 2007). Using digital resources also allows

students to self-pace and control their own curriculum which can appeal to the varied

knowledge levels of each student (Khalil et al 2005). Digital representations offer

positives to both the learner and instructor; however, students were unable to completely

agree that they were comfortable beginning their mental model formation with digital

diagrams. Digital diagrams traditionally contain more color and details, which can

overwhelm students with less familiarity with structures (Lin et al 2017). Previous

research supports a blended approach that uses digital resources as a complement to

physical resources provided which encourages long-term knowledge formation (Khalil et

al 2015).

Butcher (2006) agrees that visual representations are not equal in their

effectiveness for each learner. This variation in effectiveness comes from both diagram

specific factors (e.g., number of labels, style, color, etc. ) - and the student’s background

(e.g., spatial awareness ability, knowledge of the subject, ability to comprehend visual

information, etc.; Offerdahl 2017). Students’ responses indicate that they preferred

simplified structures with every structure identified. When applying this preference to the

cognitive load theory proposed by Lin et all (2017) there are some interesting

intersections.

Cognitive load theory states that each student has a maximum cognitive load they

can comprehend before information is no longer stored effectively. It is recommended

that educators take this theory into account when choosing diagrams to incorporate into

the curriculum. Educators should aim to not waste cognitive load on excessive or

extravagant details that consume cognitive load without adding to a student’s knowledge

(Lin et al 2017). This theory is supported by students’ preference of a simplified diagram,

however, students preferring every structure being labeled rather than only relevant

structures is a contrast from what the theory predicts. Excessive visual details add no

educational value to diagrams and can distract learners while, comparatively, labeling

every structure can help students form connections and associations between previously

attained information. However, having to distinguish between relevant and irrelevant

structures when focusing on concepts could have a positive effect on mental model

formation as Fenesi et al (2017) found association between the amounts of effort required

during initial understanding of the diagram and the representation associated in the

students' mind; the more effort required initially to understand the diagram encourages a

more in-depth and critical understanding of the information.

Responses gathered helped to inform my understanding of undergraduate

students’ preferences on types of diagrams. Before the collection of these responses,

based on personal preference, I assumed that students would prefer printed simplified,

two-dimensional diagrams and that this preference would translate to an increased

proportion of correctly identified structures on simplified diagrams.

Participant responses support published data showing that diagrams are a crucial

resource for mastering an anatomical topic (Fenesi et al 2017). Students agreed that when

preparing for a laboratory practical, using digital visual resources was crucial to fostering

their understanding of a topic. This result is extremely relevant as students and educators

are relying more heavily on digital components of the curriculum. Physical diagrams are

still important during initial mental model formation; however, students recognize the

importance of digital resources from multiple sources, which helps to create a whole,

well-informed mental model (Ainsworth 1999). Digital visual representations create an

opportunity for the integration of three-dimensional representations into the curriculum.

Based on student responses, these three-dimensional representations are important for

representing the spatial relationship of multiple anatomical structures. This conclusion is

currently relevant as traditional, in-person laboratories are not always a viable option for

every student or institution. During the COVID-19 pandemic, traditional laboratory

instruction has transition to digital instruction, which has forced instructors to adapt to

teaching in an online setting. For A&P courses, this transition to online learning means

finding a substitute to teaching with specimen and models. Three-dimensional, digital

representations are not an equal substitute for a physical laboratory; however, students do

comprehend spatial relationships using these resources (Khalil et al 2005). This spatial

relationship formed from physical laboratories or three-dimensional diagrams helps

students to relate anatomical structures to one another, which encourages mental model

formation.

Phase II

The three diagrams used in phase two were unique in their style, number of

leader lines, and colors. Allowing for analysis of a variety of diagram dependent factors

of student performance.

