Dermatology: Practical and Conceptual


Review | Dermatol Pract Concept. 2022;12(4):e2022188 1

Theory-Based Approaches to Support 
Dermoscopic Image Interpretation Education: 

A Review of the Literature
Tiffaney Tran1, Niels K Ternov2, Jochen Weber3, Catarina Barata4, Elizabeth G Berry5,  

Hung Q Doan1, Ashfaq A Marghoob3, Elizabeth V Seiverling6, Shelly Sinclair7,  
Jennifer A Stein8, Elizabeth R Stoos5, Martin G Tolsgaard9, Maya Wolfensperger10,  

Ralph P Braun10, Kelly C Nelson1

1 Department of Dermatology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA

2 Department of Plastic Surgery, Herlev Hospital, Herlev, Denmark

3 Dermatology Service, Memorial Sloan Kettering Cancer Center, New York, NY, USA

4 Institute for Systems and Robotics; Instituto Superior Técnico, University of Lisbon, Lisbon, Portugal

5 Department of Dermatology, Oregon Health & Science University, Portland, OR, USA

6 Division of Dermatology, Maine Medical Center, Portland, ME, USA; Department of Dermatology, Tufts University School of Medicine, 

Boston, MA,USA

7 Department of Biology, Davidson College, Davidson, NC, USA

8 The Ronald O. Perelman Department of Dermatology, New York University School of Medicine, New York, NY, USA

9 Copenhagen Academy for Medical Education and Simulation; Department of Obstetrics, Copenhagen University Hospital Rigshospitalet, 

Copenhagen, Denmark

10 Department of Dermatology, University Hospital of Zürich, University of Zürich, Zürich, Switzerland

Key words: dermoscopy education, image interpretation education, pattern recognition, educational theory, container model

Citation: Tran T, Ternov NK, Weber J, et al. Theory-Based Approaches to Support Dermoscopic Image Interpretation Education: A Review 
of the Literature. Dermatol Pract Concept. 2022;12(4):e2022188. DOI: https://doi.org/10.5826/dpc.1204a188

Accepted: February 1, 2022; Published: October 2022

Copyright: ©2022 Tran et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-
NonCommercial License (BY-NC-4.0), https://creativecommons.org/licenses/by-nc/4.0/, which permits unrestricted noncommercial use, 
distribution, and reproduction in any medium, provided the original authors and source are credited.

Funding: None.

Competing Interests: None.

Authorship: All authors have contributed significantly to this publication.

Corresponding Author: Kelly C. Nelson, MD, Department of Dermatology, The University of Texas MD Anderson Cancer Center, 1400 
Pressler Street, Unit 1452, Houston, TX 77030, USA, Telephone: 713-745-1113, Fax: 713-745-3597, E-mail: kcnelson1@mdanderson.org 

Introduction: Efficient interpretation of dermoscopic images relies on pattern recognition, and the 
development of expert-level proficiency typically requires extensive training and years of practice. 
While traditional methods of transferring knowledge have proven effective, technological advances 
may significantly improve upon these strategies and better equip dermoscopy learners with the pattern 
recognition skills required for real-world practice.

ABSTRACT



2 Review | Dermatol Pract Concept. 2022;12(4):e2022188

healthcare providers estimated that at least six years of ex-

perience may be necessary to develop a sufficient level of 

competency [1]. 

The container model embodies a traditional instructional 

approach centered on the idea that the acquisition of knowl-

edge is comparable to filling one mind, or one mental filing 

cabinet, with as many facts and concepts as possible [2,3]. 

This learning theory was developed on the metaphor of our 

minds acting as containers capable of accumulating and 

retaining different items, whether real or metaphorical [4]. 

 Using this metaphor, knowledge is regarded as a commodity 

to be transferred from one medium to another [3]. 

The expectation that diagnostic concepts can be placed in 

a fixed cognitive “container,” where they can then be easily 

accessed, is misleading, especially in the setting of medical ed-

ucation. The successful recall of knowledge in the real-world 

context seems to be strongly influenced by the learning con-

text [5]. However, knowledge acquired using the container 

model is usually isolated from its context and thus static and 

inflexible. In practice, knowledge, even if highly case-specific, 

is not a readily available object to retrieve but something to 

reconstruct and adapt when faced with different situations [2].

Thus, opponents to the container model point out that in 

solving new problems, learners who have diligently absorbed 

declarative knowledge (eg facts and concepts that can be 

 verbalized) under the container model may still fail to appro-

priately retrieve and apply that knowledge [6]. Alternatively, 

learners may be able to successfully adapt their knowledge in 

the problem-solving process, but this successful retrieval and 

application may require a high cognitive load [6]. 

The cognitive load theory is an instructional the-

ory  derived from current understanding of cognitive  

architecture [7]. The term cognitive load refers to the learner 

Introduction

As a visual specialty, dermatology relies on the recognition 

of characteristic features and patterns within clinical and 

dermoscopic images of skin lesions. Other fields in medi-

cine, such as radiology and pathology, also rely on pattern 

recognition to formulate diagnoses and plans of care. In 

these fields, experts are distinguished from novices by their 

speed and accuracy in interpreting medical images and com-

pleting diagnostic tasks. While analyzing images of skin 

lesions, viewers may perform global interpretation, which 

encompasses holistic processing with immediate pattern rec-

ognition. Viewers may also perform feature search, which in-

volves the identification of specific features (eg, dermoscopic 

criteria) associated with normalcy or pathology.

