Clinical reasoning may be defined as the cognitive processes involved in arriving at a diagnosis or treatment plan.1  It is central to effective medical practice, yet it remains elusive and continues to present a challenge to clinical teachers and learners.2  The antemortem misdiagnosis rate at autopsies and audits of adverse events in hospitals attest to the continuing importance of clinical reasoning, despite the introduction of guidelines and advances in technology.3,4  The level of quality and accuracy of the clinical reasoning that underlies diagnostic and management decisions is the difference between trainees and expert clinicians. The process of learning clinical reasoning may be assisted by using think aloud.

Think aloud is a research method used to study cognition and is considered the optimal method to capture thought processes.5  Cognitive psychologists view thinking as a temporal sequence of mental states. Each state contains information retrieved from long-term memory, perceptions from the environment, and the inferences generated from the recovered information. Think aloud occurs when individuals verbalize their thoughts while performing a task. It is our view that during think aloud individuals do not describe or explain what they are thinking; they simply verbalize how they are using the available information to generate a solution to a problem.5  For example, when asked to multiply 17 by 13, an individual might say, “17 × 10 is 170 and 17 × 3 is 51: adding these gives the answer 221.” When people verbalize how they are thinking as they perform a task, the sequence of thoughts is not changed by the added instruction to “think aloud.”5 

Expert clinicians gather and use clinical information to generate accurate diagnoses. However, this is usually done in an implicit and frequently automatic process. Think aloud is a method of making clinical reasoning processes more explicit, so that inexperienced clinicians are able to learn from these processes. Think aloud methods provide explanations and descriptions of complex information, and parallels the clinical reasoning process. Think aloud explicitly allows trainees to see how high-content knowledge is used by expert clinicians, who are able to select the important information and generate links and associations to organize this information. Research has shown how trainees and expert clinicians are able to identify important pieces of information for a diagnosis.6  However, only expert clinicians are also able to identify and organize the important information to generate a correct diagnosis. Think aloud allows trainees to hear how expert clinicians selectively organize information to solve a patient problem, and it also allows supervising clinicians to understand the way trainees are thinking when using think aloud to explain their developing reasoning process. It transforms the thought processes of expert clinicians, which are otherwise automatic and implicit, into explicit and concrete explanations. Instead of learning how to deal with complex cases merely by exposure to many cases, think aloud provides a systematic way for trainees to develop their clinical reasoning skills for complex cases.7 

Evidence to support think aloud as a way to capture thought processes comes from task analysis and functional magnetic resonance imaging studies.5,8  In task analysis, an individual's thought processes and his or her steps used to solve a problem are made explicit.5  The contention that verbalization accurately reflects the thought processes is supported by the fact that verbalized thoughts are consistent with the sequence of intermediate steps used to solve a problem, and other sequences of verbalized thoughts are used to generate correct answers. Laboratory analyses of recordings of reaction times, error rates, patterns of brain recordings, and sequences of eye fixations add further support to the contention that verbalization during think aloud reflects thought processes.5  Functional magnetic resonance imaging shows how brain activity differs when physicians answer a question, and when they think aloud about how they arrive at their chosen answer.8  These findings add further evidence that think aloud is a measure of the process of critical cognition. In think aloud, it is the thinking processes involved that are important. How we arrive at an answer is what is instructive, rather than just focusing on the answer itself.5 

Much of expert clinicians' reasoning is based on pattern recognition, which is implicit.9  Pattern recognition occurs very quickly and without deliberation. We suggest that teachers retrospectively consider why they arrived at a particular diagnosis based on pattern recognition. Was this based on experience with a similar case previously, or was it that the constellation of symptoms and signs were characteristic or pathognomonic of a particular condition? Most experienced clinicians who are aware of the danger of premature closure when using pattern recognition would seek immediate confirmatory evidence from further history and physical examination to support their intuition or diagnosis based on pattern recognition. The reason for the pattern recognition, and how to avoid the associated risk of premature closure, can be demonstrated to trainees. Think aloud reveals steps in the reasoning process and makes explicit how decisions are made—outlining the process of making the diagnosis, rather than just focusing on the diagnosis. All thinking processes involved in clinical reasoning shown in box 1 can be captured during think aloud. Think aloud shows that clinical reasoning is an iterative process in which the repetition of collecting and analyzing data continues until a final diagnosis is made.10 

How can think aloud be used in graduate medical education to learn and assess clinical reasoning? We suggest that trainees be encouraged to pause intermittently as they present their clinical findings to their peers or supervisors, and use think aloud to explain what and how they are thinking. This can be done in small group teaching sessions or when they are presenting their findings to a supervisor (best done without the patient being present). When trainees do this, their clinical reasoning can be assessed and common errors can be remedied.11  If their supervisors explain how they themselves are thinking as part of this session, trainees can learn how an expert thinks. The table shows an outline of how think aloud can be used.