Participants had an hour between receiving the diagrams and being asked to

identity structures from the diagrams. An intuitive assumption is that this short time

frame would increase the likelihood that participants would memorize leader line

positions rather than use the diagrams to form a mental model. If this assumption was

correct, I would expect to see a higher level of correctness associated with the structures

on the control diagrams compared to structures on any manipulated diagrams. However,

the proportion of correctly identified structures on both the bone and brain control

diagrams were significantly lower than for the manipulated diagrams. For the sarcomere

diagram, however, the proportion of correctness was highest for the structure with the

control treatment. This difference may be attributed to the style of the sarcomere

diagram. The sarcomere diagram is the simplest of the three diagrams; the original

diagram had fewer lines and used color to differentiate between structures. Presumably

these factors contribute to the sarcomere diagram being the easiest for students to recall.

The sarcomere diagram had the highest proportion of correctly identified structures for all

treatments except the treatment with fewer lines. The diagram-dependent factors present

in the sarcomere diagram exclude the sarcomere diagram from conclusions drawn from

the bone and brain diagrams because these factors create too many points of variations to

treat the three diagrams as equal. Participants were given the same three diagrams to

review for the same amount of time which excludes subject dependent causes of variation

such as time spent reviewing the diagrams and the type of diagram reviewed by

participant. The differences between these three diagrams can be quantified by looking at

the proportion of correctly identified anatomical structures by participants. The bone and

brain control diagrams have similar proportion of correctly identified structures at 44.6%

and 55.4% respectively while structures on the sarcomere diagram were identified

correctly 76.9% of the time.

The manipulated bone and brain diagrams had a higher proportion of correctness

for structures on the manipulated diagrams. This increased proportion leads to several

possible conclusions. One possible conclusion is the manipulated diagrams presented the

anatomical structures in a manner that were not significantly different enough to elicit a

change in the proportion of correctly identified structures. Another conclusion implies

that participants initiated some level of mental model formation rather than memorization

of leader line placement. A final conclusion is that the bone and brain diagrams with a

reduced number of leader lines were more simplistic and therefore easier for students to

interpret, which correlates to an increased proportion of correctness. The conclusion that

the reduced number of leader lines lead to a more simplistic diagram can be dismissed

because the results indicate an increased portion of correct answers for all manipulated

diagrams, not only the diagrams with a reduced number of leader lines. The conclusion

attributing the increased proportion of correctly identify structures to students being

unaware of any manipulations to the diagram and preforming slightly above average can

be disproven by the differences between the proportion of correctly identified structures

for the control diagrams compared to the manipulated diagrams. My conclusion

attributing the increased proportion of correct answers to increased mental model

formation is not easily dismissed by the data from phase II which encourages further

exploration of these ideas. Attributing the increased proportion of correct answers to

increased mental model formation; I identify factors responsible for increased mental

model formation. The bone and brain diagrams are more complex than the sarcomere

diagram and the increased complexity requires participants to expend more cognitive

resources to comprehend anatomical structures, which may encourage a more in-depth

understating of the structure (Fenesi et al 2017). This understanding is better equipped to

be projected onto manipulations of the previously studied diagram to correctly identify

the anatomical structure.

Another factor that may increase correct identification in the manipulated bone

and brain diagram compared to the sarcomere diagram is the style of the diagrams. The

brain diagram communicates the most anatomically-realistic features compared to the

other two diagrams. The brain diagram does not use color to distinguish between

structures and requires an attention to detail to differentiate between structures. The brain

diagram’s lack of differentiation between structures by visual aide requires some amount

of prior knowledge application to identify structures found in the diagram. Required prior

knowledge can be disadvantageous and overwhelming for students with low prior

knowledge (Lin et al 2017). The bone diagram has other unique features compared to the

brain and sarcomere diagram that further differentiate the three diagrams. The bone

diagram uses vibrant colors to denote various bones in the skull. The use of color as a

means of distinction between structures can be distracting and make it more difficult for

students to integrate the anatomical structures represented in this diagram into their prior

knowledge (Lin et al 2017). In contrast, Khalil et al (2005) found that students preferred

diagrams that used color as a tool to differentiate between anatomical structures. Color

adds another aspect to the cognitive load of the student to comprehend which, by the

cognitive load theory, would decrease student’s ability to recall and correctly identify

structures (Lin et al 2017). My results contradict this theory as a higher proportion of

correctly identified structures in both the brain and bone diagram is seen compared to the

less cognitively taxing sarcomere diagram.