Experts and novices differ in the order that they perform 

these mental processes during medical image analysis. Nov-

ices usually start by attempting to search deliberately for 

features and comparing their findings with their prior, al-

beit limited, knowledge and experiences. In contrast, experts 

typically reach a diagnostic conclusion relatively quickly 

through global interpretation and then seek to justify their 

conclusion with an efficient feature search.

Visual diagnostic skills at the expert level require the 

ability to perform global interpretation—or efficient pattern 

recognition without deliberate search strategies. Traditional 

educational methods that primarily drill declarative knowl-

edge (e.g., specific features) have been largely ineffective in 

teaching novices the pattern recognition skills required for 

real-world practice. With these conventional training mo-

dalities, expert-level proficiency in dermoscopy has typically 

required extensive training and years of practice. A diagnos-

tic accuracy study performed among medical students and 

Objectives: A narrative review of the literature was performed to explore emerging directions in med-
ical image interpretation education that may enhance dermoscopy education. This article represents 
the first of a two-part review series on this topic.

Methods: To promote innovation in dermoscopy education, the International Skin Imaging Collabo-
rative (ISIC) assembled a 12-member Education Working Group that comprises international dermos-
copy experts and educational scientists. Based on a preliminary literature review and their experiences 
as educators, the group developed and refined a list of innovative approaches through multiple rounds 
of discussion and feedback. For each approach, literature searches were performed for relevant articles.

Results: Through a consensus-based approach, the group identified a number of emerging directions 
in image interpretation education. The following theory-based approaches will be discussed in this 
first part: whole-task learning, microlearning, perceptual learning, and adaptive learning.

Conclusions: Compared to traditional methods, these theory-based approaches may enhance dermos-
copy education by making learning more engaging and interactive and reducing the amount of time 
required to develop expert-level pattern recognition skills. Further exploration is needed to determine 
how these approaches can be seamlessly and successfully integrated to optimize dermoscopy education. 



Review | Dermatol Pract Concept. 2022;12(4):e2022188 3

Objectives

This article represents the first of a two-part review series 

on novel instructional approaches in image interpretation 

education that could translate to dermoscopic educational 

interventions. In this first part, we will present a collection 

of theory-based approaches —such as whole case learning, 

microlearning, perceptual learning, and adaptive learning— 

that could enhance dermoscopic image interpretation edu-

cation. While these emerging directions may also apply to 

general dermatology education, the scope of this series is 

limited to dermoscopy education.

Methods

To promote innovation in dermoscopy education, the In-

ternational Skin Imaging Collaborative (ISIC) assembled a 

12-member Education Working Group that comprises in-

ternational dermoscopy experts and educational scientists.

For this initiative, the group convened virtually on a regular

basis to discuss novel methods in medical image interpreta-

tion education that could be translate to dermoscopy train-

ing programs. Based on a preliminary literature review as

well as their experiences as educators, the group developed

and refined a list of innovative approaches through multiple

rounds of discussion and feedback.

For each approach, literature searches were performed 

in the PubMed and Google Scholar databases for relevant 

English-language articles. Search strategies included terms 

for concepts of dermoscopy education, image interpretation 

education, and health science education in addition to the 

instructional approach under investigation. Articles pub-

lished since 2000 were preferred for inclusion, but articles 

published before 2000 were also considered, especially when 

seeking to understand the historical and theoretical under-

pinnings of some approaches. Additional articles were iden-

tified among the references of retrieved articles and through 

discussions with educational scientists.

Relevant literature findings on the educational theories, 

methods, and concepts identified during the consensus pro-

cess are presented in this review series. The theory-based 

approaches described in the first part of this series include: 

whole-task learning, microlearning, perceptual learning, and 

adaptive learning.

Results

Whole-Task Learning (4-C/ID)
Overview

Whole-task learning is a time-efficient instructional design 

in medical education that teaches complex skill development 

mental bandwidth to complete tasks [7]. It is based on the 

idea that working memory—where information is stored 

temporarily—has limited capacity to store and use new in-

formation [7]. As a result, if the amount of new information 

exceeds mental capacity, further learning and accurate deci-

sion making will be impaired [8]. It is important for educa-

tors to consider cognitive load when designing instructional 

materials to maintain the overall load within the optimal 

range for learning and performance.

For difficult tasks, learners may primarily rely on delib-

erate and purposeful thinking, and this process may strain 

their working memory, negatively affecting their ability to 

complete the tasks. In cognitive psychology, the dual process 

theory recognizes two thought systems: slow  (deliberate) 

thinking and fast (automatic) thinking [9]. For medical im-

age interpretation, the dual process theory manifests as a 

two-component diagnostic strategy in which the fast system 

facilitates pattern recognition and the slow system facilitates 

analytical reasoning [10]. Within a medical simulation, ac-

cessing relevant knowledge and skills while assessing the 

simulated clinical environment may create a high extraneous 

cognitive load for learners, causing poor performance [8]. 