TABLE

Trainee's Presentation and Supervisor's Think Aloud

Trainee's Presentation and Supervisor's Think Aloud
Trainee's Presentation and Supervisor's Think Aloud
box 1 Cognitive Processes Involved in Clinical Reasoning

Intuition and hypothetical deduction

  • How data are analyzed as they are collected and aggregated

  • How this analysis is used to support diagnoses

  • How the preliminary analysis is used to collect further highly relevant, specific data

  • How possible diagnoses are compared and contrasted

  • How tentative diagnoses are reviewed as new data emerge

  • How additional data are collected for clarification when information is encountered that does not support the tentative diagnosis

  • How clinicians regulate judgments and minimize errors and biases

box 2 Advantages of the Think Aloud Technique
  • It occurs in real time during patient management

  • It requires minimal teacher training

  • All intermediate and critical steps used in clinical reasoning are made explicit

  • It demonstrates the use of intuitive and hypothetical deductive processes

  • It is useful for both teaching and assessment

  • It is incorporated into clinical practice without significantly slowing down clinical activity

While there are a number of ways to teach and learn clinical reasoning, think aloud has unique advantages.12  Based on our experience of including anecdotal feedback from clinical teachers following training in the think aloud technique, advantages of this approach are shown in box 2.

We have conducted workshops at 2 international conferences where clinicians were instructed in how to use think aloud.13,14  The feedback was very positive. Attendees agreed that think aloud is a useful technique to teach and assess clinical reasoning in clinical practice, and it provides an actual teaching strategy for what is a complex process.

To better understand the value of the think aloud technique for teaching and learning clinical reasoning, well-designed research such as comparing the effect of using teach aloud to other methods of teaching clinical reasoning, qualitative research to establish themes on what aspects of think aloud are effective, and a longitudinal study to assess the long-term effects of introducing teach aloud on clinical practice will be key next steps.

References

1
Durning
SJ
,
Artino
AR
Jr,
Schuwirth
L
,
van der Vleuten
C.
Clarifying assumptions to enhance our understanding and assessment of clinical reasoning
.
Acad Med
.
2013
;
88
(
4
):
442
448
.
2
Charlin
B
,
Lubarsky
S
,
Millette
B
,
Crevier
F
,
Audétat
MC
,
Charbonneau
A
,
et al
.
Clinical reasoning processes: unravelling complexity through graphic representation
.
Med Educ
.
2012
;
46
(
5
):
454
463
.
3
Shojania
KG
,
Burton
EC
,
McDonald
KM
,
Goldman
L.
Changes in rates of autopsy-detected diagnostic errors over time: a systematic review
.
JAMA
.
2003
;
289
(
21
):
2849
2856
.
4
Wilson
RM
,
Harrison
BT
,
Gibberd
RW
,
Hamilton
JD.
An analysis of the causes of adverse events from the Quality in Australian Health Care Study
.
Med J Aust
.
1999
;
170
(
9
):
411
415
.
5
Ericsson
KA
,
Simon
HA.
Protocol Analysis: Verbal Reports as Data
.
Cambridge, MA
:
The MIT Press;
1993
.
6
Joseph
GM
,
Patel
VL.
Domain knowledge and hypothesis generation in diagnostic reasoning
.
Med Decis Making
.
1990
;
10
(
1
):
31
46
.
7
Arocha
JF
,
Patel
VL.
Novice diagnostic reasoning in medicine: accounting for evidence
.
J Learn Sc
.
1995
;
4
(
4
):
355
384
.
8
Durning
SJ
,
Artino
AR
Jr,
Beckman
TJ
,
Graner
J
,
van der Vleuten
C
,
Holmboe
E
,
et al
.
Does the think-aloud protocol reflect thinking? Exploring functional neuroimaging differences with thinking (answering multiple choice questions) versus thinking aloud
.
Med Teach
.
2013
;
35
(
9
):
720
726
.
9
Norman
G
,
Young
M
,
Brooks
L.
Non-analytical models of clinical reasoning: the role of experience
.
Med Educ
.
2007
;
41
(
12
):
1140
1145
.
10
Pinnock
R
,
Welch
P
,
Plummer
D
,
Young
L.
Can grounded theory provide a framework for clinical reasoning?
MedEdPublish
.
2014
;
3
(
26
). .
11
Audétat
MC
,
Dory
V
,
Nendaz
M
,
Vanpee
D
,
Pestiaux
D
,
Junod Perron
N
,
et al
.
What is so difficult about managing clinical reasoning difficulties?
Med Educ
.
2012
;
46
(
2
):
216
227
.
12
Pinnock
R
,
Welch
P.
Learning clinical reasoning
.
J Paediatr Child Health
.
2014
;
50
(
4
):
253
257
.
13
Pinnock
R
,
Young
L
,
Spence
F
,
Henning
M
,
Hazell
W.
Teaching and learning clinical reasoning in everyday practice
.
Workshop presented at: AMEE 2013; August 24–28, 2013
;
Prague, Czech Republic
. .
14
Pinnock
R
,
Young
L
,
Spence
F
,
Henning
H
,
Hazell
W.
Teaching and learning clinical reasoning in everyday practice
.
Preconference workshop presented at: AMEE 2014; August 30–September 3, 2014; Milan, Italy
. .