In a more macroscopic view of my analysis, all participants were shown the same

diagrams with the same manipulations; the only difference was the presentation. three

variations of presentations were used: (A) which featured the diagrams grouped by their

original diagram, (B) which featured the diagrams randomly grouped, and (C) which

featured the diagrams grouped by their treatment. I hypothesized that version (A),

grouping diagrams based on their original diagram, would have a higher proportion of

correctly identified structures. Organizing the diagrams together in this manner would

theoretically allow students to maintain focus on one mental model at a time instead of

switching between diagrams. An ANOVA analysis was performed on the phase two data

and found that version (B) correlated to the highest proportion of correct answers. The

diagrams in version (B) were randomly arranged with no patters in diagram type or

diagram treatment. This indicates that the sequence of diagram presentation does have a

significant effect on student performance; however, the root of this effect is unidentified.

Blocking experiential design (e.g., version order of questions) was used to attempt to

minimize the effect of subject dependent variables on results. However, the small sample

size of this project increases the effect of subject dependent variables on results. Future

projects should consider the effect of question order during experimental design.

Participant confidence offers key insights to explore identification of anatomical

structures through using the Dunning-Kruger effect. The Dunning-Kruger effect finds an

inverse relationship between confidence in a correct answer and the level of knowledge

the individual has on a particular subject (Dunning and Kruger 2009). Applying this

relationship to my project, I would expect to see an inverse relationship between the

proportions of correct answers to the confidence in a correct answer. I see a modest

positive correlation between these two variables; however, students generally

overestimated the correctness of their responses. Structures that were easy to identify and

associated with a high proportion of correctness were also associated with a high rating of

confidence. However, structures that were difficult to identify and associated with a low

proportion of correctness were also associated with a moderately high rating of

confidence. This leads to one of two conclusions either (1) participants have low overall

subject area knowledge and cannot accurately assess their responses or (2) participants

have a moderate understanding of the subject area that allows them to feel confident in

the more easily identifiable structure but not as confident on more difficult structures. It

was surprising to see differentiation between student's confidence of easily identified

structures compared to more difficult structures. Students have developed enough of a

knowledge base to accurately assess their performance on easy structures; however, their

knowledge base is not comprehensive enough for students to be able to accurately assess

their responses to more cognitively-taxing structures. As students continue to expand

their knowledge base, they will develop a more accurate understanding of self-

assessment, which would correlate to a decrease in response correctness confidence

according to Dunning-Kruger.

The results of phase two illuminated areas of my project that future projects could

alter to improve and draw more definite conclusions. In this project, participants were

tested on three diagrams with varying visual representation factors. This meant the

proportion of correctly identified structure was the result of several independent

variables. Future studies could isolate individual variables to obtain more concise data.

For example, minimizing the differences in diagram dependent factors by using diagrams

with similar color, attention to detail, scale, and theme would allow for results to reflect

manipulations in the treatment of each diagram instead of a compound of variables.

Continuing to isolate variables responsible to changes in participants’ response, future

projects would be tempted to apply each manipulation to each diagram. However, if

every possible combination of manipulations and diagrams were presented to

participants, the identification of the anatomical structure would become obvious to

participants. Instead, future projects should focus on minimizing the queuing effect when

selecting the sequence order and manipulation for each diagram. Another aspect that

future projects could modify is the use of diagrams from a different source independent

of the curriculum or the textbook (Amerman 2018). The diagrams participants were

asked to identify structures from are sourced from their textbook, which means that

students were previously familiar with the provided diagram. Students with previous

experience with the diagrams have a greater opportunity to have already formed a mental

model of the anatomical topic. Future projects should ask students to identify structures

that have not been previously covered in the Human Anatomy and Physiology sequence.