The high cognitive load experienced by learners in medical 

simulations may be explained by their extensive use of slow 

thinking as opposed to fast thinking, the former requiring 

considerable mental effort and use of mental resources.

Fast thinking, or non-analytical reasoning, is a key com-

ponent of expert performance for diagnostic tasks that rely 

on pattern recognition, such as skin lesion classification in 

dermatology and X-ray interpretation in radiology [11]. For 

image interpretation education, an important goal is for nov-

ices to gradually develop a degree of automaticity in pattern 

recognition, which translates to a low cognitive load [10]. 

Traditional teaching methods based on the container model 

have been generally ineffective in both teaching flexible 

knowledge and training automaticity in novices.

In dermoscopy education, educators who use the con-

tainer model usually provide instruction to passive learn-

ers on a defined set of diagnostic features, adding to their 

 “containers.” In teaching learners to detect the relative pres-

ence or absence of a feature, this approach in effect requires 

real-world stimuli to likewise fit a binary interpretation 

(eg   present/absent, melanoma/non-melanoma). However, 

real-world stimuli frequently present on a continuum, or a 

sliding scale, where concerning features may be completely 

non-existent, obviously present, or extremely subtle.

Dermoscopy education requires instructional approaches 

that transfer flexible knowledge on the continuous nature 

of features and their clinically relevant contexts. This review 

seeks to explore an array of emerging theory-based approaches 

in image interpretation education that may displace the con-

tainer model and enhance dermoscopy training programs.



4 Review | Dermatol Pract Concept. 2022;12(4):e2022188

Applications in Dermoscopy Education

In whole-task learning, authentic scenarios, structured in a 

way to facilitate skill transfer to clinical encounters, serve 

as the framework for learning. This approach contrasts with 

conventional teaching models that focus on didactic lectures, 

which are then supported by hypothetical scenarios. Whole-

task learning could be applied to dermoscopy education 

to promote problem-solving skills and foster professional 

independence. Learning tasks may involve addressing skin 

complaints in hypothetical patient encounters. Learners then 

receive instruction on the dermoscopic appearance of com-

mon dermatologic diagnoses (supportive information).

Dermoscopic training programs that involve case-

based learning could be adapted to whole-task learning by 

re-structuring the curriculum with cases at the forefront and 

re-imagining each case as a series of learning tasks [17]. For 

example, dermoscopic cases are introduced prior to receiv-

ing instruction. As they navigate through cases, learners may 

receive further information on specific dermoscopic features 

and management approaches in the form of didactic lec-

tures, multimedia content, or other teaching materials [18]. 

After completing the didactic portion, learners may then en-

gage in repetitive practice to develop task automaticity and 

efficiency.

Microlearning
Overview

Microlearning is an instructional approach that involves 

segmenting the curriculum into short bursts, or small bites, 

of learning [19]. In contrast to traditional training sessions 

with “massed” practice, microlearning sessions may involve 

spaced review and distributed practice, increasing on-task 

attention and decreasing mind wandering [20]. According 

to the “forgetting curve,” memory retention declines over 

time as learners tend to forget much of their learned material 

within hours or days [19]. Microlearning seeks to address 

this trend by introducing and re-introducing lessons in short 

bursts. Through distributed practice, microlearning pro-

motes the transfer of information from short-term to long-

term memory storage [19]. 

With the microlearning approach, learners experience 

low cognitive load since working memory does not become 

overstrained, and this maintains learning capacity [21]. In 

reducing mental fatigue, this strategy increases learning re-

tention and efficiency [19]. While microlearning lessons are 

usually self-paced, learners tend to complete them faster 

given their high level of engagement [19]. 

Applications in Medical Education

In recent years, microlearning modules have become 

more readily available to learners with the emergence of 

through authentic clinical scenarios [12]. For learners, spon-

taneous transfer of knowledge from the learning situation 

to the clinical environment is challenging [13]. Through 

incorporation of real-life problems and fragmentation 

of instruction, whole-task learning aims to teach founda-

tional knowledge in a way that fosters the “transfer out” of 

knowledge to actual practice. This approach also seeks to 

facilitate skill transfer in a manner that attends to cognitive 

load [12]. 

A specific whole-task learning strategy is four-c omponent 

instructional design (4-C/ID). The four components in 4-C/ID  

are: (1) learning tasks, (2) supportive information, (3) just-

in-time information, and (4) part-task practice [12]. Learn-

ing tasks, which function as the backbone of 4-C/ID, are 

authentic tasks sequenced from simple to complex in terms 

of difficulty and organized into “task classes.” Supportive in-

formation may be presented at the beginning of a task class 

and provide foundational knowledge. Just-in-time informa-

tion may be provided right when the learner needs it for a 

specific task. Part-task practice is an optional component in 

which the learner is given the opportunity to practice a spe-

cific task in order to develop a degree of automaticity. By 

shifting the focus of learning from lectures to clinical scenar-

ios, learners may better appreciate the educational content 

and its relevance to their professional roles [14]. 

Whole-task learning is similar to case-based learning in 

that both emphasize realistic clinical situations in the in-

structional design. In case-based learning, learners engage 

in group-based discussions of authentic patient cases and 

receive guidance and feedback from instructors [15]. Learn-

ers are usually expected to prepare on their own through 

self- directed learning in advance of the case-based learning 

sessions [15]. In whole-task learning, learners are presented 

with authentic clinical scenarios prior to receiving formal 

instruction. As learners navigate the scenarios, they are pro-

vided further information relevant to the scenarios in a struc-

tured delivery format.

Applications in Medical Education

Task-based learning has been applied in surgical education 

in recent years. A randomized controlled study conducted 

among surgical interns implemented task-based learning in 

an inanimate surgical skills laboratory setting [16]. Com-

pared to the control group, the intervention group per-

formed better on post-intervention assessments and required 

less time to complete the clinical procedure [16]. Another 

qualitative study evaluated the feasibility and efficacy of 

whole-task learning in a web-based doctoral-level pharma-

cotherapy course and garnered positive results [14]. Learners 

expressed that by posing authentic scenarios, the complex 

delivery format provided them an opportunity to identify 

with their future health profession [14]. 



5

education, a microlearning module in which learners exclu-

sively practice diagnosing seborrheic keratosis (SKs) would 

result in blocked practice, while one requiring a learner to 

distinguish between SKs, benign nevi, and melanomas, pre-

sented in a random order, would result in interleaved prac-

tice. In a before-and-after study for a dermoscopy training 

program, blocked practice for benign lesions resulted in high 

specificity for benign lesions but poor sensitivity for malig-

nant lesions in that participants would frequently categorize 

melanomas as, for instance, SKs [28]. Sensitivity for malig-

nant lesions subsequently improved with the adoption of in-

terleaved practice [28]. 

By segmenting complex tasks into smaller units, micro-

learning represents a powerful teaching tool for dermoscopy 

education. It may enable an efficient transfer of expert-level 

pattern recognition skills to novices, especially when im-

plemented through technology tools such as smartphone 

apps. In bridging the gap between formal and informal 

learning, the use of microlearning technology may enhance 

learner engagement and motivation as well as knowledge 

 retention [29]. Microlearning modules may also be suitable 

for gamification in which game design principles are applied 

to enhance the learning experience and activate intrinsic re-

ward pathways. 

Perceptual Learning
Overview

Perceptual learning is a learning method that challenges the 

container model theory by promoting the idea of experience 

as fundamental to developing expertise [30]. In neuropsy-

chology, perceptual learning refers to the changes that occur 

in neural circuitry as a result of experience, resulting in the 

development of sensory discrimination [31]. This phenom-

enon explains how we learn to discriminate between faces, 

speech sounds, and musical pitches. For visual discrimina-

tion training, this approach relies on repeated exposures to 

numerous stimuli (eg visual features) so that one learns to 

perceive subtle differences between the stimuli. The concept 

of perceptual learning may be applied to visual specialties 

in which the educator teaches key diagnostic features and 

then creates opportunities for learners to practice recogniz-

ing these features with feedback.

For medical image interpretation education, the two 

components of perceptual learning are discovery and 

 fluency [6]. In the discovery phase, students learn to identify 

new information relevant to the diagnostic task by ignor-

ing less relevant information and extracting the more salient 

points. Using inattentional selectivity, learners may process 

a large amount of information from a case. Fluency comes 

with practice and refers to the student ability to efficiently 

recognize the information needed for diagnostic tasks.

multimedia content that can be easily accessed via personal 

devices. In a Dutch non-randomized study involving medical 

and biomedical university students, investigators employed 

an open-source mobile application (or “app”) to teach cir-

culation and respiration using microlearning and spaced 

review [22]. For a month before the exam, learners used 

the app to complete training modules with practice assess-

ments that reviewed educational content and provided feed-

back. Intensive app users performed significantly better on 

the final exam compared to moderate users and non-users, 

though these results may also be correlated with increased 

time spent learning [23]. 

Applications in Dermoscopy Education

In dermoscopy education, a real-life example of microle-

arning can be found in a telementoring framework model 

called Project ECHO (Extension for Community Health 

Outcomes) [24]. As an effective alternative to on-site men-

toring, tele-mentoring allows learners to process the cases 

with real-time guidance from dermoscopy experts [25]. In 

Project ECHO, teaching sessions occur on a monthly ba-

sis and pair a didactic micro-lecture with learner presen-

tations of real-life challenging cases encountered during 

patient care. A before-and-after study among primary care 

providers demonstrated that ECHO attendance increased 

participants’ ability to interpret dermoscopic images of skin 

cancer [25]. 

Another example of microlearning in dermoscopy 

can be found in the educational webcasts posted by the 

International Dermoscopy Society (IDS). These web-

casts include short YouTube videos of 5 to 10 minutes 

in length organized into disease-based learning (Level 1), 

 morphology-based learning (Level 2), and context-based 

learning (Level 3) as well as case-based learning [26]. To 

facilitate conceptual understanding, these webcasts could 

be expanded by posting dermoscopic images with practice 

questions plus key points in a microlearning format on a 

weekly or monthly basis.