Asking participants to identify structures outside of the scope of the course further

ensures that student’s formation of mental models is dependent on their one-hour

exposure to the diagram.

Simpler diagrams are traditionally assumed to be easier for less experienced

learners to understand (Lin et al 2017; Fenesi et al 2017); however, my results suggest

that details found on diagrams have important functions by serving as memory cues and

spatial markers, which assist students’ recall. Kottmeyer et al (2020) found that students

with low diagram compression ability especially benefited from extra supporting details

featured on diagrams. By removing these details from the diagram, students require more

time to orient themselves to the visual representation. The stress of reorienting to a visual

representation can lead to poor student performance. Results from phase two supports

this conclusion- removing leader lines from the diagrams was associated with a decreased

proportion of correct answers. This reorientation, however, is not dependent on the

placement of either the beginning or ending of the leader line.

The goal of this project was to explore intersections between student preferences

and student performance in anatomical diagrams. Students’ responses indicated that

forming spatial awareness between anatomical structures is important for mental model

formation. These relationships are formed from a combination of provided two-

dimensional diagrams and digital three-dimensional representations. When viewing two-

dimensional diagrams, students preferred simplified diagrams with leader lines for every

anatomical structure represented. This preference translated into a higher proportion of

correctly identified structures associated with the diagrams containing supporting visual

details and the use of visual cues to separate structures. Students also performed better

and felt more confident on the diagram with the least amount of leader lines total. The

conclusions made from this project support current research which suggest that while

students prefer three-dimensional diagrams (Tan et al 2012), students preform best on

detailed, two-dimensional diagrams (Fenesi et al 2017). The practical applications of this

project have the potential to better inform educators as they choose diagrams to integrate

into the classroom.

LITERATURE CITED

Ainsworth, S.. 1999. The functions of multiple representations. Computers & Education

33 131-152. PII: S0360-1315(99)00029-9

Amerman, E.. 2018. Human Anatomy & Physiology (Mastering A&P). 2

ed., Pearson.

Britson, C. 2020. Effect of Word Bank Provision for Lab Practical’s on Student

Performance in Human Anatomy and Physiology I and II courses. Journal of the

Human Anatomy and Physiology Society 24: 536-45. DOI:

10.21692/haps.2020.004

Butcher, K.R.. 2006. Learning from text with diagrams: Promoting mental model

development and inference generation. Journal of Educational Psychology 98.182-

197. DOI: 10.1037/0022-0663.98.1.182

Fenesi, B., Mackinnon, C., Cheng, L., Kim, J. A., Wainman, B. C.. 2017. The Effect of

Image Quality, Repeated Study, and Assessment Method on Anatomy

Learning. Anatomical Sciences Education. 10: 249-261. DOI: 10.1002/ase.1657

Khalil, M.K., Johnson, T.E., Lamar, C.H. 2005. Comparison of Computer-Based and

Paper-Based Imagery Strategies in Learning Anatomy. Clinical Anatomy 18: 457-

464.

Kottmeyer, A. M., Van Meter, P., Cameron, C.. 2020. Diagram comprehension ability of

college students in an introductory biology course. Advanced Physiology

Education. 44: 169-180. DOI: 10.1152/advan.00146.2018

Kriz, S., Hegarty, M.. 2007. Top-down and bottom-up influences on learning from

animations. International Journal of Human-Computer Studies 65 911-930. DOI:

10.1016.j.ijhcs.2007.06.005

Kruger, J., Dunning, D.. 2009. Unskilled and Unaware of It: How Difficulties in

Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments.

Psychology, 1, 30-46. DOI: 10.1037//0022-3514.77.6.1121

Lin, Y. Y., Holmqvist, K., Miyoshi, K., Ashida, H.. 2017. Effects of detailed Illustrations

on Science Learning: An Eye Tracking Study. Springer Inst Sci (45). 45: 557-581.