Since microlearning can be applied to drill certain topics 

or specific skills, educators may consider whether to imple-

ment “blocked” or “interleaved” practice. Many programs 

involve “blocked” practice in which the learner practices 

specific skills (eg A, B, C) one at a time in isolation (eg 

AAA BBB CCC) [27]. An alternative to “blocked” 
practice is “interleaved” practice in which learners 

practice multiple different skills in an intermixed order 

(eg ABC BCA CAB). In interleaved practice, the amount 

of practice devoted to a specific skill becomes spaced, or 

distributed, across the learn-ing session [27]. 

By continuously exposing learners to multiple relevant 

topics, interleaved practice may be more effective in pre-

paring learners for real-life applications. In dermoscopy 

Review | Dermatol Pract Concept. 2022;12(4):e2022188



6 Review | Dermatol Pract Concept. 2022;12(4):e2022188

Students may have different starting points for a given topic, 

or they may learn at different paces based on their individual 

abilities and the instructional method being used. Adaptive 

approaches represent a solution to these problems: by re-

sponding to the learner response times and accuracy rates, 

adaptive algorithms can repeat content, or adapt content dif-

ficulty, to optimize the learning process [36]. 

Adaptive response time-based sequencing (ARTS) is an 

example of an adaptive learning approach that customizes 

the learning sequence based on performance data [37]. Once 

the algorithm has detected mastery of a specific concept ac-

cording to objective learning criteria, it can retire that con-

cept and shift to focus on the learner weaker areas. Learning 

criteria should correlate with a given level of proficiency 

and could involve a number of accurate responses provided 

within a specified amount of time, correlating with a degree 

of automaticity. Training is considered complete when all cri-

teria are met.

Applications in Medical Education

Adaptive algorithms have been successfully applied in 

teaching transesophageal echocardiography (TEE) image 

interpretation. Like dermoscopic image interpretation, TEE 

interpretation involves recognition of diagnostic patterns. In 

one teaching method, an algorithm modified the sequence 

of and time intervals between different TEE cases to suit 

each learner needs [38]. It evaluated both response speed 

and accuracy to determine whether to retire or re-sequence a 

specific concept. This method proved effective in improving 

response time and accuracy and optimizing performance for 

TEE learners. 

For electrocardiography (EKG) interpretation, adap-

tive learning has also been successful in promoting content 

mastery. Reading an EKG, like evaluating a skin lesion in 

dermoscopy, requires pattern recognition skills that nov-

ices are expected to obtain via experiential learning [30]. 

With ARTS, the pace of learning was adapted for each EKG 

learner based on response time and accuracy. As with the 

previous example, a concept was retired only if the learner 

achieved the target response time while maintaining accu-

racy. If both measures were not achieved, the concept was 

re-sequenced into the learning sequence.

Applications in Dermoscopy Education

For dermoscopy education, adaptive algorithms may grad-

ually increase the difficulty of dermoscopic images based 

on learner performance to generate faster improvements in 

performance. Alternatively, if a learner repeatedly fails to 

recognize a specific dermoscopic feature, additional images 

containing the feature may be shown until the learner starts 

to “see” the feature. Since learners encounter new content 

according to a personalized training schedule, demotivation 

Applications in Medical Education

Through perceptual learning, learners receive exposure to 

real-life examples, engage in repetitive practice, and gradu-

ally learn to recognize important diagnostic features quickly 

and accurately. Perceptual learning has been applied to ra-

diology and electrocardiogram (EKG) image interpretation 

training, where learners have demonstrated gains in accu-

racy and fluency [6,30]. More recently, perceptual learning 

has been applied to dermatology education, where learners 

classify clinical images of rashes and skin lesions by mor-

phology, configuration, and distribution [32]. Through per-

ceptual learning, learners demonstrated the ability to quickly 

and accurately identify skin lesion characteristics at a level 

comparable to that of expert dermatologists.

Applications in Dermoscopy Education

In a dermoscopy training program for primary care pro-

viders, educators applied a heuristic training approach that 

resembled the discovery phase of the perceptual learning ap-

proach [33]. In the heuristic strategy, learners are expected 

to devise their own heuristics, or mental shortcuts, for future 

decision-making based on their experiences. Following an 

introductory didactic training session on classical dermo-

scopic features, learners in the heuristic training arm were 

provided the opportunity to view a series of dermoscopic 

images with minimal guidance from instructors. Labeled 

with the diagnosis only, these images did not contain further 

annotation or description, and learners were expected to 

discover salient features on their own. On post-intervention 

assessments, learners in the heuristic training arm performed 

as well as learners who had received feedback on the salient 

features in those images.

Adaptive Learning
Overview

Adaptive learning is an educational approach that optimizes 

learning for the individual learner through innovative tech-

nology tools [6]. This approach features an adaptive algo-

rithm that tailors the individual learning sequence according 

to their strengths and weaknesses. Adaptive algorithms re-

semble an automated form of the deliberate practice strategy 

commonly used to achieve expert performance in music and 

sports. In deliberate practice, a teacher evaluates student per-

formance and recommends practice activities (training tasks) 

and practice objectives (training goals) based on the teacher 

prior experiences and the student needs [34]. Students follow 

teachers recommendations, practice with full concentration, 

and receive or self-generate immediate feedback [34]. 

In traditional medical education, pre-determined lecture 

or training schedules could not be easily adapted or mod-

ified to accommodate the individual student needs [35]. 



Review | Dermatol Pract Concept. 2022;12(4):e2022188 7

We envision a hypothetical dermoscopy training pro-

gram that combines the strengths of each approach pre-

sented in this article. In this program, learning concepts, such 

as a specific dermoscopic diagnoses, would be organized as 

their own unit. Each unit may be prefaced by a real-world 

clinical scenario that promotes whole-task learning. Educa-

tional content on the dermoscopic diagnosis (eg clinical pre-

sentation, dermoscopic appearance) could then be presented 

via microlearning modules that deliver instruction in small 

segments to minimize extraneous cognitive load.

Each microlearning module may also include multiple 

example images of each diagnostic feature plus new cases 

for perceptual learning. These example images and cases 

could be hosted on a user-friendly application that contains 

elements of game design and provides immediate feedback 

to learners. Adaptive learning algorithms built into the ap-

plication would either re-sequence or retire cases according 

and mental fatigue may be reduced [39]. Adaptive learning 

ensures that each student acquires the knowledge needed, 

in whatever sequence and at whatever pace, to reach their 

desired level of mastery. In addition, innovative technolo-

gies with user interface interactions may enable performance 

tracking and personalized modifications for efficient learning.

Conclusions

A summary of the instructional approaches explored in the 

first part of this review series is included in Table 1. In gen-

eral, training programs that apply microlearning modules, 

perceptual learning cases, and/or adaptive learning algo-

rithms may enable novices to acquire expert-level knowl-

edge in an effective manner. Meanwhile, whole-task learning 

equips learners for real-life clinical situations using hypo-

thetical clinical scenarios.

Table 1. Summary of the educational theories presented in the first part of this review series plus 
examples of existing or potential applications in dermoscopy education.

Educational 
Theory Description Application(s) in Dermoscopy Education

Container 
Model

• Learners receive passive instruction and fill 
their mental “container” with as many facts and 
concepts as possible.

• Acquired knowledge is usually static and 
inflexible because it is isolated from its 
context.

Existing Applications
• didactic lectures
• rules-based algorithms

Whole-Task 
Learning

• Curriculum design is based on authentic clinical 
scenarios and comprises 4 components: (1) 
learning tasks, (2) supportive information, (3) 
just-in-time information, and (4) part-time 
practice.

Potential Application
• curriculum design: structured as a series of 

clinical scenarios based on real-life cases

Micro-learning • Educational content is segmented into short 
bursts, or small bites, of learning that may be 
spaced apart.

• This approach is expected to enhance 
engagement, increase knowledge retention, and 
decrease mental fatigue.

Existing Application
• Project ECHO, developed by dermatology 

faculty at MaineHealth

Perceptual 
Learning

• For medical image interpretation, fine visual 
discrimination skills are developed through 
repeated exposures to numerous examples of 
important visual features.

• With feedback and practice, novices may learn 
to efficiently extract important features and 
ignore irrelevant ones.

Existing Applications
• YouDermoscopy, created and developed by 

Meeter Congressi
Potential Applications
• library of training cases: learners classify 

hundreds of images and receive feedback on 
performance

Adaptive 
Learning 

• Adaptive algorithms respond to the individual 
performance data and make personalized 
modifications to the training schedule.

• Each individual training schedule is tailored to 
his/her strengths and weaknesses in order to 
optimize learning outcomes.

Potential Applications
• learning modules: learners complete assessments; 

adaptive algorithms retire specific learning 
concepts based on objective mastery criteria

ECHO = Extension for Community Healthcare Outcomes.



8 Review | Dermatol Pract Concept. 2022;12(4):e2022188

9. Kahneman D. Thinking, Fast and Slow. New York, NY, US: Far-

rar, Straus and Giroux; 2011.

10. Gronchi G, Giovannelli F. Dual process theory of thought and de-

fault mode network: a possible neural foundation of fast thinking. 

Front Psychol. 2018;9:1237. DOI: 10.3389/fpsyg.2018.01237. 

PMID: 30065692. PMCID: PMC6056761.

11. Norman G, Young M, Brooks L. Non-analytical models of clin-

ical reasoning: the role of experience. Med Educ. 2007;41(12): 

1140-1145. DOI: 10.1111/j.1365-2923.2007.02914.x. PMID: 

1800 4990.

12. Daniel M, Stojan J, Wolff M, et al. Applying four-component 

instructional design to develop a  case presentation curriculum. 

Perspect Med Educ. 2018;7(4):276-280. DOI: 10.1007/s40037-

018-0443-8. PMID: 29992438. PMCID: PMC6086819.

13. Castillo J-M, Park YS, Harris I, et al. A critical narrative review 

of transfer of basic science knowledge in health professions 

education. Med Educ. 2018;52(6):592-604. DOI: 10.1111/

medu.13519. PMID: 29417600.