DOI: 10.1007/211251-017-9417-1

Mayer, R. E., Sims, V. K..1994. For Whom Is a Picture Worth a Thousand

Words? Extension of a Dual-Coding Theory of Multimedia Learning. Journal of

Educational Psychology (86). Vol 86, No 3: 389-401

Offeredahl, E. G., Arneson, J. B., Byrne, N.. 2017. Lighten the Load: Scaffolding Visual

Literacy in Biochemistry and Molecular Biology. CBE- Life Sciences Education,

15(1), 1-11.

Seel, N. M., Strittmatter, P.. 1989. Presentation of information by media and its effect on

mental models. Advances in Physiology, 58, 37-57. DOI: 10.1016/S0166-

4115(08)62146-9

Tan, S., Hu, A., Wilson, T., Ladak, H., Haase, P., Fung, K.. 2012. Role of a computer-

generated three-dimensional laryngeal model in anatomy teaching for advanced

learners. The Journal of Laryngology & Otology 126, 395-401. DOI:

10.1017.Soo22215111002830

West, C. K., Farmer, J. A., Wolff, P. M.. 1991. Instructional design implications from

cognitive science. Englewood Cliffs, NJ: Prentice Hall.

Wu, H., Krajcik, J. S., Soloway, E.. 2000. Promoting Conceptual Understanding of

Chemical Representations: Students’ Use of a Visualization Tool in the Classroom.

National Association of Research in Science Teaching.

Table 1: Human Anatomy and Physiology I student responses to Likert-style statements

on phase I survey which recorded undergraduate opinions on diagrams and their function

in the A&P curriculum. (SA= strongly agree, A= agree, N= neutral, D= disagree, SD=

strongly disagree; df=4)

Statement

An important part of my preparation for an anatomy

laboratory practical is understanding the spatial relationship

between structures in an organism.

If I have access to three-dimensional interactive anatomical

software, I would prefer to use it to study anatomical figures

than a two-dimensional image in a textbook or lab manual.

I feel comfortable learning to identify anatomical structures

using computer anatomical models rather than diagrams or

pictures of dissecting specimens.

When initially learning about various structures, I prefer

simplified diagrams or pictures that highlight one particular

structure rather than a diagram or picture that is detail

oriented.

I prefer images with leader lines that identify every structure

in an image rather than leader lines that only identify the

relevant structures in an image.

I prefer to learn anatomical structures on a three-dimensional

image rather than a two-dimensional image.

When I study a two-dimensional model, I can apply

anatomical knowledge to a three-dimensional subject easily.

When studying anatomical structures on a computer-based

model, I can apply my anatomical knowledge easily to an

image or diagram of a dissected specimen.

While studying anatomical structures, I seek computer-based

models from sources outside of the classroom to

contextualize my anatomical knowledge.

After studying an anatomical structure on a simplified

computer-based image or diagram, I am spatially aware of the

structure so that I’m able to apply my knowledge to a more

complex diagram.

Figure 1: Human Anatomy and Physiology II students' performance on Phase II which

consisted of correct identification of a single anatomical structure on each of the twelve

manipulated diagrams (Amerman 2018; Appendix C). Participant responses are

represented in this figure as the proportion of the number of correctly identified

anatomical structures to the number of incorrectly identified structures out of the thirty-

one participants.

Figure 2: Human Anatomy and Physiology II students’ performance in Phase II which

consisted of a participant self-assessment of their confidence in the correctness of their

identification of an anatomical structure from one of the twelve manipulated diagrams

(Amerman 2018; Appendix C). Participants rated their confidence on a scale from one to

ten; ten being the most confidence in the correctness of their answer and one being the

least confident in the correctness of their answer (in this figure, confidence ratings were

scaled to one hundred).

APPENDIX A

Phase I Survey Questions:

Preliminary Information:

Have you taken any other biology courses at the University of Mississippi?

A) Yes

B) No

If you answered yes:

- How many 100 level credits? _________

- How many 200 level credits? _________

- How many 300 level credits? _________

- How many 400 level credits? _________

Questionnaire:

Mark the response most representative of your feelings on the following statements.

1. An important part of my preparation for an anatomy laboratory practical is

understanding the spatial relationship between structures in an organism.