14. Pittenger AL, Olson-Kellogg B. Leveraging learning technol-

ogies for collaborative writing in an online pharmacotherapy 

course. Distance Educ. 2012;33(1):61-80. DOI: 10.1080/0158 

7919.2012.667960

15. Bi M, Zhao Z, Yang J, Wang Y. Comparison of case-based learn-

ing and traditional method in teaching postgraduate students of 

medical oncology. Med Teach. 2019;41(10):1124-1128. DOI: 

10.1080/0142159X.2019.1617414. PMID: 31215320.

16. Velmahos GC, Toutouzas KG, Sillin LF, et al. Cognitive task anal-

ysis for teaching technical skills in an inanimate surgical skills 

laboratory. Am J Surg. 2004;187(1):114-119. DOI: 10.1016/j.

amjsurg.2002.12.005. PMID: 14706600.

17. Sherbino J, Dore KL, Wood TJ, et al. The relationship be-

tween response time and diagnostic accuracy. Acad Med. 

2012;87(6):785-791. DOI: 10.1097/ACM.0b013e318253acbd. 

PMID: 22534592.

18. Janssen-Noordman AM, Merrienboer JJ, van der Vleuten CP, 

Scherpbier AJ. Design of integrated practice for learning pro-

fessional competences. Med Teach. 2006;28(5):447-452. DOI: 

10.1080/01421590600825276. PMID: 16973459.

19. Shail MS. Using micro-learning on mobile applications to increase 

knowledge retention and work performance: a review of liter-

ature. Cureus. 2019;11(8):e5307. DOI: 10.7759/cureus.5307. 

PMID: 31511813. PMCID: PMC6716752.

20. Metcalfe J, Xu J. People mind wander more during massed 

than spaced inductive learning. J Exp Psychol Learn Mem 

Cogn. 2016;42(6):978-984. DOI: 10.1037/xlm0000216. PMID: 

26618908.

21. Raman M, McLaughlin K, Violato C, Rostom A, Allard JP, 

Coderre  S. Teaching in small portions dispersed over time en-

hances  long-term knowledge retention. Med Teach. 2010;32(3): 

250-255. DOI: 10.3109/01421590903197019. PMID: 20218841.

22. Lameris AL, Hoenderop JG, Bindels RJ, Eijsvogels TM. The 

impact of formative testing on study behaviour and study per-

formance of (bio)medical students: a smartphone application 

intervention study. BMC Med Educ. 2015;15:72. DOI: 10.1186/

s12909-015-0351-0. PMID: 25889923. PMCID: PMC4404663.

23. Thijssen DHJ, Hopman MTE, van Wijngaarden MT, Hoenderop 

JGJ, Bindels RJM, Eijsvogels TMH. The impact of feedback 

during formative testing on study behaviour and performance 

of (bio)medical students: a randomised controlled study. BMC 

Med Educ. 2019;19(1):97. DOI: 10.1186/s12909-019-1534-x. 

PMID: 30943962. PMCID: PMC6446354.

to learner performance, indicated by response time and ac-

curacy. Upon completion of all microlearning modules for 

the unit, learners are offered the opportunity to revisit the 

clinical scenario from the beginning of that unit.

To evaluate the impact of these educational approaches 

on dermoscopic image interpretation skills, educators may 

assess learner performance at multiple time points using val-

idated instruments that measure both fluency and accuracy 

in diagnosing lesions. Investigators may also perform pro-

spective audits of clinical diagnoses versus histopathological 

diagnoses among participants in the training program.

With the container model, learners were taught to in-

terpret images using a given set of diagnostic features and/

or rule-based algorithms. However, learners often struggled 

with manipulating and applying acquired knowledge to new 

images. Compared to traditional methods, emerging ap-

proaches in image interpretation education are more inter-

active and learner-centered. These approaches may improve 

learning outcomes by grounding learning in real-life clinical 

scenarios (whole-task learning) or delivering instruction in 

short segments (microlearning). For dermoscopic training, 

perceptual learning and adaptive learning may be especially 

valuable in that they provide immediate feedback and adapt 

the pace of learning to learner performance, respectively. The 

second part of this series will continue exploring instruc-

tional strategies and methods in image interpretation educa-

tion that could also support dermoscopy education.

References

1. Ternov NK, Vestergaard T, Hölmich LR, et al. Reliable test of 

clinicians’ mastery in skin cancer diagnostics. Arch Dermatol 

Res. 2021;313(4):235-243. DOI: 10.1007/s00403-020-02097-8. 

PMID: 32596742.

2. Bereiter C, Scardamalia M. Beyond Bloom’s taxonomy: re-

thinking knowledge for the knowledge age. In: Hargreaves A, 

 Lieberman A, Fullan M, Hopkins D, eds. International Hand-

book of Educational Change. Vol 5. Dordrecht: Springer; 1998: 

675-692.

3. Bowden D. The limits of containment: text-as-container in com-

position studies. Coll Compos Commun. 1993;44(3):364-379. 

DOI: 10.2307/358989.

4. Lakoff G, Johnson M. Metaphors We Live By. Chicago, IL, US: 

The University of Chicago Press; 2003.

5. Norman G, Bordage G, Page G, Keane D. How specific is case 

specificity? Med Educ. 2006;40(7):618-623. DOI: 10.1111/ 

j.1365-2929.2006.02511.x. PMID: 16836533.

6. Kellman PJ. Adaptive and perceptual learning technologies 

in medical education and training. Mil Med. 2013;178(10): 

98-106.  DOI: 10.7205/MILMED-D-13-00218. PMID: 24084310.

7. Sweller J. Cognitive load theory. In: Mestre JP, Ross BH, eds. The 

Psychology of Learning and Motivation: Cognition in Education. 

Vol 55. San Diego, CA, US: Elsevier Academic Press; 2011:37-76.

8. Fraser KL, Ayres P, Sweller J. Cognitive load theory for the de-

sign of medical simulations. Simul Healthc. 2015;10(5):295-307. 

DOI: 10.1097/SIH.0000000000000097. PMID: 26154251.



Review | Dermatol Pract Concept. 2022;12(4):e2022188 9

distribution. J Am Acad Dermatol. 2015;72(3):489-495. DOI: 

10.1016/j.jaad.2014.11.016. PMID: 25592621.

33. Seiverling EV, Ahrns HT, Greene A, et al. Teaching benign skin 

lesions as a strategy to improve the triage amalgamated dermo-

scopic algorithm (TADA). J Am Board Fam Med. 2019;32(1): 

96-102. DOI: 10.3122/jabfm.2019.01.180049. PMID: 3061 

0147. PMCID: PMC8039813.

34. Ericsson KA. Acquisition and maintenance of medical exper-

tise: a perspective from the expert-performance approach with 

deliberate practice. Acad Med. 2015;90(11):1471-1486. DOI: 

10.1097/ACM.0000000000000939. PMID: 26375267.

35. Mettler E, Massey CM, Kellman PJ. A comparison of adaptive 

and fixed schedules of practice. J Exp Psychol Gen. 2016;145(7): 

897-917. DOI: 10.1037/xge0000170. PMID: 27123574.  PMCID:  

PMC6028005.

36. Mettler E, Massey C, Kellman P. Improving adaptive learning 

technology through the use of response times. Proc Annu Conf 

Cogn Sci Soc. 2011;33:6.

37. Mettler E, Kellman PJ. Adaptive response-time-based category 

sequencing in perceptual learning. Vis Res. 2014;99:111-123. 

DOI: 10.1016/j.visres.2013.12.009. PMID: 24380704. PMCID: 

PMC6124487.

38. Romito BT, Krasne S, Kellman PJ, Dhillon A. The impact of 

a perceptual and adaptive learning module on transoesoph-

ageal echocardiography interpretation by anaesthesiology 

residents. Br J Anaesth. 2016;117(4):477-481. Br J Anaesth. 

2016 Oct;117(4):477-481. DOI: 10.1093/bja/aew295. PMID: 

28077535.

39. Möckel T, Beste C, Wascher E. The effects of time on task in 

response selection: An ERP study of mental fatigue. Sci Rep. 

2015;5(1):10113. DOI: 10.1038/srep10113. PMID: 26054837. 

PMCID: PMC4460573.

24. Arora S, Thornton K, Murata G, et al. Outcomes of treatment 

for hepatitis C virus infection by primary care providers. 

N Engl J Med. 2011;364(23):2199-2207. DOI: 10.1056/NEJ-

Moa1009370. PMID: 21631316. PMCID: PMC3820419.

25. Seiverling EV, Li D, Stevens K, Cyr P, Dorr G, Ahrns H.   

Distance  learning and spaced review to complement  dermoscopy  

training for primary care. Dermatol Pract Concept. 2021; 

11(2):e2021030. DOI: 10.5826/dpc.1102a30. PMID: 3395 4013.  

PMCID: PMC8060017.

26. International Dermoscopy Society (IDS). Webcasts. International 

Dermoscopy Society. Available from https://dermoscopy-ids.org 

/education-elearning/podcasts/. Published 2021. Updated March 

19, 2021. Accessed October 12, 2021.

27. Taylor K, Rohrer D. The effects of interleaved practice. Appl 

Cogn Psychol. 2010;24:837-848. DOI: 10.1002/acp.1598.

28. Ahrns HT, Butt M, Khalsa A, et al. Dermoscopy for family med-

icine residents: teaching benign neoplasms. Paper presented at: 

North American Primary Care Research Group Annual Meeting; 

November 20, 2017, 2017; Montreal, Quebec, Canada.

29. Buchem I, Hamelmann H. Microlearning: a strategy for ongoing 

professional development. eLearning Papers. 2010;21:1-15.

30. Krasne S, Stevens CD, Kellman PJ, Niemann JT. Mastering 

electrocardiogram interpretation skills through a perceptual 

and adaptive learning module. AEM Educ Train. 2020 May 

5;5(2):e10454. DOI: 10.1002/aet2.10454. PMID: 33796803. 

PMCID: PMC7995930.

31. Kellman PJ. Perceptual learning. In: Pashler H, Gallistel R, eds. 

Stevens’ Handbook of Experimental Psychology: Learning, 

Motivation, and Emotion. Vol 3. 3 ed.: John Wiley & Sons Inc; 

2002:259-299.

32. Rimoin L, Altieri L, Craft N, Krasne S, Kellman PJ. Training pat-

tern recognition of skin lesion morphology, configuration, and